Upload folder using huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/README.md +202 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/adapter_config.json +34 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/adapter_model.safetensors +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/config.json +45 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/non_lora_trainables.bin +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/trainer_state.json +0 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/README.md +202 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/adapter_config.json +34 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/adapter_model.safetensors +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/config.json +45 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/non_lora_trainables.bin +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/trainer_state.json +2226 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/README.md +202 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/adapter_config.json +34 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/adapter_model.safetensors +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/config.json +45 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/non_lora_trainables.bin +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/trainer_state.json +0 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/README.md +202 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/adapter_config.json +34 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/adapter_model.safetensors +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/config.json +45 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/non_lora_trainables.bin +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/trainer_state.json +2226 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/README.md +202 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/adapter_config.json +34 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/adapter_model.safetensors +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/config.json +45 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/non_lora_trainables.bin +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/trainer_state.json +0 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/README.md +202 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/adapter_config.json +34 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/adapter_model.safetensors +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/config.json +45 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/non_lora_trainables.bin +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/trainer_state.json +2226 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/README.md +202 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/adapter_config.json +34 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/adapter_model.safetensors +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/config.json +45 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/non_lora_trainables.bin +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/trainer_state.json +0 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/README.md +202 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/adapter_config.json +34 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/adapter_model.safetensors +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/config.json +45 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/non_lora_trainables.bin +3 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/trainer_state.json +0 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_3_epochs_1_GA_2_lora/README.md +202 -0
- mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_3_epochs_1_GA_2_lora/adapter_config.json +34 -0
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: ./weights/Bunny-v1_1-Llama-3-8B-V
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.11.1
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 256,
|
14 |
+
"lora_dropout": 0.1,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 128,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": null,
|
22 |
+
"target_modules": [
|
23 |
+
"o_proj",
|
24 |
+
"up_proj",
|
25 |
+
"gate_proj",
|
26 |
+
"down_proj",
|
27 |
+
"q_proj",
|
28 |
+
"v_proj",
|
29 |
+
"k_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c6471582fd50057e84dfb08f495660314c24a16a94f8ac9426da1edcd17b1603
|
3 |
+
size 671150064
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/config.json
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
3 |
+
"architectures": [
|
4 |
+
"BunnyLlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"auto_map": {
|
9 |
+
"AutoConfig": "configuration_bunny_llama.BunnyLlamaConfig",
|
10 |
+
"AutoModelForCausalLM": "modeling_bunny_llama.BunnyLlamaForCausalLM"
|
11 |
+
},
|
12 |
+
"bos_token_id": 128000,
|
13 |
+
"continuous_training": false,
|
14 |
+
"eos_token_id": 128001,
|
15 |
+
"freeze_mm_mlp_adapter": false,
|
16 |
+
"hidden_act": "silu",
|
17 |
+
"hidden_size": 4096,
|
18 |
+
"image_aspect_ratio": "pad",
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 14336,
|
21 |
+
"max_position_embeddings": 8192,
|
22 |
+
"mm_hidden_size": 3456,
|
23 |
+
"mm_projector_lr": null,
|
24 |
+
"mm_projector_type": "mlp2x_gelu",
|
25 |
+
"mm_vision_tower": "./weights/siglip-so400m-patch14-384",
|
26 |
+
"model_type": "bunny-llama",
|
27 |
+
"num_attention_heads": 32,
|
28 |
+
"num_hidden_layers": 32,
|
29 |
+
"num_key_value_heads": 8,
|
30 |
+
"pretraining_tp": 1,
|
31 |
+
"rms_norm_eps": 1e-05,
|
32 |
+
"rope_scaling": null,
|
33 |
+
"rope_theta": 500000.0,
|
34 |
+
"tie_word_embeddings": false,
|
35 |
+
"tokenizer_model_max_length": 2048,
|
36 |
+
"tokenizer_padding_side": "right",
|
37 |
+
"torch_dtype": "float16",
|
38 |
+
"transformers_version": "4.41.2",
|
39 |
+
"tune_mm_mlp_adapter": false,
|
40 |
+
"unfreeze_vision_tower": true,
|
41 |
+
"use_cache": true,
|
42 |
+
"use_mm_proj": true,
|
43 |
+
"use_s2": true,
|
44 |
+
"vocab_size": 128256
|
45 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/non_lora_trainables.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:582698d24359f6be8e4df167566915dd311ba98e0ec3ef337f6141b3712b3e53
|
3 |
+
size 918507402
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_2_lora/trainer_state.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: ./weights/Bunny-v1_1-Llama-3-8B-V
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.11.1
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 256,
|
14 |
+
"lora_dropout": 0.1,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 128,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": null,
|
22 |
+
"target_modules": [
|
23 |
+
"gate_proj",
|
24 |
+
"up_proj",
|
25 |
+
"q_proj",
|
26 |
+
"k_proj",
|
27 |
+
"o_proj",
|
28 |
+
"v_proj",
|
29 |
+
"down_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:509f4fe725368c830f31a07e9dfa637e7409475215bca031f12ab92ad28fc1d5
|
3 |
+
size 671150064
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/config.json
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
3 |
+
"architectures": [
|
4 |
+
"BunnyLlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"auto_map": {
|
9 |
+
"AutoConfig": "configuration_bunny_llama.BunnyLlamaConfig",
|
10 |
+
"AutoModelForCausalLM": "modeling_bunny_llama.BunnyLlamaForCausalLM"
|
11 |
+
},
|
12 |
+
"bos_token_id": 128000,
|
13 |
+
"continuous_training": false,
|
14 |
+
"eos_token_id": 128001,
|
15 |
+
"freeze_mm_mlp_adapter": false,
|
16 |
+
"hidden_act": "silu",
|
17 |
+
"hidden_size": 4096,
|
18 |
+
"image_aspect_ratio": "pad",
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 14336,
|
21 |
+
"max_position_embeddings": 8192,
|
22 |
+
"mm_hidden_size": 3456,
|
23 |
+
"mm_projector_lr": null,
|
24 |
+
"mm_projector_type": "mlp2x_gelu",
|
25 |
+
"mm_vision_tower": "./weights/siglip-so400m-patch14-384",
|
26 |
+
"model_type": "bunny-llama",
|
27 |
+
"num_attention_heads": 32,
|
28 |
+
"num_hidden_layers": 32,
|
29 |
+
"num_key_value_heads": 8,
|
30 |
+
"pretraining_tp": 1,
|
31 |
+
"rms_norm_eps": 1e-05,
|
32 |
+
"rope_scaling": null,
|
33 |
+
"rope_theta": 500000.0,
|
34 |
+
"tie_word_embeddings": false,
|
35 |
+
"tokenizer_model_max_length": 2048,
|
36 |
+
"tokenizer_padding_side": "right",
|
37 |
+
"torch_dtype": "float16",
|
38 |
+
"transformers_version": "4.41.2",
|
39 |
+
"tune_mm_mlp_adapter": false,
|
40 |
+
"unfreeze_vision_tower": true,
|
41 |
+
"use_cache": true,
|
42 |
+
"use_mm_proj": true,
|
43 |
+
"use_s2": true,
|
44 |
+
"vocab_size": 128256
|
45 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/non_lora_trainables.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1cfa623b92512d18b2427e22625319dbb311b37891dcfc42a982b687758f3ac6
|
3 |
+
size 918507402
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_1_epochs_1_GA_4_lora/trainer_state.json
ADDED
@@ -0,0 +1,2226 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9984,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 312,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0032,
|
13 |
+
"grad_norm": 0.8230906866528206,
|
14 |
+
"learning_rate": 2e-05,
|
15 |
+
"loss": 1.3718,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.0064,
|
20 |
+
"grad_norm": 0.8011482349224612,
|
21 |
+
"learning_rate": 4e-05,
|
22 |
+
"loss": 1.3884,
|
23 |
+
"step": 2
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.0096,
|
27 |
+
"grad_norm": 0.6777465654408708,
|
28 |
+
"learning_rate": 6e-05,
|
29 |
+
"loss": 1.2942,
|
30 |
+
"step": 3
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.0128,
|
34 |
+
"grad_norm": 0.6985842469244755,
|
35 |
+
"learning_rate": 8e-05,
|
36 |
+
"loss": 1.3011,
|
37 |
+
"step": 4
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.016,
|
41 |
+
"grad_norm": 0.6358810285065617,
|
42 |
+
"learning_rate": 0.0001,
|
43 |
+
"loss": 1.2236,
|
44 |
+
"step": 5
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.0192,
|
48 |
+
"grad_norm": 0.8043615850550121,
|
49 |
+
"learning_rate": 0.00012,
|
50 |
+
"loss": 1.1222,
|
51 |
+
"step": 6
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.0224,
|
55 |
+
"grad_norm": 0.8766371190581345,
|
56 |
+
"learning_rate": 0.00014,
|
57 |
+
"loss": 1.0481,
|
58 |
+
"step": 7
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.0256,
|
62 |
+
"grad_norm": 0.5896469063055397,
|
63 |
+
"learning_rate": 0.00016,
|
64 |
+
"loss": 1.0115,
|
65 |
+
"step": 8
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.0288,
|
69 |
+
"grad_norm": 0.5140042859696677,
|
70 |
+
"learning_rate": 0.00018,
|
71 |
+
"loss": 1.0032,
|
72 |
+
"step": 9
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.032,
|
76 |
+
"grad_norm": 0.47047377131530216,
|
77 |
+
"learning_rate": 0.0002,
|
78 |
+
"loss": 0.9718,
|
79 |
+
"step": 10
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.0352,
|
83 |
+
"grad_norm": 0.4689357680369753,
|
84 |
+
"learning_rate": 0.00019999458931878073,
|
85 |
+
"loss": 1.0164,
|
86 |
+
"step": 11
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.0384,
|
90 |
+
"grad_norm": 0.49055471890156765,
|
91 |
+
"learning_rate": 0.0001999783578606323,
|
92 |
+
"loss": 1.0478,
|
93 |
+
"step": 12
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.0416,
|
97 |
+
"grad_norm": 0.48360570699467204,
|
98 |
+
"learning_rate": 0.00019995130738201966,
|
99 |
+
"loss": 1.0179,
|
100 |
+
"step": 13
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.0448,
|
104 |
+
"grad_norm": 0.4385882347361944,
|
105 |
+
"learning_rate": 0.0001999134408101731,
|
106 |
+
"loss": 0.8839,
|
107 |
+
"step": 14
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.048,
|
111 |
+
"grad_norm": 0.5842432990089418,
|
112 |
+
"learning_rate": 0.00019986476224277165,
|
113 |
+
"loss": 1.0378,
|
114 |
+
"step": 15
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.0512,
|
118 |
+
"grad_norm": 0.4798987625762095,
|
119 |
+
"learning_rate": 0.00019980527694749952,
|
120 |
+
"loss": 0.9014,
|
121 |
+
"step": 16
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.0544,
|
125 |
+
"grad_norm": 0.43385099220728984,
|
126 |
+
"learning_rate": 0.00019973499136147606,
|
127 |
+
"loss": 0.9908,
|
128 |
+
"step": 17
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.0576,
|
132 |
+
"grad_norm": 0.4363177117395925,
|
133 |
+
"learning_rate": 0.0001996539130905593,
|
134 |
+
"loss": 0.9302,
|
135 |
+
"step": 18
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.0608,
|
139 |
+
"grad_norm": 0.4184869797333112,
|
140 |
+
"learning_rate": 0.0001995620509085228,
|
141 |
+
"loss": 0.9156,
|
142 |
+
"step": 19
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.064,
|
146 |
+
"grad_norm": 0.4093043649081553,
|
147 |
+
"learning_rate": 0.00019945941475610623,
|
148 |
+
"loss": 0.9582,
|
149 |
+
"step": 20
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.0672,
|
153 |
+
"grad_norm": 0.40265123100906836,
|
154 |
+
"learning_rate": 0.0001993460157399396,
|
155 |
+
"loss": 0.8491,
|
156 |
+
"step": 21
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.0704,
|
160 |
+
"grad_norm": 0.40866258779980846,
|
161 |
+
"learning_rate": 0.0001992218661313415,
|
162 |
+
"loss": 0.9515,
|
163 |
+
"step": 22
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.0736,
|
167 |
+
"grad_norm": 0.47714968174592676,
|
168 |
+
"learning_rate": 0.00019908697936499103,
|
169 |
+
"loss": 0.983,
|
170 |
+
"step": 23
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.0768,
|
174 |
+
"grad_norm": 0.3725347901708648,
|
175 |
+
"learning_rate": 0.00019894137003747403,
|
176 |
+
"loss": 0.9136,
|
177 |
+
"step": 24
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.08,
|
181 |
+
"grad_norm": 0.48768461879424546,
|
182 |
+
"learning_rate": 0.00019878505390570362,
|
183 |
+
"loss": 0.9486,
|
184 |
+
"step": 25
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.0832,
|
188 |
+
"grad_norm": 0.46702127014841616,
|
189 |
+
"learning_rate": 0.00019861804788521493,
|
190 |
+
"loss": 0.9692,
|
191 |
+
"step": 26
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.0864,
|
195 |
+
"grad_norm": 0.39034215128065525,
|
196 |
+
"learning_rate": 0.00019844037004833473,
|
197 |
+
"loss": 0.8706,
|
198 |
+
"step": 27
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.0896,
|
202 |
+
"grad_norm": 0.4730253707882492,
|
203 |
+
"learning_rate": 0.00019825203962222572,
|
204 |
+
"loss": 0.9078,
|
205 |
+
"step": 28
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.0928,
|
209 |
+
"grad_norm": 0.3566887338551931,
|
210 |
+
"learning_rate": 0.0001980530769868059,
|
211 |
+
"loss": 0.8921,
|
212 |
+
"step": 29
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.096,
|
216 |
+
"grad_norm": 0.37413026403570376,
|
217 |
+
"learning_rate": 0.00019784350367254322,
|
218 |
+
"loss": 0.8723,
|
219 |
+
"step": 30
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.0992,
|
223 |
+
"grad_norm": 0.4494091415776092,
|
224 |
+
"learning_rate": 0.0001976233423581255,
|
225 |
+
"loss": 0.9454,
|
226 |
+
"step": 31
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.1024,
|
230 |
+
"grad_norm": 0.4380782092404729,
|
231 |
+
"learning_rate": 0.0001973926168680066,
|
232 |
+
"loss": 0.9703,
|
233 |
+
"step": 32
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.1056,
|
237 |
+
"grad_norm": 0.4024207233737277,
|
238 |
+
"learning_rate": 0.00019715135216982798,
|
239 |
+
"loss": 0.8925,
|
240 |
+
"step": 33
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.1088,
|
244 |
+
"grad_norm": 0.40406197533093235,
|
245 |
+
"learning_rate": 0.0001968995743717171,
|
246 |
+
"loss": 0.9361,
|
247 |
+
"step": 34
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.112,
|
251 |
+
"grad_norm": 0.3618341982376051,
|
252 |
+
"learning_rate": 0.00019663731071946206,
|
253 |
+
"loss": 0.8614,
|
254 |
+
"step": 35
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.1152,
|
258 |
+
"grad_norm": 0.3658614124884964,
|
259 |
+
"learning_rate": 0.00019636458959356316,
|
260 |
+
"loss": 0.8498,
|
261 |
+
"step": 36
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.1184,
|
265 |
+
"grad_norm": 0.5573838774300858,
|
266 |
+
"learning_rate": 0.0001960814405061619,
|
267 |
+
"loss": 0.8736,
|
268 |
+
"step": 37
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.1216,
|
272 |
+
"grad_norm": 0.38557296694847226,
|
273 |
+
"learning_rate": 0.00019578789409784727,
|
274 |
+
"loss": 0.8929,
|
275 |
+
"step": 38
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.1248,
|
279 |
+
"grad_norm": 0.40079642763917983,
|
280 |
+
"learning_rate": 0.00019548398213434007,
|
281 |
+
"loss": 0.8423,
|
282 |
+
"step": 39
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.128,
|
286 |
+
"grad_norm": 0.38961763082023404,
|
287 |
+
"learning_rate": 0.00019516973750305532,
|
288 |
+
"loss": 0.8835,
|
289 |
+
"step": 40
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.1312,
|
293 |
+
"grad_norm": 0.4099214309122936,
|
294 |
+
"learning_rate": 0.00019484519420954354,
|
295 |
+
"loss": 0.9405,
|
296 |
+
"step": 41
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.1344,
|
300 |
+
"grad_norm": 0.39275472621957447,
|
301 |
+
"learning_rate": 0.00019451038737381077,
|
302 |
+
"loss": 0.8549,
|
303 |
+
"step": 42
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.1376,
|
307 |
+
"grad_norm": 0.37890491998796727,
|
308 |
+
"learning_rate": 0.00019416535322651818,
|
309 |
+
"loss": 0.9153,
|
310 |
+
"step": 43
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.1408,
|
314 |
+
"grad_norm": 0.3677543896244243,
|
315 |
+
"learning_rate": 0.00019381012910506146,
|
316 |
+
"loss": 0.8453,
|
317 |
+
"step": 44
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.144,
|
321 |
+
"grad_norm": 0.3798879355149729,
|
322 |
+
"learning_rate": 0.00019344475344953012,
|
323 |
+
"loss": 0.8136,
|
324 |
+
"step": 45
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.1472,
|
328 |
+
"grad_norm": 0.39541256509109524,
|
329 |
+
"learning_rate": 0.00019306926579854821,
|
330 |
+
"loss": 0.8917,
|
331 |
+
"step": 46
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.1504,
|
335 |
+
"grad_norm": 0.3933918528877009,
|
336 |
+
"learning_rate": 0.00019268370678499533,
|
337 |
+
"loss": 0.8155,
|
338 |
+
"step": 47
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.1536,
|
342 |
+
"grad_norm": 0.40698599023324944,
|
343 |
+
"learning_rate": 0.0001922881181316097,
|
344 |
+
"loss": 0.8666,
|
345 |
+
"step": 48
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.1568,
|
349 |
+
"grad_norm": 0.40042005946240466,
|
350 |
+
"learning_rate": 0.00019188254264647337,
|
351 |
+
"loss": 0.8412,
|
352 |
+
"step": 49
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.16,
|
356 |
+
"grad_norm": 0.4607168760067279,
|
357 |
+
"learning_rate": 0.0001914670242183795,
|
358 |
+
"loss": 0.8868,
|
359 |
+
"step": 50
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.1632,
|
363 |
+
"grad_norm": 0.39972125720735685,
|
364 |
+
"learning_rate": 0.0001910416078120832,
|
365 |
+
"loss": 0.8918,
|
366 |
+
"step": 51
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.1664,
|
370 |
+
"grad_norm": 0.4188280311405362,
|
371 |
+
"learning_rate": 0.0001906063394634356,
|
372 |
+
"loss": 0.902,
|
373 |
+
"step": 52
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.1696,
|
377 |
+
"grad_norm": 0.42940697765165986,
|
378 |
+
"learning_rate": 0.00019016126627440237,
|
379 |
+
"loss": 0.8641,
|
380 |
+
"step": 53
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.1728,
|
384 |
+
"grad_norm": 0.3545329739920464,
|
385 |
+
"learning_rate": 0.00018970643640796642,
|
386 |
+
"loss": 0.8735,
|
387 |
+
"step": 54
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.176,
|
391 |
+
"grad_norm": 0.39694907744821073,
|
392 |
+
"learning_rate": 0.000189241899082916,
|
393 |
+
"loss": 0.7712,
|
394 |
+
"step": 55
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.1792,
|
398 |
+
"grad_norm": 0.3873462097880066,
|
399 |
+
"learning_rate": 0.00018876770456851877,
|
400 |
+
"loss": 0.8613,
|
401 |
+
"step": 56
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.1824,
|
405 |
+
"grad_norm": 0.39944251069469355,
|
406 |
+
"learning_rate": 0.0001882839041790818,
|
407 |
+
"loss": 0.8342,
|
408 |
+
"step": 57
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.1856,
|
412 |
+
"grad_norm": 0.38873644333253415,
|
413 |
+
"learning_rate": 0.00018779055026839868,
|
414 |
+
"loss": 0.8602,
|
415 |
+
"step": 58
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.1888,
|
419 |
+
"grad_norm": 0.41970967893409644,
|
420 |
+
"learning_rate": 0.00018728769622408423,
|
421 |
+
"loss": 0.8577,
|
422 |
+
"step": 59
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.192,
|
426 |
+
"grad_norm": 0.3660391396843375,
|
427 |
+
"learning_rate": 0.00018677539646179707,
|
428 |
+
"loss": 0.8452,
|
429 |
+
"step": 60
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.1952,
|
433 |
+
"grad_norm": 0.4567680983316884,
|
434 |
+
"learning_rate": 0.00018625370641935129,
|
435 |
+
"loss": 0.9641,
|
436 |
+
"step": 61
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.1984,
|
440 |
+
"grad_norm": 0.4347858356915908,
|
441 |
+
"learning_rate": 0.00018572268255071718,
|
442 |
+
"loss": 0.9423,
|
443 |
+
"step": 62
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.2016,
|
447 |
+
"grad_norm": 0.3337070071383876,
|
448 |
+
"learning_rate": 0.00018518238231991218,
|
449 |
+
"loss": 0.8076,
|
450 |
+
"step": 63
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.2048,
|
454 |
+
"grad_norm": 0.4286830947909843,
|
455 |
+
"learning_rate": 0.00018463286419478255,
|
456 |
+
"loss": 0.9527,
|
457 |
+
"step": 64
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.208,
|
461 |
+
"grad_norm": 0.35295999710631726,
|
462 |
+
"learning_rate": 0.00018407418764067627,
|
463 |
+
"loss": 0.8146,
|
464 |
+
"step": 65
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.2112,
|
468 |
+
"grad_norm": 0.40069838158976223,
|
469 |
+
"learning_rate": 0.00018350641311400812,
|
470 |
+
"loss": 0.9182,
|
471 |
+
"step": 66
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.2144,
|
475 |
+
"grad_norm": 0.428778831044889,
|
476 |
+
"learning_rate": 0.0001829296020557174,
|
477 |
+
"loss": 0.9332,
|
478 |
+
"step": 67
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.2176,
|
482 |
+
"grad_norm": 0.35155177053946185,
|
483 |
+
"learning_rate": 0.00018234381688461942,
|
484 |
+
"loss": 0.8472,
|
485 |
+
"step": 68
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.2208,
|
489 |
+
"grad_norm": 0.3393523853315711,
|
490 |
+
"learning_rate": 0.0001817491209906506,
|
491 |
+
"loss": 0.8116,
|
492 |
+
"step": 69
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.224,
|
496 |
+
"grad_norm": 0.39467587805954535,
|
497 |
+
"learning_rate": 0.00018114557872800905,
|
498 |
+
"loss": 0.8882,
|
499 |
+
"step": 70
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.2272,
|
503 |
+
"grad_norm": 0.45738467437649594,
|
504 |
+
"learning_rate": 0.00018053325540819045,
|
505 |
+
"loss": 0.9899,
|
506 |
+
"step": 71
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.2304,
|
510 |
+
"grad_norm": 0.40860575562909096,
|
511 |
+
"learning_rate": 0.0001799122172929206,
|
512 |
+
"loss": 0.8518,
|
513 |
+
"step": 72
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.2336,
|
517 |
+
"grad_norm": 0.37358375585659853,
|
518 |
+
"learning_rate": 0.00017928253158698473,
|
519 |
+
"loss": 0.8671,
|
520 |
+
"step": 73
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.2368,
|
524 |
+
"grad_norm": 0.3614438681293151,
|
525 |
+
"learning_rate": 0.0001786442664309554,
|
526 |
+
"loss": 0.7581,
|
527 |
+
"step": 74
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.24,
|
531 |
+
"grad_norm": 0.39185167616797406,
|
532 |
+
"learning_rate": 0.0001779974908938184,
|
533 |
+
"loss": 0.8273,
|
534 |
+
"step": 75
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.2432,
|
538 |
+
"grad_norm": 0.3725033489438709,
|
539 |
+
"learning_rate": 0.0001773422749654988,
|
540 |
+
"loss": 0.7593,
|
541 |
+
"step": 76
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.2464,
|
545 |
+
"grad_norm": 0.3879109255323267,
|
546 |
+
"learning_rate": 0.00017667868954928694,
|
547 |
+
"loss": 0.8456,
|
548 |
+
"step": 77
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.2496,
|
552 |
+
"grad_norm": 0.3449533879291914,
|
553 |
+
"learning_rate": 0.00017600680645416583,
|
554 |
+
"loss": 0.8246,
|
555 |
+
"step": 78
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.2528,
|
559 |
+
"grad_norm": 0.3557304717641571,
|
560 |
+
"learning_rate": 0.00017532669838704035,
|
561 |
+
"loss": 0.8002,
|
562 |
+
"step": 79
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.256,
|
566 |
+
"grad_norm": 0.39624711741320795,
|
567 |
+
"learning_rate": 0.00017463843894486937,
|
568 |
+
"loss": 0.845,
|
569 |
+
"step": 80
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.2592,
|
573 |
+
"grad_norm": 0.4040277144139735,
|
574 |
+
"learning_rate": 0.0001739421026067017,
|
575 |
+
"loss": 0.9255,
|
576 |
+
"step": 81
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.2624,
|
580 |
+
"grad_norm": 0.3573834496881195,
|
581 |
+
"learning_rate": 0.00017323776472561627,
|
582 |
+
"loss": 0.8664,
|
583 |
+
"step": 82
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.2656,
|
587 |
+
"grad_norm": 0.39861349237455607,
|
588 |
+
"learning_rate": 0.00017252550152056795,
|
589 |
+
"loss": 0.869,
|
590 |
+
"step": 83
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.2688,
|
594 |
+
"grad_norm": 0.4122338593577007,
|
595 |
+
"learning_rate": 0.0001718053900681397,
|
596 |
+
"loss": 0.8542,
|
597 |
+
"step": 84
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.272,
|
601 |
+
"grad_norm": 0.38711562232847824,
|
602 |
+
"learning_rate": 0.00017107750829420176,
|
603 |
+
"loss": 0.8265,
|
604 |
+
"step": 85
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.2752,
|
608 |
+
"grad_norm": 0.39947214892388133,
|
609 |
+
"learning_rate": 0.00017034193496547902,
|
610 |
+
"loss": 0.9227,
|
611 |
+
"step": 86
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.2784,
|
615 |
+
"grad_norm": 0.36818193645695974,
|
616 |
+
"learning_rate": 0.00016959874968102735,
|
617 |
+
"loss": 0.8025,
|
618 |
+
"step": 87
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.2816,
|
622 |
+
"grad_norm": 0.3563867965150087,
|
623 |
+
"learning_rate": 0.00016884803286362,
|
624 |
+
"loss": 0.799,
|
625 |
+
"step": 88
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.2848,
|
629 |
+
"grad_norm": 0.3863926930438656,
|
630 |
+
"learning_rate": 0.00016808986575104465,
|
631 |
+
"loss": 0.9039,
|
632 |
+
"step": 89
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.288,
|
636 |
+
"grad_norm": 0.35592980299837496,
|
637 |
+
"learning_rate": 0.00016732433038731242,
|
638 |
+
"loss": 0.8311,
|
639 |
+
"step": 90
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.2912,
|
643 |
+
"grad_norm": 0.4204313739806602,
|
644 |
+
"learning_rate": 0.0001665515096137797,
|
645 |
+
"loss": 0.9226,
|
646 |
+
"step": 91
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.2944,
|
650 |
+
"grad_norm": 0.3723535694546385,
|
651 |
+
"learning_rate": 0.00016577148706018328,
|
652 |
+
"loss": 0.8311,
|
653 |
+
"step": 92
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.2976,
|
657 |
+
"grad_norm": 0.41234782362203953,
|
658 |
+
"learning_rate": 0.00016498434713559088,
|
659 |
+
"loss": 0.8979,
|
660 |
+
"step": 93
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.3008,
|
664 |
+
"grad_norm": 0.4053425895286585,
|
665 |
+
"learning_rate": 0.00016419017501926656,
|
666 |
+
"loss": 0.8344,
|
667 |
+
"step": 94
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.304,
|
671 |
+
"grad_norm": 0.36505412209597476,
|
672 |
+
"learning_rate": 0.0001633890566514535,
|
673 |
+
"loss": 0.8345,
|
674 |
+
"step": 95
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.3072,
|
678 |
+
"grad_norm": 0.382611792068928,
|
679 |
+
"learning_rate": 0.00016258107872407375,
|
680 |
+
"loss": 0.8749,
|
681 |
+
"step": 96
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.3104,
|
685 |
+
"grad_norm": 0.37741695374713863,
|
686 |
+
"learning_rate": 0.0001617663286713474,
|
687 |
+
"loss": 0.8493,
|
688 |
+
"step": 97
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.3136,
|
692 |
+
"grad_norm": 0.4676602590760305,
|
693 |
+
"learning_rate": 0.00016094489466033043,
|
694 |
+
"loss": 0.8432,
|
695 |
+
"step": 98
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.3168,
|
699 |
+
"grad_norm": 0.34327944895308454,
|
700 |
+
"learning_rate": 0.00016011686558137448,
|
701 |
+
"loss": 0.7974,
|
702 |
+
"step": 99
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.32,
|
706 |
+
"grad_norm": 0.3556557830313786,
|
707 |
+
"learning_rate": 0.0001592823310385073,
|
708 |
+
"loss": 0.8677,
|
709 |
+
"step": 100
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.3232,
|
713 |
+
"grad_norm": 0.4186166790777886,
|
714 |
+
"learning_rate": 0.0001584413813397364,
|
715 |
+
"loss": 0.8586,
|
716 |
+
"step": 101
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.3264,
|
720 |
+
"grad_norm": 0.37112136732584244,
|
721 |
+
"learning_rate": 0.00015759410748727662,
|
722 |
+
"loss": 0.8209,
|
723 |
+
"step": 102
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.3296,
|
727 |
+
"grad_norm": 0.38959686649650355,
|
728 |
+
"learning_rate": 0.00015674060116770236,
|
729 |
+
"loss": 0.8077,
|
730 |
+
"step": 103
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.3328,
|
734 |
+
"grad_norm": 0.36221559570247797,
|
735 |
+
"learning_rate": 0.00015588095474202595,
|
736 |
+
"loss": 0.8142,
|
737 |
+
"step": 104
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.336,
|
741 |
+
"grad_norm": 0.3408395021986205,
|
742 |
+
"learning_rate": 0.00015501526123570277,
|
743 |
+
"loss": 0.825,
|
744 |
+
"step": 105
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.3392,
|
748 |
+
"grad_norm": 0.46256890237184806,
|
749 |
+
"learning_rate": 0.00015414361432856475,
|
750 |
+
"loss": 0.9299,
|
751 |
+
"step": 106
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.3424,
|
755 |
+
"grad_norm": 0.4279687120096611,
|
756 |
+
"learning_rate": 0.0001532661083446829,
|
757 |
+
"loss": 0.9313,
|
758 |
+
"step": 107
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.3456,
|
762 |
+
"grad_norm": 0.499080817408835,
|
763 |
+
"learning_rate": 0.00015238283824216015,
|
764 |
+
"loss": 0.9257,
|
765 |
+
"step": 108
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.3488,
|
769 |
+
"grad_norm": 0.3677888976110564,
|
770 |
+
"learning_rate": 0.00015149389960285558,
|
771 |
+
"loss": 0.8234,
|
772 |
+
"step": 109
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.352,
|
776 |
+
"grad_norm": 0.40300933145036566,
|
777 |
+
"learning_rate": 0.00015059938862204127,
|
778 |
+
"loss": 0.8661,
|
779 |
+
"step": 110
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.3552,
|
783 |
+
"grad_norm": 0.35272019193019816,
|
784 |
+
"learning_rate": 0.00014969940209799248,
|
785 |
+
"loss": 0.7685,
|
786 |
+
"step": 111
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.3584,
|
790 |
+
"grad_norm": 0.42072850665932426,
|
791 |
+
"learning_rate": 0.00014879403742151283,
|
792 |
+
"loss": 0.838,
|
793 |
+
"step": 112
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.3616,
|
797 |
+
"grad_norm": 0.3843947444096128,
|
798 |
+
"learning_rate": 0.00014788339256539544,
|
799 |
+
"loss": 0.7295,
|
800 |
+
"step": 113
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.3648,
|
804 |
+
"grad_norm": 0.43191105712976785,
|
805 |
+
"learning_rate": 0.0001469675660738206,
|
806 |
+
"loss": 0.8535,
|
807 |
+
"step": 114
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.368,
|
811 |
+
"grad_norm": 0.3584146254614676,
|
812 |
+
"learning_rate": 0.00014604665705169237,
|
813 |
+
"loss": 0.8303,
|
814 |
+
"step": 115
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.3712,
|
818 |
+
"grad_norm": 0.4020587727121328,
|
819 |
+
"learning_rate": 0.00014512076515391375,
|
820 |
+
"loss": 0.8827,
|
821 |
+
"step": 116
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.3744,
|
825 |
+
"grad_norm": 0.42244806005338637,
|
826 |
+
"learning_rate": 0.00014418999057460276,
|
827 |
+
"loss": 0.8637,
|
828 |
+
"step": 117
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.3776,
|
832 |
+
"grad_norm": 0.3490560388254176,
|
833 |
+
"learning_rate": 0.0001432544340362501,
|
834 |
+
"loss": 0.7685,
|
835 |
+
"step": 118
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.3808,
|
839 |
+
"grad_norm": 0.3610932978019899,
|
840 |
+
"learning_rate": 0.00014231419677881966,
|
841 |
+
"loss": 0.7689,
|
842 |
+
"step": 119
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.384,
|
846 |
+
"grad_norm": 0.35696651622274633,
|
847 |
+
"learning_rate": 0.00014136938054879283,
|
848 |
+
"loss": 0.8692,
|
849 |
+
"step": 120
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.3872,
|
853 |
+
"grad_norm": 0.37982273980822234,
|
854 |
+
"learning_rate": 0.00014042008758815818,
|
855 |
+
"loss": 0.8633,
|
856 |
+
"step": 121
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.3904,
|
860 |
+
"grad_norm": 0.49338348781038344,
|
861 |
+
"learning_rate": 0.00013946642062334766,
|
862 |
+
"loss": 0.7882,
|
863 |
+
"step": 122
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.3936,
|
867 |
+
"grad_norm": 0.41407314429068576,
|
868 |
+
"learning_rate": 0.00013850848285411994,
|
869 |
+
"loss": 0.8352,
|
870 |
+
"step": 123
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.3968,
|
874 |
+
"grad_norm": 0.35863668034543733,
|
875 |
+
"learning_rate": 0.000137546377942393,
|
876 |
+
"loss": 0.824,
|
877 |
+
"step": 124
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.4,
|
881 |
+
"grad_norm": 0.34638170296412163,
|
882 |
+
"learning_rate": 0.00013658021000102636,
|
883 |
+
"loss": 0.7332,
|
884 |
+
"step": 125
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.4032,
|
888 |
+
"grad_norm": 0.33192770455483456,
|
889 |
+
"learning_rate": 0.00013561008358255468,
|
890 |
+
"loss": 0.8012,
|
891 |
+
"step": 126
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.4064,
|
895 |
+
"grad_norm": 0.4018840350246986,
|
896 |
+
"learning_rate": 0.00013463610366787392,
|
897 |
+
"loss": 0.8274,
|
898 |
+
"step": 127
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.4096,
|
902 |
+
"grad_norm": 0.46935625670519754,
|
903 |
+
"learning_rate": 0.00013365837565488064,
|
904 |
+
"loss": 0.9137,
|
905 |
+
"step": 128
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.4128,
|
909 |
+
"grad_norm": 0.3467457829185042,
|
910 |
+
"learning_rate": 0.0001326770053470668,
|
911 |
+
"loss": 0.8057,
|
912 |
+
"step": 129
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.416,
|
916 |
+
"grad_norm": 0.3713608562339026,
|
917 |
+
"learning_rate": 0.0001316920989420703,
|
918 |
+
"loss": 0.778,
|
919 |
+
"step": 130
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.4192,
|
923 |
+
"grad_norm": 0.2953063312947437,
|
924 |
+
"learning_rate": 0.00013070376302018287,
|
925 |
+
"loss": 0.7741,
|
926 |
+
"step": 131
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.4224,
|
930 |
+
"grad_norm": 0.315778283756295,
|
931 |
+
"learning_rate": 0.00012971210453281674,
|
932 |
+
"loss": 0.7968,
|
933 |
+
"step": 132
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.4256,
|
937 |
+
"grad_norm": 0.3777293731399955,
|
938 |
+
"learning_rate": 0.000128717230790931,
|
939 |
+
"loss": 0.8476,
|
940 |
+
"step": 133
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.4288,
|
944 |
+
"grad_norm": 0.36565481085206214,
|
945 |
+
"learning_rate": 0.00012771924945341906,
|
946 |
+
"loss": 0.7424,
|
947 |
+
"step": 134
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.432,
|
951 |
+
"grad_norm": 0.40097648303660766,
|
952 |
+
"learning_rate": 0.00012671826851545851,
|
953 |
+
"loss": 0.8961,
|
954 |
+
"step": 135
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.4352,
|
958 |
+
"grad_norm": 0.4335469508570514,
|
959 |
+
"learning_rate": 0.0001257143962968246,
|
960 |
+
"loss": 0.9337,
|
961 |
+
"step": 136
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.4384,
|
965 |
+
"grad_norm": 0.3571253000150674,
|
966 |
+
"learning_rate": 0.00012470774143016853,
|
967 |
+
"loss": 0.7772,
|
968 |
+
"step": 137
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.4416,
|
972 |
+
"grad_norm": 0.4175995621863514,
|
973 |
+
"learning_rate": 0.00012369841284926188,
|
974 |
+
"loss": 0.7899,
|
975 |
+
"step": 138
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.4448,
|
979 |
+
"grad_norm": 0.3555584844534414,
|
980 |
+
"learning_rate": 0.00012268651977720866,
|
981 |
+
"loss": 0.8058,
|
982 |
+
"step": 139
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.448,
|
986 |
+
"grad_norm": 0.4478506698284743,
|
987 |
+
"learning_rate": 0.00012167217171462566,
|
988 |
+
"loss": 0.8602,
|
989 |
+
"step": 140
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.4512,
|
993 |
+
"grad_norm": 0.3924154628163152,
|
994 |
+
"learning_rate": 0.0001206554784277931,
|
995 |
+
"loss": 0.8566,
|
996 |
+
"step": 141
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.4544,
|
1000 |
+
"grad_norm": 0.3413680221828951,
|
1001 |
+
"learning_rate": 0.00011963654993677645,
|
1002 |
+
"loss": 0.8442,
|
1003 |
+
"step": 142
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.4576,
|
1007 |
+
"grad_norm": 0.37597687671024504,
|
1008 |
+
"learning_rate": 0.00011861549650352069,
|
1009 |
+
"loss": 0.7896,
|
1010 |
+
"step": 143
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.4608,
|
1014 |
+
"grad_norm": 0.35508126757199976,
|
1015 |
+
"learning_rate": 0.00011759242861991855,
|
1016 |
+
"loss": 0.7702,
|
1017 |
+
"step": 144
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.464,
|
1021 |
+
"grad_norm": 0.35676098581829857,
|
1022 |
+
"learning_rate": 0.00011656745699585371,
|
1023 |
+
"loss": 0.7463,
|
1024 |
+
"step": 145
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.4672,
|
1028 |
+
"grad_norm": 0.34262057715324995,
|
1029 |
+
"learning_rate": 0.00011554069254722051,
|
1030 |
+
"loss": 0.8023,
|
1031 |
+
"step": 146
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.4704,
|
1035 |
+
"grad_norm": 0.3160775424152765,
|
1036 |
+
"learning_rate": 0.00011451224638392129,
|
1037 |
+
"loss": 0.753,
|
1038 |
+
"step": 147
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.4736,
|
1042 |
+
"grad_norm": 0.41408430747051145,
|
1043 |
+
"learning_rate": 0.00011348222979784289,
|
1044 |
+
"loss": 0.8722,
|
1045 |
+
"step": 148
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.4768,
|
1049 |
+
"grad_norm": 0.3807991104363706,
|
1050 |
+
"learning_rate": 0.00011245075425081328,
|
1051 |
+
"loss": 0.7902,
|
1052 |
+
"step": 149
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.48,
|
1056 |
+
"grad_norm": 0.3444267731692557,
|
1057 |
+
"learning_rate": 0.00011141793136253986,
|
1058 |
+
"loss": 0.7729,
|
1059 |
+
"step": 150
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.4832,
|
1063 |
+
"grad_norm": 0.3198788705719864,
|
1064 |
+
"learning_rate": 0.0001103838728985307,
|
1065 |
+
"loss": 0.7863,
|
1066 |
+
"step": 151
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.4864,
|
1070 |
+
"grad_norm": 0.41645947104870057,
|
1071 |
+
"learning_rate": 0.000109348690758,
|
1072 |
+
"loss": 0.8442,
|
1073 |
+
"step": 152
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.4896,
|
1077 |
+
"grad_norm": 0.3473642274949473,
|
1078 |
+
"learning_rate": 0.00010831249696175918,
|
1079 |
+
"loss": 0.7755,
|
1080 |
+
"step": 153
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.4928,
|
1084 |
+
"grad_norm": 0.351233087536269,
|
1085 |
+
"learning_rate": 0.0001072754036400944,
|
1086 |
+
"loss": 0.7799,
|
1087 |
+
"step": 154
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.496,
|
1091 |
+
"grad_norm": 0.39062438800982935,
|
1092 |
+
"learning_rate": 0.00010623752302063283,
|
1093 |
+
"loss": 0.8262,
|
1094 |
+
"step": 155
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.4992,
|
1098 |
+
"grad_norm": 0.4661216475962772,
|
1099 |
+
"learning_rate": 0.00010519896741619803,
|
1100 |
+
"loss": 0.8754,
|
1101 |
+
"step": 156
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.5024,
|
1105 |
+
"grad_norm": 0.3680543907349618,
|
1106 |
+
"learning_rate": 0.00010415984921265609,
|
1107 |
+
"loss": 0.8511,
|
1108 |
+
"step": 157
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.5056,
|
1112 |
+
"grad_norm": 0.3554554750391017,
|
1113 |
+
"learning_rate": 0.00010312028085675391,
|
1114 |
+
"loss": 0.7905,
|
1115 |
+
"step": 158
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.5088,
|
1119 |
+
"grad_norm": 0.34003331830378203,
|
1120 |
+
"learning_rate": 0.00010208037484395114,
|
1121 |
+
"loss": 0.8518,
|
1122 |
+
"step": 159
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.512,
|
1126 |
+
"grad_norm": 0.35691497758794416,
|
1127 |
+
"learning_rate": 0.00010104024370624644,
|
1128 |
+
"loss": 0.8042,
|
1129 |
+
"step": 160
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.5152,
|
1133 |
+
"grad_norm": 0.3680532317532297,
|
1134 |
+
"learning_rate": 0.0001,
|
1135 |
+
"loss": 0.826,
|
1136 |
+
"step": 161
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.5184,
|
1140 |
+
"grad_norm": 0.4306874440922751,
|
1141 |
+
"learning_rate": 9.895975629375359e-05,
|
1142 |
+
"loss": 0.8114,
|
1143 |
+
"step": 162
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.5216,
|
1147 |
+
"grad_norm": 0.3927827556296759,
|
1148 |
+
"learning_rate": 9.791962515604887e-05,
|
1149 |
+
"loss": 0.8477,
|
1150 |
+
"step": 163
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.5248,
|
1154 |
+
"grad_norm": 0.3688233147013582,
|
1155 |
+
"learning_rate": 9.687971914324607e-05,
|
1156 |
+
"loss": 0.787,
|
1157 |
+
"step": 164
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.528,
|
1161 |
+
"grad_norm": 0.3460381638988584,
|
1162 |
+
"learning_rate": 9.584015078734395e-05,
|
1163 |
+
"loss": 0.7861,
|
1164 |
+
"step": 165
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.5312,
|
1168 |
+
"grad_norm": 0.3293873896538617,
|
1169 |
+
"learning_rate": 9.480103258380198e-05,
|
1170 |
+
"loss": 0.7599,
|
1171 |
+
"step": 166
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.5344,
|
1175 |
+
"grad_norm": 0.33067678398241424,
|
1176 |
+
"learning_rate": 9.376247697936719e-05,
|
1177 |
+
"loss": 0.7788,
|
1178 |
+
"step": 167
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.5376,
|
1182 |
+
"grad_norm": 0.30272346196435956,
|
1183 |
+
"learning_rate": 9.272459635990562e-05,
|
1184 |
+
"loss": 0.7594,
|
1185 |
+
"step": 168
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.5408,
|
1189 |
+
"grad_norm": 0.35694173144440655,
|
1190 |
+
"learning_rate": 9.168750303824084e-05,
|
1191 |
+
"loss": 0.8066,
|
1192 |
+
"step": 169
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.544,
|
1196 |
+
"grad_norm": 0.36394962483417836,
|
1197 |
+
"learning_rate": 9.065130924199998e-05,
|
1198 |
+
"loss": 0.6855,
|
1199 |
+
"step": 170
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.5472,
|
1203 |
+
"grad_norm": 0.4052142574936814,
|
1204 |
+
"learning_rate": 8.961612710146934e-05,
|
1205 |
+
"loss": 0.8238,
|
1206 |
+
"step": 171
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.5504,
|
1210 |
+
"grad_norm": 0.4466361193230218,
|
1211 |
+
"learning_rate": 8.858206863746018e-05,
|
1212 |
+
"loss": 0.7949,
|
1213 |
+
"step": 172
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.5536,
|
1217 |
+
"grad_norm": 0.3691146825930911,
|
1218 |
+
"learning_rate": 8.754924574918675e-05,
|
1219 |
+
"loss": 0.7591,
|
1220 |
+
"step": 173
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.5568,
|
1224 |
+
"grad_norm": 0.3884324905017403,
|
1225 |
+
"learning_rate": 8.651777020215712e-05,
|
1226 |
+
"loss": 0.7671,
|
1227 |
+
"step": 174
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.56,
|
1231 |
+
"grad_norm": 0.37235345511558177,
|
1232 |
+
"learning_rate": 8.548775361607872e-05,
|
1233 |
+
"loss": 0.7954,
|
1234 |
+
"step": 175
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.5632,
|
1238 |
+
"grad_norm": 0.43342144747289674,
|
1239 |
+
"learning_rate": 8.445930745277953e-05,
|
1240 |
+
"loss": 0.7674,
|
1241 |
+
"step": 176
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.5664,
|
1245 |
+
"grad_norm": 0.3601056959148259,
|
1246 |
+
"learning_rate": 8.343254300414628e-05,
|
1247 |
+
"loss": 0.6814,
|
1248 |
+
"step": 177
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.5696,
|
1252 |
+
"grad_norm": 0.34765672643846884,
|
1253 |
+
"learning_rate": 8.240757138008149e-05,
|
1254 |
+
"loss": 0.8115,
|
1255 |
+
"step": 178
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.5728,
|
1259 |
+
"grad_norm": 0.3542918688839561,
|
1260 |
+
"learning_rate": 8.138450349647936e-05,
|
1261 |
+
"loss": 0.7751,
|
1262 |
+
"step": 179
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.576,
|
1266 |
+
"grad_norm": 0.43405703341062163,
|
1267 |
+
"learning_rate": 8.036345006322359e-05,
|
1268 |
+
"loss": 0.883,
|
1269 |
+
"step": 180
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.5792,
|
1273 |
+
"grad_norm": 0.35990538609349737,
|
1274 |
+
"learning_rate": 7.934452157220694e-05,
|
1275 |
+
"loss": 0.8274,
|
1276 |
+
"step": 181
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.5824,
|
1280 |
+
"grad_norm": 0.36566849513888866,
|
1281 |
+
"learning_rate": 7.832782828537437e-05,
|
1282 |
+
"loss": 0.8316,
|
1283 |
+
"step": 182
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.5856,
|
1287 |
+
"grad_norm": 0.3880712586476513,
|
1288 |
+
"learning_rate": 7.731348022279134e-05,
|
1289 |
+
"loss": 0.7823,
|
1290 |
+
"step": 183
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.5888,
|
1294 |
+
"grad_norm": 0.40070345592700185,
|
1295 |
+
"learning_rate": 7.630158715073813e-05,
|
1296 |
+
"loss": 0.8682,
|
1297 |
+
"step": 184
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.592,
|
1301 |
+
"grad_norm": 0.452126500075801,
|
1302 |
+
"learning_rate": 7.52922585698315e-05,
|
1303 |
+
"loss": 0.7349,
|
1304 |
+
"step": 185
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.5952,
|
1308 |
+
"grad_norm": 0.39133686156881714,
|
1309 |
+
"learning_rate": 7.428560370317542e-05,
|
1310 |
+
"loss": 0.8166,
|
1311 |
+
"step": 186
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.5984,
|
1315 |
+
"grad_norm": 0.30433046880932474,
|
1316 |
+
"learning_rate": 7.328173148454151e-05,
|
1317 |
+
"loss": 0.7799,
|
1318 |
+
"step": 187
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.6016,
|
1322 |
+
"grad_norm": 0.38912324970709505,
|
1323 |
+
"learning_rate": 7.228075054658096e-05,
|
1324 |
+
"loss": 0.874,
|
1325 |
+
"step": 188
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.6048,
|
1329 |
+
"grad_norm": 0.34571755403615634,
|
1330 |
+
"learning_rate": 7.1282769209069e-05,
|
1331 |
+
"loss": 0.7972,
|
1332 |
+
"step": 189
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.608,
|
1336 |
+
"grad_norm": 0.3759979893898139,
|
1337 |
+
"learning_rate": 7.028789546718326e-05,
|
1338 |
+
"loss": 0.7436,
|
1339 |
+
"step": 190
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.6112,
|
1343 |
+
"grad_norm": 0.3396623833976361,
|
1344 |
+
"learning_rate": 6.929623697981718e-05,
|
1345 |
+
"loss": 0.771,
|
1346 |
+
"step": 191
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.6144,
|
1350 |
+
"grad_norm": 0.3930329153264238,
|
1351 |
+
"learning_rate": 6.830790105792973e-05,
|
1352 |
+
"loss": 0.7294,
|
1353 |
+
"step": 192
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.6176,
|
1357 |
+
"grad_norm": 0.384939784498545,
|
1358 |
+
"learning_rate": 6.732299465293322e-05,
|
1359 |
+
"loss": 0.8072,
|
1360 |
+
"step": 193
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.6208,
|
1364 |
+
"grad_norm": 0.3670712491930503,
|
1365 |
+
"learning_rate": 6.63416243451194e-05,
|
1366 |
+
"loss": 0.8139,
|
1367 |
+
"step": 194
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.624,
|
1371 |
+
"grad_norm": 0.37471193485584176,
|
1372 |
+
"learning_rate": 6.536389633212609e-05,
|
1373 |
+
"loss": 0.8026,
|
1374 |
+
"step": 195
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.6272,
|
1378 |
+
"grad_norm": 0.3466481941938704,
|
1379 |
+
"learning_rate": 6.43899164174453e-05,
|
1380 |
+
"loss": 0.7856,
|
1381 |
+
"step": 196
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.6304,
|
1385 |
+
"grad_norm": 0.334681278560714,
|
1386 |
+
"learning_rate": 6.341978999897365e-05,
|
1387 |
+
"loss": 0.7457,
|
1388 |
+
"step": 197
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.6336,
|
1392 |
+
"grad_norm": 0.3580934295092547,
|
1393 |
+
"learning_rate": 6.245362205760704e-05,
|
1394 |
+
"loss": 0.8062,
|
1395 |
+
"step": 198
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.6368,
|
1399 |
+
"grad_norm": 0.41942104569534644,
|
1400 |
+
"learning_rate": 6.149151714588009e-05,
|
1401 |
+
"loss": 0.8307,
|
1402 |
+
"step": 199
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.64,
|
1406 |
+
"grad_norm": 0.3044902604615677,
|
1407 |
+
"learning_rate": 6.053357937665237e-05,
|
1408 |
+
"loss": 0.6707,
|
1409 |
+
"step": 200
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.6432,
|
1413 |
+
"grad_norm": 0.37144444775815594,
|
1414 |
+
"learning_rate": 5.957991241184184e-05,
|
1415 |
+
"loss": 0.7813,
|
1416 |
+
"step": 201
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.6464,
|
1420 |
+
"grad_norm": 0.34248491706444256,
|
1421 |
+
"learning_rate": 5.863061945120719e-05,
|
1422 |
+
"loss": 0.8051,
|
1423 |
+
"step": 202
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.6496,
|
1427 |
+
"grad_norm": 0.34748171044370163,
|
1428 |
+
"learning_rate": 5.768580322118034e-05,
|
1429 |
+
"loss": 0.7627,
|
1430 |
+
"step": 203
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.6528,
|
1434 |
+
"grad_norm": 0.38512453359767645,
|
1435 |
+
"learning_rate": 5.6745565963749925e-05,
|
1436 |
+
"loss": 0.8318,
|
1437 |
+
"step": 204
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.656,
|
1441 |
+
"grad_norm": 0.36707414698679175,
|
1442 |
+
"learning_rate": 5.5810009425397294e-05,
|
1443 |
+
"loss": 0.8305,
|
1444 |
+
"step": 205
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 0.6592,
|
1448 |
+
"grad_norm": 0.48217984342912773,
|
1449 |
+
"learning_rate": 5.487923484608629e-05,
|
1450 |
+
"loss": 0.9014,
|
1451 |
+
"step": 206
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 0.6624,
|
1455 |
+
"grad_norm": 0.38863801931906744,
|
1456 |
+
"learning_rate": 5.395334294830765e-05,
|
1457 |
+
"loss": 0.8525,
|
1458 |
+
"step": 207
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 0.6656,
|
1462 |
+
"grad_norm": 0.35518873432323794,
|
1463 |
+
"learning_rate": 5.3032433926179395e-05,
|
1464 |
+
"loss": 0.7778,
|
1465 |
+
"step": 208
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 0.6688,
|
1469 |
+
"grad_norm": 0.34123260040379993,
|
1470 |
+
"learning_rate": 5.211660743460458e-05,
|
1471 |
+
"loss": 0.775,
|
1472 |
+
"step": 209
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 0.672,
|
1476 |
+
"grad_norm": 0.4001757666899788,
|
1477 |
+
"learning_rate": 5.1205962578487155e-05,
|
1478 |
+
"loss": 0.7731,
|
1479 |
+
"step": 210
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 0.6752,
|
1483 |
+
"grad_norm": 0.3315496723721597,
|
1484 |
+
"learning_rate": 5.030059790200756e-05,
|
1485 |
+
"loss": 0.7922,
|
1486 |
+
"step": 211
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 0.6784,
|
1490 |
+
"grad_norm": 0.3258037685568246,
|
1491 |
+
"learning_rate": 4.940061137795876e-05,
|
1492 |
+
"loss": 0.699,
|
1493 |
+
"step": 212
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.6816,
|
1497 |
+
"grad_norm": 0.3884518590202934,
|
1498 |
+
"learning_rate": 4.850610039714444e-05,
|
1499 |
+
"loss": 0.7866,
|
1500 |
+
"step": 213
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 0.6848,
|
1504 |
+
"grad_norm": 0.7104627436824442,
|
1505 |
+
"learning_rate": 4.761716175783989e-05,
|
1506 |
+
"loss": 0.8089,
|
1507 |
+
"step": 214
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 0.688,
|
1511 |
+
"grad_norm": 0.3013941102836104,
|
1512 |
+
"learning_rate": 4.673389165531714e-05,
|
1513 |
+
"loss": 0.7054,
|
1514 |
+
"step": 215
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 0.6912,
|
1518 |
+
"grad_norm": 0.3731169801442719,
|
1519 |
+
"learning_rate": 4.585638567143529e-05,
|
1520 |
+
"loss": 0.7737,
|
1521 |
+
"step": 216
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 0.6944,
|
1525 |
+
"grad_norm": 0.30263267234838137,
|
1526 |
+
"learning_rate": 4.498473876429726e-05,
|
1527 |
+
"loss": 0.7029,
|
1528 |
+
"step": 217
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 0.6976,
|
1532 |
+
"grad_norm": 0.3659751206161679,
|
1533 |
+
"learning_rate": 4.411904525797408e-05,
|
1534 |
+
"loss": 0.8146,
|
1535 |
+
"step": 218
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 0.7008,
|
1539 |
+
"grad_norm": 0.36421909907458466,
|
1540 |
+
"learning_rate": 4.325939883229766e-05,
|
1541 |
+
"loss": 0.7769,
|
1542 |
+
"step": 219
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 0.704,
|
1546 |
+
"grad_norm": 0.3403495561782968,
|
1547 |
+
"learning_rate": 4.240589251272342e-05,
|
1548 |
+
"loss": 0.8074,
|
1549 |
+
"step": 220
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 0.7072,
|
1553 |
+
"grad_norm": 0.4116298314841219,
|
1554 |
+
"learning_rate": 4.155861866026364e-05,
|
1555 |
+
"loss": 0.7561,
|
1556 |
+
"step": 221
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.7104,
|
1560 |
+
"grad_norm": 0.4017877316418641,
|
1561 |
+
"learning_rate": 4.071766896149273e-05,
|
1562 |
+
"loss": 0.7802,
|
1563 |
+
"step": 222
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 0.7136,
|
1567 |
+
"grad_norm": 0.3933048903616485,
|
1568 |
+
"learning_rate": 3.988313441862553e-05,
|
1569 |
+
"loss": 0.8302,
|
1570 |
+
"step": 223
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 0.7168,
|
1574 |
+
"grad_norm": 0.4053820034136759,
|
1575 |
+
"learning_rate": 3.9055105339669595e-05,
|
1576 |
+
"loss": 0.7537,
|
1577 |
+
"step": 224
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 0.72,
|
1581 |
+
"grad_norm": 0.3422570001344906,
|
1582 |
+
"learning_rate": 3.823367132865265e-05,
|
1583 |
+
"loss": 0.7513,
|
1584 |
+
"step": 225
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 0.7232,
|
1588 |
+
"grad_norm": 0.3672321719467038,
|
1589 |
+
"learning_rate": 3.741892127592625e-05,
|
1590 |
+
"loss": 0.7552,
|
1591 |
+
"step": 226
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 0.7264,
|
1595 |
+
"grad_norm": 0.38655052693729053,
|
1596 |
+
"learning_rate": 3.6610943348546526e-05,
|
1597 |
+
"loss": 0.8076,
|
1598 |
+
"step": 227
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 0.7296,
|
1602 |
+
"grad_norm": 0.32046006517959386,
|
1603 |
+
"learning_rate": 3.580982498073344e-05,
|
1604 |
+
"loss": 0.7549,
|
1605 |
+
"step": 228
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 0.7328,
|
1609 |
+
"grad_norm": 0.45524335147940526,
|
1610 |
+
"learning_rate": 3.501565286440914e-05,
|
1611 |
+
"loss": 0.8619,
|
1612 |
+
"step": 229
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 0.736,
|
1616 |
+
"grad_norm": 0.29405543478538154,
|
1617 |
+
"learning_rate": 3.422851293981676e-05,
|
1618 |
+
"loss": 0.7585,
|
1619 |
+
"step": 230
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 0.7392,
|
1623 |
+
"grad_norm": 0.38285886702055727,
|
1624 |
+
"learning_rate": 3.3448490386220355e-05,
|
1625 |
+
"loss": 0.8357,
|
1626 |
+
"step": 231
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 0.7424,
|
1630 |
+
"grad_norm": 0.3801889952884557,
|
1631 |
+
"learning_rate": 3.2675669612687565e-05,
|
1632 |
+
"loss": 0.8901,
|
1633 |
+
"step": 232
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 0.7456,
|
1637 |
+
"grad_norm": 0.35279011141194516,
|
1638 |
+
"learning_rate": 3.191013424895536e-05,
|
1639 |
+
"loss": 0.7481,
|
1640 |
+
"step": 233
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 0.7488,
|
1644 |
+
"grad_norm": 0.3471907142593491,
|
1645 |
+
"learning_rate": 3.115196713638e-05,
|
1646 |
+
"loss": 0.7855,
|
1647 |
+
"step": 234
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 0.752,
|
1651 |
+
"grad_norm": 0.36619187308979667,
|
1652 |
+
"learning_rate": 3.040125031897264e-05,
|
1653 |
+
"loss": 0.7625,
|
1654 |
+
"step": 235
|
1655 |
+
},
|
1656 |
+
{
|
1657 |
+
"epoch": 0.7552,
|
1658 |
+
"grad_norm": 0.3762428412156853,
|
1659 |
+
"learning_rate": 2.9658065034520978e-05,
|
1660 |
+
"loss": 0.8157,
|
1661 |
+
"step": 236
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 0.7584,
|
1665 |
+
"grad_norm": 0.3171161168873978,
|
1666 |
+
"learning_rate": 2.892249170579826e-05,
|
1667 |
+
"loss": 0.7799,
|
1668 |
+
"step": 237
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 0.7616,
|
1672 |
+
"grad_norm": 0.3592730455514312,
|
1673 |
+
"learning_rate": 2.8194609931860316e-05,
|
1674 |
+
"loss": 0.7478,
|
1675 |
+
"step": 238
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 0.7648,
|
1679 |
+
"grad_norm": 0.4134931519699558,
|
1680 |
+
"learning_rate": 2.7474498479432087e-05,
|
1681 |
+
"loss": 0.8437,
|
1682 |
+
"step": 239
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.768,
|
1686 |
+
"grad_norm": 0.30546020641842614,
|
1687 |
+
"learning_rate": 2.6762235274383772e-05,
|
1688 |
+
"loss": 0.7224,
|
1689 |
+
"step": 240
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 0.7712,
|
1693 |
+
"grad_norm": 0.3454408457555957,
|
1694 |
+
"learning_rate": 2.6057897393298324e-05,
|
1695 |
+
"loss": 0.7727,
|
1696 |
+
"step": 241
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 0.7744,
|
1700 |
+
"grad_norm": 0.3680677506255253,
|
1701 |
+
"learning_rate": 2.536156105513062e-05,
|
1702 |
+
"loss": 0.7745,
|
1703 |
+
"step": 242
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 0.7776,
|
1707 |
+
"grad_norm": 0.36274461235683025,
|
1708 |
+
"learning_rate": 2.4673301612959654e-05,
|
1709 |
+
"loss": 0.8305,
|
1710 |
+
"step": 243
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 0.7808,
|
1714 |
+
"grad_norm": 0.4459754472676995,
|
1715 |
+
"learning_rate": 2.399319354583418e-05,
|
1716 |
+
"loss": 0.8917,
|
1717 |
+
"step": 244
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 0.784,
|
1721 |
+
"grad_norm": 0.34453176080889125,
|
1722 |
+
"learning_rate": 2.3321310450713062e-05,
|
1723 |
+
"loss": 0.7476,
|
1724 |
+
"step": 245
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 0.7872,
|
1728 |
+
"grad_norm": 0.31196539201987367,
|
1729 |
+
"learning_rate": 2.265772503450122e-05,
|
1730 |
+
"loss": 0.7468,
|
1731 |
+
"step": 246
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 0.7904,
|
1735 |
+
"grad_norm": 0.3184608885787424,
|
1736 |
+
"learning_rate": 2.2002509106181624e-05,
|
1737 |
+
"loss": 0.7402,
|
1738 |
+
"step": 247
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 0.7936,
|
1742 |
+
"grad_norm": 0.36443373379957883,
|
1743 |
+
"learning_rate": 2.1355733569044635e-05,
|
1744 |
+
"loss": 0.7567,
|
1745 |
+
"step": 248
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 0.7968,
|
1749 |
+
"grad_norm": 0.4417924711660864,
|
1750 |
+
"learning_rate": 2.0717468413015283e-05,
|
1751 |
+
"loss": 0.823,
|
1752 |
+
"step": 249
|
1753 |
+
},
|
1754 |
+
{
|
1755 |
+
"epoch": 0.8,
|
1756 |
+
"grad_norm": 0.3236759284366048,
|
1757 |
+
"learning_rate": 2.008778270707944e-05,
|
1758 |
+
"loss": 0.6885,
|
1759 |
+
"step": 250
|
1760 |
+
},
|
1761 |
+
{
|
1762 |
+
"epoch": 0.8032,
|
1763 |
+
"grad_norm": 0.39242767874833534,
|
1764 |
+
"learning_rate": 1.946674459180955e-05,
|
1765 |
+
"loss": 0.9275,
|
1766 |
+
"step": 251
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 0.8064,
|
1770 |
+
"grad_norm": 0.3427482665469463,
|
1771 |
+
"learning_rate": 1.8854421271990964e-05,
|
1772 |
+
"loss": 0.7372,
|
1773 |
+
"step": 252
|
1774 |
+
},
|
1775 |
+
{
|
1776 |
+
"epoch": 0.8096,
|
1777 |
+
"grad_norm": 0.3626863241157415,
|
1778 |
+
"learning_rate": 1.8250879009349398e-05,
|
1779 |
+
"loss": 0.8017,
|
1780 |
+
"step": 253
|
1781 |
+
},
|
1782 |
+
{
|
1783 |
+
"epoch": 0.8128,
|
1784 |
+
"grad_norm": 0.34605782828987924,
|
1785 |
+
"learning_rate": 1.7656183115380577e-05,
|
1786 |
+
"loss": 0.8484,
|
1787 |
+
"step": 254
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 0.816,
|
1791 |
+
"grad_norm": 0.4383327443511539,
|
1792 |
+
"learning_rate": 1.707039794428259e-05,
|
1793 |
+
"loss": 0.8186,
|
1794 |
+
"step": 255
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 0.8192,
|
1798 |
+
"grad_norm": 0.3447222581311919,
|
1799 |
+
"learning_rate": 1.649358688599191e-05,
|
1800 |
+
"loss": 0.7213,
|
1801 |
+
"step": 256
|
1802 |
+
},
|
1803 |
+
{
|
1804 |
+
"epoch": 0.8224,
|
1805 |
+
"grad_norm": 0.31970536995245263,
|
1806 |
+
"learning_rate": 1.5925812359323745e-05,
|
1807 |
+
"loss": 0.7434,
|
1808 |
+
"step": 257
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 0.8256,
|
1812 |
+
"grad_norm": 0.3656450810361097,
|
1813 |
+
"learning_rate": 1.5367135805217458e-05,
|
1814 |
+
"loss": 0.7676,
|
1815 |
+
"step": 258
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 0.8288,
|
1819 |
+
"grad_norm": 0.36298269369872477,
|
1820 |
+
"learning_rate": 1.4817617680087825e-05,
|
1821 |
+
"loss": 0.7487,
|
1822 |
+
"step": 259
|
1823 |
+
},
|
1824 |
+
{
|
1825 |
+
"epoch": 0.832,
|
1826 |
+
"grad_norm": 0.42152653173895593,
|
1827 |
+
"learning_rate": 1.4277317449282834e-05,
|
1828 |
+
"loss": 0.8926,
|
1829 |
+
"step": 260
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 0.8352,
|
1833 |
+
"grad_norm": 0.38946242385235036,
|
1834 |
+
"learning_rate": 1.3746293580648717e-05,
|
1835 |
+
"loss": 0.8176,
|
1836 |
+
"step": 261
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 0.8384,
|
1840 |
+
"grad_norm": 0.41804903127540954,
|
1841 |
+
"learning_rate": 1.3224603538202929e-05,
|
1842 |
+
"loss": 0.7996,
|
1843 |
+
"step": 262
|
1844 |
+
},
|
1845 |
+
{
|
1846 |
+
"epoch": 0.8416,
|
1847 |
+
"grad_norm": 0.31093486947817817,
|
1848 |
+
"learning_rate": 1.2712303775915802e-05,
|
1849 |
+
"loss": 0.7247,
|
1850 |
+
"step": 263
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 0.8448,
|
1854 |
+
"grad_norm": 0.3187140163567531,
|
1855 |
+
"learning_rate": 1.220944973160133e-05,
|
1856 |
+
"loss": 0.7473,
|
1857 |
+
"step": 264
|
1858 |
+
},
|
1859 |
+
{
|
1860 |
+
"epoch": 0.848,
|
1861 |
+
"grad_norm": 0.3872254324451249,
|
1862 |
+
"learning_rate": 1.1716095820918216e-05,
|
1863 |
+
"loss": 0.8821,
|
1864 |
+
"step": 265
|
1865 |
+
},
|
1866 |
+
{
|
1867 |
+
"epoch": 0.8512,
|
1868 |
+
"grad_norm": 0.4051984951107359,
|
1869 |
+
"learning_rate": 1.1232295431481222e-05,
|
1870 |
+
"loss": 0.7603,
|
1871 |
+
"step": 266
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 0.8544,
|
1875 |
+
"grad_norm": 0.40122800893302063,
|
1876 |
+
"learning_rate": 1.0758100917083991e-05,
|
1877 |
+
"loss": 0.8414,
|
1878 |
+
"step": 267
|
1879 |
+
},
|
1880 |
+
{
|
1881 |
+
"epoch": 0.8576,
|
1882 |
+
"grad_norm": 0.35538652933955217,
|
1883 |
+
"learning_rate": 1.0293563592033595e-05,
|
1884 |
+
"loss": 0.7676,
|
1885 |
+
"step": 268
|
1886 |
+
},
|
1887 |
+
{
|
1888 |
+
"epoch": 0.8608,
|
1889 |
+
"grad_norm": 0.3137491139738254,
|
1890 |
+
"learning_rate": 9.838733725597615e-06,
|
1891 |
+
"loss": 0.7501,
|
1892 |
+
"step": 269
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 0.864,
|
1896 |
+
"grad_norm": 0.45963025353581044,
|
1897 |
+
"learning_rate": 9.393660536564408e-06,
|
1898 |
+
"loss": 0.7936,
|
1899 |
+
"step": 270
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 0.8672,
|
1903 |
+
"grad_norm": 0.3348526950988372,
|
1904 |
+
"learning_rate": 8.958392187916841e-06,
|
1905 |
+
"loss": 0.736,
|
1906 |
+
"step": 271
|
1907 |
+
},
|
1908 |
+
{
|
1909 |
+
"epoch": 0.8704,
|
1910 |
+
"grad_norm": 0.39896905950270944,
|
1911 |
+
"learning_rate": 8.532975781620512e-06,
|
1912 |
+
"loss": 0.7654,
|
1913 |
+
"step": 272
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 0.8736,
|
1917 |
+
"grad_norm": 0.37722430967890513,
|
1918 |
+
"learning_rate": 8.117457353526625e-06,
|
1919 |
+
"loss": 0.7327,
|
1920 |
+
"step": 273
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 0.8768,
|
1924 |
+
"grad_norm": 0.31478006761399024,
|
1925 |
+
"learning_rate": 7.711881868390291e-06,
|
1926 |
+
"loss": 0.7615,
|
1927 |
+
"step": 274
|
1928 |
+
},
|
1929 |
+
{
|
1930 |
+
"epoch": 0.88,
|
1931 |
+
"grad_norm": 0.4188263703262344,
|
1932 |
+
"learning_rate": 7.3162932150046885e-06,
|
1933 |
+
"loss": 0.8783,
|
1934 |
+
"step": 275
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 0.8832,
|
1938 |
+
"grad_norm": 0.31942616334359136,
|
1939 |
+
"learning_rate": 6.930734201451816e-06,
|
1940 |
+
"loss": 0.725,
|
1941 |
+
"step": 276
|
1942 |
+
},
|
1943 |
+
{
|
1944 |
+
"epoch": 0.8864,
|
1945 |
+
"grad_norm": 0.2997679274265762,
|
1946 |
+
"learning_rate": 6.555246550469907e-06,
|
1947 |
+
"loss": 0.7615,
|
1948 |
+
"step": 277
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 0.8896,
|
1952 |
+
"grad_norm": 0.36551094625988517,
|
1953 |
+
"learning_rate": 6.189870894938587e-06,
|
1954 |
+
"loss": 0.7969,
|
1955 |
+
"step": 278
|
1956 |
+
},
|
1957 |
+
{
|
1958 |
+
"epoch": 0.8928,
|
1959 |
+
"grad_norm": 0.34379292762124014,
|
1960 |
+
"learning_rate": 5.834646773481811e-06,
|
1961 |
+
"loss": 0.8058,
|
1962 |
+
"step": 279
|
1963 |
+
},
|
1964 |
+
{
|
1965 |
+
"epoch": 0.896,
|
1966 |
+
"grad_norm": 0.35790166821440883,
|
1967 |
+
"learning_rate": 5.489612626189245e-06,
|
1968 |
+
"loss": 0.7782,
|
1969 |
+
"step": 280
|
1970 |
+
},
|
1971 |
+
{
|
1972 |
+
"epoch": 0.8992,
|
1973 |
+
"grad_norm": 0.33879760718838586,
|
1974 |
+
"learning_rate": 5.154805790456485e-06,
|
1975 |
+
"loss": 0.7636,
|
1976 |
+
"step": 281
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 0.9024,
|
1980 |
+
"grad_norm": 0.31407783216446544,
|
1981 |
+
"learning_rate": 4.830262496944693e-06,
|
1982 |
+
"loss": 0.7323,
|
1983 |
+
"step": 282
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 0.9056,
|
1987 |
+
"grad_norm": 0.38757334898188933,
|
1988 |
+
"learning_rate": 4.516017865659949e-06,
|
1989 |
+
"loss": 0.8443,
|
1990 |
+
"step": 283
|
1991 |
+
},
|
1992 |
+
{
|
1993 |
+
"epoch": 0.9088,
|
1994 |
+
"grad_norm": 0.39258218690616686,
|
1995 |
+
"learning_rate": 4.21210590215273e-06,
|
1996 |
+
"loss": 0.8223,
|
1997 |
+
"step": 284
|
1998 |
+
},
|
1999 |
+
{
|
2000 |
+
"epoch": 0.912,
|
2001 |
+
"grad_norm": 0.3313155165113673,
|
2002 |
+
"learning_rate": 3.918559493838114e-06,
|
2003 |
+
"loss": 0.7951,
|
2004 |
+
"step": 285
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 0.9152,
|
2008 |
+
"grad_norm": 0.32605323503196637,
|
2009 |
+
"learning_rate": 3.6354104064368566e-06,
|
2010 |
+
"loss": 0.799,
|
2011 |
+
"step": 286
|
2012 |
+
},
|
2013 |
+
{
|
2014 |
+
"epoch": 0.9184,
|
2015 |
+
"grad_norm": 0.393570523991794,
|
2016 |
+
"learning_rate": 3.3626892805379562e-06,
|
2017 |
+
"loss": 0.741,
|
2018 |
+
"step": 287
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 0.9216,
|
2022 |
+
"grad_norm": 0.37378520088314604,
|
2023 |
+
"learning_rate": 3.100425628282899e-06,
|
2024 |
+
"loss": 0.7429,
|
2025 |
+
"step": 288
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 0.9248,
|
2029 |
+
"grad_norm": 0.3179898573134195,
|
2030 |
+
"learning_rate": 2.848647830172024e-06,
|
2031 |
+
"loss": 0.7852,
|
2032 |
+
"step": 289
|
2033 |
+
},
|
2034 |
+
{
|
2035 |
+
"epoch": 0.928,
|
2036 |
+
"grad_norm": 0.33638562625723384,
|
2037 |
+
"learning_rate": 2.607383131993424e-06,
|
2038 |
+
"loss": 0.7518,
|
2039 |
+
"step": 290
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 0.9312,
|
2043 |
+
"grad_norm": 0.36906155605346674,
|
2044 |
+
"learning_rate": 2.3766576418745022e-06,
|
2045 |
+
"loss": 0.7948,
|
2046 |
+
"step": 291
|
2047 |
+
},
|
2048 |
+
{
|
2049 |
+
"epoch": 0.9344,
|
2050 |
+
"grad_norm": 0.5865294985065448,
|
2051 |
+
"learning_rate": 2.1564963274568027e-06,
|
2052 |
+
"loss": 0.7865,
|
2053 |
+
"step": 292
|
2054 |
+
},
|
2055 |
+
{
|
2056 |
+
"epoch": 0.9376,
|
2057 |
+
"grad_norm": 0.3357869205531904,
|
2058 |
+
"learning_rate": 1.9469230131940907e-06,
|
2059 |
+
"loss": 0.8067,
|
2060 |
+
"step": 293
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 0.9408,
|
2064 |
+
"grad_norm": 0.3214550435563662,
|
2065 |
+
"learning_rate": 1.7479603777742938e-06,
|
2066 |
+
"loss": 0.7578,
|
2067 |
+
"step": 294
|
2068 |
+
},
|
2069 |
+
{
|
2070 |
+
"epoch": 0.944,
|
2071 |
+
"grad_norm": 0.3617697253780003,
|
2072 |
+
"learning_rate": 1.559629951665298e-06,
|
2073 |
+
"loss": 0.7783,
|
2074 |
+
"step": 295
|
2075 |
+
},
|
2076 |
+
{
|
2077 |
+
"epoch": 0.9472,
|
2078 |
+
"grad_norm": 0.3851342510588961,
|
2079 |
+
"learning_rate": 1.3819521147851123e-06,
|
2080 |
+
"loss": 0.7932,
|
2081 |
+
"step": 296
|
2082 |
+
},
|
2083 |
+
{
|
2084 |
+
"epoch": 0.9504,
|
2085 |
+
"grad_norm": 0.34426702777854773,
|
2086 |
+
"learning_rate": 1.2149460942964098e-06,
|
2087 |
+
"loss": 0.7609,
|
2088 |
+
"step": 297
|
2089 |
+
},
|
2090 |
+
{
|
2091 |
+
"epoch": 0.9536,
|
2092 |
+
"grad_norm": 0.36253585088408663,
|
2093 |
+
"learning_rate": 1.05862996252597e-06,
|
2094 |
+
"loss": 0.7984,
|
2095 |
+
"step": 298
|
2096 |
+
},
|
2097 |
+
{
|
2098 |
+
"epoch": 0.9568,
|
2099 |
+
"grad_norm": 0.4054570274935403,
|
2100 |
+
"learning_rate": 9.130206350089765e-07,
|
2101 |
+
"loss": 0.8,
|
2102 |
+
"step": 299
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 0.96,
|
2106 |
+
"grad_norm": 0.34938837916613863,
|
2107 |
+
"learning_rate": 7.781338686584927e-07,
|
2108 |
+
"loss": 0.8201,
|
2109 |
+
"step": 300
|
2110 |
+
},
|
2111 |
+
{
|
2112 |
+
"epoch": 0.9632,
|
2113 |
+
"grad_norm": 0.37777115892854124,
|
2114 |
+
"learning_rate": 6.539842600603918e-07,
|
2115 |
+
"loss": 0.8213,
|
2116 |
+
"step": 301
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 0.9664,
|
2120 |
+
"grad_norm": 0.37766797195751295,
|
2121 |
+
"learning_rate": 5.405852438937764e-07,
|
2122 |
+
"loss": 0.7736,
|
2123 |
+
"step": 302
|
2124 |
+
},
|
2125 |
+
{
|
2126 |
+
"epoch": 0.9696,
|
2127 |
+
"grad_norm": 0.34012917606966536,
|
2128 |
+
"learning_rate": 4.3794909147720773e-07,
|
2129 |
+
"loss": 0.7852,
|
2130 |
+
"step": 303
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 0.9728,
|
2134 |
+
"grad_norm": 0.39195816403535333,
|
2135 |
+
"learning_rate": 3.4608690944071263e-07,
|
2136 |
+
"loss": 0.7284,
|
2137 |
+
"step": 304
|
2138 |
+
},
|
2139 |
+
{
|
2140 |
+
"epoch": 0.976,
|
2141 |
+
"grad_norm": 0.38813789832520107,
|
2142 |
+
"learning_rate": 2.6500863852395584e-07,
|
2143 |
+
"loss": 0.8645,
|
2144 |
+
"step": 305
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 0.9792,
|
2148 |
+
"grad_norm": 0.32915922567617034,
|
2149 |
+
"learning_rate": 1.947230525005006e-07,
|
2150 |
+
"loss": 0.7437,
|
2151 |
+
"step": 306
|
2152 |
+
},
|
2153 |
+
{
|
2154 |
+
"epoch": 0.9824,
|
2155 |
+
"grad_norm": 0.3351156902277233,
|
2156 |
+
"learning_rate": 1.3523775722834587e-07,
|
2157 |
+
"loss": 0.7069,
|
2158 |
+
"step": 307
|
2159 |
+
},
|
2160 |
+
{
|
2161 |
+
"epoch": 0.9856,
|
2162 |
+
"grad_norm": 0.32461016981926744,
|
2163 |
+
"learning_rate": 8.655918982689581e-08,
|
2164 |
+
"loss": 0.7698,
|
2165 |
+
"step": 308
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 0.9888,
|
2169 |
+
"grad_norm": 0.34243368327755425,
|
2170 |
+
"learning_rate": 4.8692617980350406e-08,
|
2171 |
+
"loss": 0.8092,
|
2172 |
+
"step": 309
|
2173 |
+
},
|
2174 |
+
{
|
2175 |
+
"epoch": 0.992,
|
2176 |
+
"grad_norm": 0.35449923757603197,
|
2177 |
+
"learning_rate": 2.164213936770576e-08,
|
2178 |
+
"loss": 0.7537,
|
2179 |
+
"step": 310
|
2180 |
+
},
|
2181 |
+
{
|
2182 |
+
"epoch": 0.9952,
|
2183 |
+
"grad_norm": 0.37595905935200546,
|
2184 |
+
"learning_rate": 5.410681219286673e-09,
|
2185 |
+
"loss": 0.7974,
|
2186 |
+
"step": 311
|
2187 |
+
},
|
2188 |
+
{
|
2189 |
+
"epoch": 0.9984,
|
2190 |
+
"grad_norm": 0.3296041419119035,
|
2191 |
+
"learning_rate": 0.0,
|
2192 |
+
"loss": 0.7362,
|
2193 |
+
"step": 312
|
2194 |
+
},
|
2195 |
+
{
|
2196 |
+
"epoch": 0.9984,
|
2197 |
+
"step": 312,
|
2198 |
+
"total_flos": 730661894488064.0,
|
2199 |
+
"train_loss": 0.8330854674180349,
|
2200 |
+
"train_runtime": 9213.2243,
|
2201 |
+
"train_samples_per_second": 1.085,
|
2202 |
+
"train_steps_per_second": 0.034
|
2203 |
+
}
|
2204 |
+
],
|
2205 |
+
"logging_steps": 1.0,
|
2206 |
+
"max_steps": 312,
|
2207 |
+
"num_input_tokens_seen": 0,
|
2208 |
+
"num_train_epochs": 1,
|
2209 |
+
"save_steps": 500,
|
2210 |
+
"stateful_callbacks": {
|
2211 |
+
"TrainerControl": {
|
2212 |
+
"args": {
|
2213 |
+
"should_epoch_stop": false,
|
2214 |
+
"should_evaluate": false,
|
2215 |
+
"should_log": false,
|
2216 |
+
"should_save": false,
|
2217 |
+
"should_training_stop": false
|
2218 |
+
},
|
2219 |
+
"attributes": {}
|
2220 |
+
}
|
2221 |
+
},
|
2222 |
+
"total_flos": 730661894488064.0,
|
2223 |
+
"train_batch_size": 8,
|
2224 |
+
"trial_name": null,
|
2225 |
+
"trial_params": null
|
2226 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: ./weights/Bunny-v1_1-Llama-3-8B-V
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.11.1
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 256,
|
14 |
+
"lora_dropout": 0.1,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 128,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": null,
|
22 |
+
"target_modules": [
|
23 |
+
"up_proj",
|
24 |
+
"o_proj",
|
25 |
+
"q_proj",
|
26 |
+
"down_proj",
|
27 |
+
"v_proj",
|
28 |
+
"k_proj",
|
29 |
+
"gate_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bb223c3820e4d3ba11a4de32d2e8233430f8ee290e3539d12aefaa3e3db8d9e5
|
3 |
+
size 671150064
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/config.json
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
3 |
+
"architectures": [
|
4 |
+
"BunnyLlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"auto_map": {
|
9 |
+
"AutoConfig": "configuration_bunny_llama.BunnyLlamaConfig",
|
10 |
+
"AutoModelForCausalLM": "modeling_bunny_llama.BunnyLlamaForCausalLM"
|
11 |
+
},
|
12 |
+
"bos_token_id": 128000,
|
13 |
+
"continuous_training": false,
|
14 |
+
"eos_token_id": 128001,
|
15 |
+
"freeze_mm_mlp_adapter": false,
|
16 |
+
"hidden_act": "silu",
|
17 |
+
"hidden_size": 4096,
|
18 |
+
"image_aspect_ratio": "pad",
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 14336,
|
21 |
+
"max_position_embeddings": 8192,
|
22 |
+
"mm_hidden_size": 3456,
|
23 |
+
"mm_projector_lr": null,
|
24 |
+
"mm_projector_type": "mlp2x_gelu",
|
25 |
+
"mm_vision_tower": "./weights/siglip-so400m-patch14-384",
|
26 |
+
"model_type": "bunny-llama",
|
27 |
+
"num_attention_heads": 32,
|
28 |
+
"num_hidden_layers": 32,
|
29 |
+
"num_key_value_heads": 8,
|
30 |
+
"pretraining_tp": 1,
|
31 |
+
"rms_norm_eps": 1e-05,
|
32 |
+
"rope_scaling": null,
|
33 |
+
"rope_theta": 500000.0,
|
34 |
+
"tie_word_embeddings": false,
|
35 |
+
"tokenizer_model_max_length": 2048,
|
36 |
+
"tokenizer_padding_side": "right",
|
37 |
+
"torch_dtype": "float16",
|
38 |
+
"transformers_version": "4.41.2",
|
39 |
+
"tune_mm_mlp_adapter": false,
|
40 |
+
"unfreeze_vision_tower": true,
|
41 |
+
"use_cache": true,
|
42 |
+
"use_mm_proj": true,
|
43 |
+
"use_s2": true,
|
44 |
+
"vocab_size": 128256
|
45 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/non_lora_trainables.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7a0dab0cefd97719e632dbe34186b207b24721cd439d644851337f06d4b9e2aa
|
3 |
+
size 918507402
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_2_lora/trainer_state.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: ./weights/Bunny-v1_1-Llama-3-8B-V
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.11.1
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 256,
|
14 |
+
"lora_dropout": 0.1,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 128,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": null,
|
22 |
+
"target_modules": [
|
23 |
+
"q_proj",
|
24 |
+
"up_proj",
|
25 |
+
"k_proj",
|
26 |
+
"o_proj",
|
27 |
+
"v_proj",
|
28 |
+
"down_proj",
|
29 |
+
"gate_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b97c5d7de491d336dcc2b1c913eec57bc2ade51a9e0214178292b044b106f4bf
|
3 |
+
size 671150064
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/config.json
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
3 |
+
"architectures": [
|
4 |
+
"BunnyLlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"auto_map": {
|
9 |
+
"AutoConfig": "configuration_bunny_llama.BunnyLlamaConfig",
|
10 |
+
"AutoModelForCausalLM": "modeling_bunny_llama.BunnyLlamaForCausalLM"
|
11 |
+
},
|
12 |
+
"bos_token_id": 128000,
|
13 |
+
"continuous_training": false,
|
14 |
+
"eos_token_id": 128001,
|
15 |
+
"freeze_mm_mlp_adapter": false,
|
16 |
+
"hidden_act": "silu",
|
17 |
+
"hidden_size": 4096,
|
18 |
+
"image_aspect_ratio": "pad",
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 14336,
|
21 |
+
"max_position_embeddings": 8192,
|
22 |
+
"mm_hidden_size": 3456,
|
23 |
+
"mm_projector_lr": null,
|
24 |
+
"mm_projector_type": "mlp2x_gelu",
|
25 |
+
"mm_vision_tower": "./weights/siglip-so400m-patch14-384",
|
26 |
+
"model_type": "bunny-llama",
|
27 |
+
"num_attention_heads": 32,
|
28 |
+
"num_hidden_layers": 32,
|
29 |
+
"num_key_value_heads": 8,
|
30 |
+
"pretraining_tp": 1,
|
31 |
+
"rms_norm_eps": 1e-05,
|
32 |
+
"rope_scaling": null,
|
33 |
+
"rope_theta": 500000.0,
|
34 |
+
"tie_word_embeddings": false,
|
35 |
+
"tokenizer_model_max_length": 2048,
|
36 |
+
"tokenizer_padding_side": "right",
|
37 |
+
"torch_dtype": "float16",
|
38 |
+
"transformers_version": "4.41.2",
|
39 |
+
"tune_mm_mlp_adapter": false,
|
40 |
+
"unfreeze_vision_tower": true,
|
41 |
+
"use_cache": true,
|
42 |
+
"use_mm_proj": true,
|
43 |
+
"use_s2": true,
|
44 |
+
"vocab_size": 128256
|
45 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/non_lora_trainables.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:510b7354cbdb44580b8fcff4c0cba0358329aad7b5f7e48dcb45928b558fe28f
|
3 |
+
size 918507402
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_2_epochs_1_GA_4_lora/trainer_state.json
ADDED
@@ -0,0 +1,2226 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9984,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 312,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0032,
|
13 |
+
"grad_norm": 0.8210982159609682,
|
14 |
+
"learning_rate": 2e-05,
|
15 |
+
"loss": 1.3329,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.0064,
|
20 |
+
"grad_norm": 0.7751787218518346,
|
21 |
+
"learning_rate": 4e-05,
|
22 |
+
"loss": 1.3721,
|
23 |
+
"step": 2
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.0096,
|
27 |
+
"grad_norm": 0.6646539221769208,
|
28 |
+
"learning_rate": 6e-05,
|
29 |
+
"loss": 1.3088,
|
30 |
+
"step": 3
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.0128,
|
34 |
+
"grad_norm": 0.7036195246304789,
|
35 |
+
"learning_rate": 8e-05,
|
36 |
+
"loss": 1.3377,
|
37 |
+
"step": 4
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.016,
|
41 |
+
"grad_norm": 0.6168233543536914,
|
42 |
+
"learning_rate": 0.0001,
|
43 |
+
"loss": 1.1919,
|
44 |
+
"step": 5
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.0192,
|
48 |
+
"grad_norm": 0.8492868526153039,
|
49 |
+
"learning_rate": 0.00012,
|
50 |
+
"loss": 1.2512,
|
51 |
+
"step": 6
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.0224,
|
55 |
+
"grad_norm": 0.7936611575617178,
|
56 |
+
"learning_rate": 0.00014,
|
57 |
+
"loss": 1.0697,
|
58 |
+
"step": 7
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.0256,
|
62 |
+
"grad_norm": 0.5862560941923312,
|
63 |
+
"learning_rate": 0.00016,
|
64 |
+
"loss": 1.0911,
|
65 |
+
"step": 8
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.0288,
|
69 |
+
"grad_norm": 0.49328071163938675,
|
70 |
+
"learning_rate": 0.00018,
|
71 |
+
"loss": 0.9762,
|
72 |
+
"step": 9
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.032,
|
76 |
+
"grad_norm": 0.4796953108328627,
|
77 |
+
"learning_rate": 0.0002,
|
78 |
+
"loss": 0.9269,
|
79 |
+
"step": 10
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.0352,
|
83 |
+
"grad_norm": 0.47431010608538954,
|
84 |
+
"learning_rate": 0.00019999458931878073,
|
85 |
+
"loss": 1.032,
|
86 |
+
"step": 11
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.0384,
|
90 |
+
"grad_norm": 0.4707062597281157,
|
91 |
+
"learning_rate": 0.0001999783578606323,
|
92 |
+
"loss": 0.9748,
|
93 |
+
"step": 12
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.0416,
|
97 |
+
"grad_norm": 0.4880961566615289,
|
98 |
+
"learning_rate": 0.00019995130738201966,
|
99 |
+
"loss": 0.9767,
|
100 |
+
"step": 13
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.0448,
|
104 |
+
"grad_norm": 0.4658440438656818,
|
105 |
+
"learning_rate": 0.0001999134408101731,
|
106 |
+
"loss": 0.9179,
|
107 |
+
"step": 14
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.048,
|
111 |
+
"grad_norm": 0.5344941086238363,
|
112 |
+
"learning_rate": 0.00019986476224277165,
|
113 |
+
"loss": 1.0129,
|
114 |
+
"step": 15
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.0512,
|
118 |
+
"grad_norm": 0.5007192463357719,
|
119 |
+
"learning_rate": 0.00019980527694749952,
|
120 |
+
"loss": 0.9953,
|
121 |
+
"step": 16
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.0544,
|
125 |
+
"grad_norm": 0.44861347739015633,
|
126 |
+
"learning_rate": 0.00019973499136147606,
|
127 |
+
"loss": 0.8987,
|
128 |
+
"step": 17
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.0576,
|
132 |
+
"grad_norm": 0.40432515099975497,
|
133 |
+
"learning_rate": 0.0001996539130905593,
|
134 |
+
"loss": 0.9949,
|
135 |
+
"step": 18
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.0608,
|
139 |
+
"grad_norm": 0.42728242685352247,
|
140 |
+
"learning_rate": 0.0001995620509085228,
|
141 |
+
"loss": 0.9649,
|
142 |
+
"step": 19
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.064,
|
146 |
+
"grad_norm": 0.4641229091681979,
|
147 |
+
"learning_rate": 0.00019945941475610623,
|
148 |
+
"loss": 0.9609,
|
149 |
+
"step": 20
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.0672,
|
153 |
+
"grad_norm": 0.3913009193318844,
|
154 |
+
"learning_rate": 0.0001993460157399396,
|
155 |
+
"loss": 0.8997,
|
156 |
+
"step": 21
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.0704,
|
160 |
+
"grad_norm": 0.3960221279711468,
|
161 |
+
"learning_rate": 0.0001992218661313415,
|
162 |
+
"loss": 0.9046,
|
163 |
+
"step": 22
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.0736,
|
167 |
+
"grad_norm": 0.4893549306818581,
|
168 |
+
"learning_rate": 0.00019908697936499103,
|
169 |
+
"loss": 0.9538,
|
170 |
+
"step": 23
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.0768,
|
174 |
+
"grad_norm": 0.3943436800759236,
|
175 |
+
"learning_rate": 0.00019894137003747403,
|
176 |
+
"loss": 0.85,
|
177 |
+
"step": 24
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.08,
|
181 |
+
"grad_norm": 0.49281687959276665,
|
182 |
+
"learning_rate": 0.00019878505390570362,
|
183 |
+
"loss": 0.8904,
|
184 |
+
"step": 25
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.0832,
|
188 |
+
"grad_norm": 0.45164645594942804,
|
189 |
+
"learning_rate": 0.00019861804788521493,
|
190 |
+
"loss": 0.9793,
|
191 |
+
"step": 26
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.0864,
|
195 |
+
"grad_norm": 0.3884822532677797,
|
196 |
+
"learning_rate": 0.00019844037004833473,
|
197 |
+
"loss": 0.8047,
|
198 |
+
"step": 27
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.0896,
|
202 |
+
"grad_norm": 0.41101776336511836,
|
203 |
+
"learning_rate": 0.00019825203962222572,
|
204 |
+
"loss": 0.9077,
|
205 |
+
"step": 28
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.0928,
|
209 |
+
"grad_norm": 1.3086028478450786,
|
210 |
+
"learning_rate": 0.0001980530769868059,
|
211 |
+
"loss": 0.9363,
|
212 |
+
"step": 29
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.096,
|
216 |
+
"grad_norm": 0.3375555003941169,
|
217 |
+
"learning_rate": 0.00019784350367254322,
|
218 |
+
"loss": 0.8523,
|
219 |
+
"step": 30
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.0992,
|
223 |
+
"grad_norm": 0.4356929720245981,
|
224 |
+
"learning_rate": 0.0001976233423581255,
|
225 |
+
"loss": 0.9504,
|
226 |
+
"step": 31
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.1024,
|
230 |
+
"grad_norm": 0.4558256010915256,
|
231 |
+
"learning_rate": 0.0001973926168680066,
|
232 |
+
"loss": 0.9619,
|
233 |
+
"step": 32
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.1056,
|
237 |
+
"grad_norm": 0.4182066475020067,
|
238 |
+
"learning_rate": 0.00019715135216982798,
|
239 |
+
"loss": 0.853,
|
240 |
+
"step": 33
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.1088,
|
244 |
+
"grad_norm": 0.42581671871982196,
|
245 |
+
"learning_rate": 0.0001968995743717171,
|
246 |
+
"loss": 0.9528,
|
247 |
+
"step": 34
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.112,
|
251 |
+
"grad_norm": 0.39591765873363666,
|
252 |
+
"learning_rate": 0.00019663731071946206,
|
253 |
+
"loss": 0.9005,
|
254 |
+
"step": 35
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.1152,
|
258 |
+
"grad_norm": 0.39119354380322896,
|
259 |
+
"learning_rate": 0.00019636458959356316,
|
260 |
+
"loss": 0.8988,
|
261 |
+
"step": 36
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.1184,
|
265 |
+
"grad_norm": 0.40090914965423335,
|
266 |
+
"learning_rate": 0.0001960814405061619,
|
267 |
+
"loss": 0.8758,
|
268 |
+
"step": 37
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.1216,
|
272 |
+
"grad_norm": 0.3945361491838484,
|
273 |
+
"learning_rate": 0.00019578789409784727,
|
274 |
+
"loss": 0.9393,
|
275 |
+
"step": 38
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.1248,
|
279 |
+
"grad_norm": 0.4394341724704068,
|
280 |
+
"learning_rate": 0.00019548398213434007,
|
281 |
+
"loss": 0.9428,
|
282 |
+
"step": 39
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.128,
|
286 |
+
"grad_norm": 0.3727456121947463,
|
287 |
+
"learning_rate": 0.00019516973750305532,
|
288 |
+
"loss": 0.8748,
|
289 |
+
"step": 40
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.1312,
|
293 |
+
"grad_norm": 0.37234199586270694,
|
294 |
+
"learning_rate": 0.00019484519420954354,
|
295 |
+
"loss": 0.8736,
|
296 |
+
"step": 41
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.1344,
|
300 |
+
"grad_norm": 0.39842513634466276,
|
301 |
+
"learning_rate": 0.00019451038737381077,
|
302 |
+
"loss": 0.9022,
|
303 |
+
"step": 42
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.1376,
|
307 |
+
"grad_norm": 0.38128602501461806,
|
308 |
+
"learning_rate": 0.00019416535322651818,
|
309 |
+
"loss": 0.8208,
|
310 |
+
"step": 43
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.1408,
|
314 |
+
"grad_norm": 0.38025302062603994,
|
315 |
+
"learning_rate": 0.00019381012910506146,
|
316 |
+
"loss": 0.8832,
|
317 |
+
"step": 44
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.144,
|
321 |
+
"grad_norm": 0.34279246176384087,
|
322 |
+
"learning_rate": 0.00019344475344953012,
|
323 |
+
"loss": 0.79,
|
324 |
+
"step": 45
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.1472,
|
328 |
+
"grad_norm": 0.39891015144038994,
|
329 |
+
"learning_rate": 0.00019306926579854821,
|
330 |
+
"loss": 0.8637,
|
331 |
+
"step": 46
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.1504,
|
335 |
+
"grad_norm": 0.33599563185761405,
|
336 |
+
"learning_rate": 0.00019268370678499533,
|
337 |
+
"loss": 0.7904,
|
338 |
+
"step": 47
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.1536,
|
342 |
+
"grad_norm": 0.405639307195739,
|
343 |
+
"learning_rate": 0.0001922881181316097,
|
344 |
+
"loss": 0.8904,
|
345 |
+
"step": 48
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.1568,
|
349 |
+
"grad_norm": 0.347008173492179,
|
350 |
+
"learning_rate": 0.00019188254264647337,
|
351 |
+
"loss": 0.8188,
|
352 |
+
"step": 49
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.16,
|
356 |
+
"grad_norm": 0.4053373760955314,
|
357 |
+
"learning_rate": 0.0001914670242183795,
|
358 |
+
"loss": 0.8767,
|
359 |
+
"step": 50
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.1632,
|
363 |
+
"grad_norm": 0.4071076068532092,
|
364 |
+
"learning_rate": 0.0001910416078120832,
|
365 |
+
"loss": 0.8753,
|
366 |
+
"step": 51
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.1664,
|
370 |
+
"grad_norm": 0.3947573279192353,
|
371 |
+
"learning_rate": 0.0001906063394634356,
|
372 |
+
"loss": 0.9461,
|
373 |
+
"step": 52
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.1696,
|
377 |
+
"grad_norm": 0.41593394777293125,
|
378 |
+
"learning_rate": 0.00019016126627440237,
|
379 |
+
"loss": 0.8543,
|
380 |
+
"step": 53
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.1728,
|
384 |
+
"grad_norm": 0.34556265770047007,
|
385 |
+
"learning_rate": 0.00018970643640796642,
|
386 |
+
"loss": 0.8457,
|
387 |
+
"step": 54
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.176,
|
391 |
+
"grad_norm": 0.4135992491712443,
|
392 |
+
"learning_rate": 0.000189241899082916,
|
393 |
+
"loss": 0.8471,
|
394 |
+
"step": 55
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.1792,
|
398 |
+
"grad_norm": 0.36656186910259925,
|
399 |
+
"learning_rate": 0.00018876770456851877,
|
400 |
+
"loss": 0.8315,
|
401 |
+
"step": 56
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.1824,
|
405 |
+
"grad_norm": 0.41888747530592474,
|
406 |
+
"learning_rate": 0.0001882839041790818,
|
407 |
+
"loss": 0.9118,
|
408 |
+
"step": 57
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.1856,
|
412 |
+
"grad_norm": 0.36251863953206737,
|
413 |
+
"learning_rate": 0.00018779055026839868,
|
414 |
+
"loss": 0.8626,
|
415 |
+
"step": 58
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.1888,
|
419 |
+
"grad_norm": 0.3748079918860433,
|
420 |
+
"learning_rate": 0.00018728769622408423,
|
421 |
+
"loss": 0.9207,
|
422 |
+
"step": 59
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.192,
|
426 |
+
"grad_norm": 0.34699582513061866,
|
427 |
+
"learning_rate": 0.00018677539646179707,
|
428 |
+
"loss": 0.7823,
|
429 |
+
"step": 60
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.1952,
|
433 |
+
"grad_norm": 0.4324436996517203,
|
434 |
+
"learning_rate": 0.00018625370641935129,
|
435 |
+
"loss": 0.9012,
|
436 |
+
"step": 61
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.1984,
|
440 |
+
"grad_norm": 0.40830183945729825,
|
441 |
+
"learning_rate": 0.00018572268255071718,
|
442 |
+
"loss": 0.8866,
|
443 |
+
"step": 62
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.2016,
|
447 |
+
"grad_norm": 0.35645475792133624,
|
448 |
+
"learning_rate": 0.00018518238231991218,
|
449 |
+
"loss": 0.8246,
|
450 |
+
"step": 63
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.2048,
|
454 |
+
"grad_norm": 0.39501413519318324,
|
455 |
+
"learning_rate": 0.00018463286419478255,
|
456 |
+
"loss": 0.8903,
|
457 |
+
"step": 64
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.208,
|
461 |
+
"grad_norm": 0.3403250694314037,
|
462 |
+
"learning_rate": 0.00018407418764067627,
|
463 |
+
"loss": 0.8002,
|
464 |
+
"step": 65
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.2112,
|
468 |
+
"grad_norm": 0.38578492015884985,
|
469 |
+
"learning_rate": 0.00018350641311400812,
|
470 |
+
"loss": 0.8823,
|
471 |
+
"step": 66
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.2144,
|
475 |
+
"grad_norm": 0.42912996214481597,
|
476 |
+
"learning_rate": 0.0001829296020557174,
|
477 |
+
"loss": 0.9337,
|
478 |
+
"step": 67
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.2176,
|
482 |
+
"grad_norm": 0.33840309416398107,
|
483 |
+
"learning_rate": 0.00018234381688461942,
|
484 |
+
"loss": 0.8152,
|
485 |
+
"step": 68
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.2208,
|
489 |
+
"grad_norm": 0.36261675945692257,
|
490 |
+
"learning_rate": 0.0001817491209906506,
|
491 |
+
"loss": 0.8119,
|
492 |
+
"step": 69
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.224,
|
496 |
+
"grad_norm": 0.3978630764800173,
|
497 |
+
"learning_rate": 0.00018114557872800905,
|
498 |
+
"loss": 0.8884,
|
499 |
+
"step": 70
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.2272,
|
503 |
+
"grad_norm": 0.4067690308983075,
|
504 |
+
"learning_rate": 0.00018053325540819045,
|
505 |
+
"loss": 0.888,
|
506 |
+
"step": 71
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.2304,
|
510 |
+
"grad_norm": 0.39761946908218565,
|
511 |
+
"learning_rate": 0.0001799122172929206,
|
512 |
+
"loss": 0.8734,
|
513 |
+
"step": 72
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.2336,
|
517 |
+
"grad_norm": 0.416514669804446,
|
518 |
+
"learning_rate": 0.00017928253158698473,
|
519 |
+
"loss": 0.8813,
|
520 |
+
"step": 73
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.2368,
|
524 |
+
"grad_norm": 0.3831820092966067,
|
525 |
+
"learning_rate": 0.0001786442664309554,
|
526 |
+
"loss": 0.8015,
|
527 |
+
"step": 74
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.24,
|
531 |
+
"grad_norm": 0.40531349166429315,
|
532 |
+
"learning_rate": 0.0001779974908938184,
|
533 |
+
"loss": 0.8021,
|
534 |
+
"step": 75
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.2432,
|
538 |
+
"grad_norm": 0.3757725806826467,
|
539 |
+
"learning_rate": 0.0001773422749654988,
|
540 |
+
"loss": 0.7819,
|
541 |
+
"step": 76
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.2464,
|
545 |
+
"grad_norm": 0.3973182683957758,
|
546 |
+
"learning_rate": 0.00017667868954928694,
|
547 |
+
"loss": 0.8807,
|
548 |
+
"step": 77
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.2496,
|
552 |
+
"grad_norm": 0.3251584668348043,
|
553 |
+
"learning_rate": 0.00017600680645416583,
|
554 |
+
"loss": 0.7627,
|
555 |
+
"step": 78
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.2528,
|
559 |
+
"grad_norm": 0.41127123846488717,
|
560 |
+
"learning_rate": 0.00017532669838704035,
|
561 |
+
"loss": 0.8387,
|
562 |
+
"step": 79
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.256,
|
566 |
+
"grad_norm": 0.37661663005517965,
|
567 |
+
"learning_rate": 0.00017463843894486937,
|
568 |
+
"loss": 0.8423,
|
569 |
+
"step": 80
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.2592,
|
573 |
+
"grad_norm": 0.5786386678521719,
|
574 |
+
"learning_rate": 0.0001739421026067017,
|
575 |
+
"loss": 0.8949,
|
576 |
+
"step": 81
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.2624,
|
580 |
+
"grad_norm": 0.362252109033277,
|
581 |
+
"learning_rate": 0.00017323776472561627,
|
582 |
+
"loss": 0.8644,
|
583 |
+
"step": 82
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.2656,
|
587 |
+
"grad_norm": 0.3786074251460458,
|
588 |
+
"learning_rate": 0.00017252550152056795,
|
589 |
+
"loss": 0.8161,
|
590 |
+
"step": 83
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.2688,
|
594 |
+
"grad_norm": 0.38704773419959004,
|
595 |
+
"learning_rate": 0.0001718053900681397,
|
596 |
+
"loss": 0.8054,
|
597 |
+
"step": 84
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.272,
|
601 |
+
"grad_norm": 0.36687924172159914,
|
602 |
+
"learning_rate": 0.00017107750829420176,
|
603 |
+
"loss": 0.8053,
|
604 |
+
"step": 85
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.2752,
|
608 |
+
"grad_norm": 0.4532905914552837,
|
609 |
+
"learning_rate": 0.00017034193496547902,
|
610 |
+
"loss": 0.8496,
|
611 |
+
"step": 86
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.2784,
|
615 |
+
"grad_norm": 0.3868107415971005,
|
616 |
+
"learning_rate": 0.00016959874968102735,
|
617 |
+
"loss": 0.8292,
|
618 |
+
"step": 87
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.2816,
|
622 |
+
"grad_norm": 0.45004745878752395,
|
623 |
+
"learning_rate": 0.00016884803286362,
|
624 |
+
"loss": 0.8386,
|
625 |
+
"step": 88
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.2848,
|
629 |
+
"grad_norm": 0.3601950439013075,
|
630 |
+
"learning_rate": 0.00016808986575104465,
|
631 |
+
"loss": 0.8144,
|
632 |
+
"step": 89
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.288,
|
636 |
+
"grad_norm": 0.39414164051104805,
|
637 |
+
"learning_rate": 0.00016732433038731242,
|
638 |
+
"loss": 0.7563,
|
639 |
+
"step": 90
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.2912,
|
643 |
+
"grad_norm": 0.4162252396854745,
|
644 |
+
"learning_rate": 0.0001665515096137797,
|
645 |
+
"loss": 0.79,
|
646 |
+
"step": 91
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.2944,
|
650 |
+
"grad_norm": 0.37960017989400824,
|
651 |
+
"learning_rate": 0.00016577148706018328,
|
652 |
+
"loss": 0.8201,
|
653 |
+
"step": 92
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.2976,
|
657 |
+
"grad_norm": 0.3838649671026415,
|
658 |
+
"learning_rate": 0.00016498434713559088,
|
659 |
+
"loss": 0.8362,
|
660 |
+
"step": 93
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.3008,
|
664 |
+
"grad_norm": 0.43789279418131016,
|
665 |
+
"learning_rate": 0.00016419017501926656,
|
666 |
+
"loss": 0.8057,
|
667 |
+
"step": 94
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.304,
|
671 |
+
"grad_norm": 0.33824523290379066,
|
672 |
+
"learning_rate": 0.0001633890566514535,
|
673 |
+
"loss": 0.7832,
|
674 |
+
"step": 95
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.3072,
|
678 |
+
"grad_norm": 0.38508464229871187,
|
679 |
+
"learning_rate": 0.00016258107872407375,
|
680 |
+
"loss": 0.8825,
|
681 |
+
"step": 96
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.3104,
|
685 |
+
"grad_norm": 0.3872280278580832,
|
686 |
+
"learning_rate": 0.0001617663286713474,
|
687 |
+
"loss": 0.8568,
|
688 |
+
"step": 97
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.3136,
|
692 |
+
"grad_norm": 0.3710845404668928,
|
693 |
+
"learning_rate": 0.00016094489466033043,
|
694 |
+
"loss": 0.8433,
|
695 |
+
"step": 98
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.3168,
|
699 |
+
"grad_norm": 0.3601850097818311,
|
700 |
+
"learning_rate": 0.00016011686558137448,
|
701 |
+
"loss": 0.8426,
|
702 |
+
"step": 99
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.32,
|
706 |
+
"grad_norm": 0.36336268281281814,
|
707 |
+
"learning_rate": 0.0001592823310385073,
|
708 |
+
"loss": 0.8459,
|
709 |
+
"step": 100
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.3232,
|
713 |
+
"grad_norm": 0.4007311050026622,
|
714 |
+
"learning_rate": 0.0001584413813397364,
|
715 |
+
"loss": 0.8332,
|
716 |
+
"step": 101
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.3264,
|
720 |
+
"grad_norm": 0.3279742689663737,
|
721 |
+
"learning_rate": 0.00015759410748727662,
|
722 |
+
"loss": 0.7898,
|
723 |
+
"step": 102
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.3296,
|
727 |
+
"grad_norm": 0.3937884919475543,
|
728 |
+
"learning_rate": 0.00015674060116770236,
|
729 |
+
"loss": 0.8439,
|
730 |
+
"step": 103
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.3328,
|
734 |
+
"grad_norm": 0.4143083432212905,
|
735 |
+
"learning_rate": 0.00015588095474202595,
|
736 |
+
"loss": 0.8246,
|
737 |
+
"step": 104
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.336,
|
741 |
+
"grad_norm": 0.35676836384855826,
|
742 |
+
"learning_rate": 0.00015501526123570277,
|
743 |
+
"loss": 0.8377,
|
744 |
+
"step": 105
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.3392,
|
748 |
+
"grad_norm": 0.4418839450505997,
|
749 |
+
"learning_rate": 0.00015414361432856475,
|
750 |
+
"loss": 0.8546,
|
751 |
+
"step": 106
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.3424,
|
755 |
+
"grad_norm": 0.38722861928111846,
|
756 |
+
"learning_rate": 0.0001532661083446829,
|
757 |
+
"loss": 0.8938,
|
758 |
+
"step": 107
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.3456,
|
762 |
+
"grad_norm": 0.4798810748731012,
|
763 |
+
"learning_rate": 0.00015238283824216015,
|
764 |
+
"loss": 0.933,
|
765 |
+
"step": 108
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.3488,
|
769 |
+
"grad_norm": 0.4192610184794806,
|
770 |
+
"learning_rate": 0.00015149389960285558,
|
771 |
+
"loss": 0.8314,
|
772 |
+
"step": 109
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.352,
|
776 |
+
"grad_norm": 0.37992908787234797,
|
777 |
+
"learning_rate": 0.00015059938862204127,
|
778 |
+
"loss": 0.8502,
|
779 |
+
"step": 110
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.3552,
|
783 |
+
"grad_norm": 0.3537079585046425,
|
784 |
+
"learning_rate": 0.00014969940209799248,
|
785 |
+
"loss": 0.837,
|
786 |
+
"step": 111
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.3584,
|
790 |
+
"grad_norm": 0.4947879619869325,
|
791 |
+
"learning_rate": 0.00014879403742151283,
|
792 |
+
"loss": 0.8655,
|
793 |
+
"step": 112
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.3616,
|
797 |
+
"grad_norm": 0.4144273342987794,
|
798 |
+
"learning_rate": 0.00014788339256539544,
|
799 |
+
"loss": 0.8024,
|
800 |
+
"step": 113
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.3648,
|
804 |
+
"grad_norm": 0.42810201253462243,
|
805 |
+
"learning_rate": 0.0001469675660738206,
|
806 |
+
"loss": 0.8646,
|
807 |
+
"step": 114
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.368,
|
811 |
+
"grad_norm": 0.43495581513946346,
|
812 |
+
"learning_rate": 0.00014604665705169237,
|
813 |
+
"loss": 0.8019,
|
814 |
+
"step": 115
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.3712,
|
818 |
+
"grad_norm": 0.40514388555481956,
|
819 |
+
"learning_rate": 0.00014512076515391375,
|
820 |
+
"loss": 0.8307,
|
821 |
+
"step": 116
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.3744,
|
825 |
+
"grad_norm": 0.5344793506502119,
|
826 |
+
"learning_rate": 0.00014418999057460276,
|
827 |
+
"loss": 0.8583,
|
828 |
+
"step": 117
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.3776,
|
832 |
+
"grad_norm": 0.3591362856868415,
|
833 |
+
"learning_rate": 0.0001432544340362501,
|
834 |
+
"loss": 0.8422,
|
835 |
+
"step": 118
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.3808,
|
839 |
+
"grad_norm": 0.43800340514585073,
|
840 |
+
"learning_rate": 0.00014231419677881966,
|
841 |
+
"loss": 0.829,
|
842 |
+
"step": 119
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.384,
|
846 |
+
"grad_norm": 0.3546231000833001,
|
847 |
+
"learning_rate": 0.00014136938054879283,
|
848 |
+
"loss": 0.8502,
|
849 |
+
"step": 120
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.3872,
|
853 |
+
"grad_norm": 0.3824038358260862,
|
854 |
+
"learning_rate": 0.00014042008758815818,
|
855 |
+
"loss": 0.8601,
|
856 |
+
"step": 121
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.3904,
|
860 |
+
"grad_norm": 0.3773025369439645,
|
861 |
+
"learning_rate": 0.00013946642062334766,
|
862 |
+
"loss": 0.7483,
|
863 |
+
"step": 122
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.3936,
|
867 |
+
"grad_norm": 0.40841295198664107,
|
868 |
+
"learning_rate": 0.00013850848285411994,
|
869 |
+
"loss": 0.8981,
|
870 |
+
"step": 123
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.3968,
|
874 |
+
"grad_norm": 0.32297152126914114,
|
875 |
+
"learning_rate": 0.000137546377942393,
|
876 |
+
"loss": 0.7604,
|
877 |
+
"step": 124
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.4,
|
881 |
+
"grad_norm": 0.38908595736780166,
|
882 |
+
"learning_rate": 0.00013658021000102636,
|
883 |
+
"loss": 0.8377,
|
884 |
+
"step": 125
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.4032,
|
888 |
+
"grad_norm": 0.30749072492921053,
|
889 |
+
"learning_rate": 0.00013561008358255468,
|
890 |
+
"loss": 0.6953,
|
891 |
+
"step": 126
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.4064,
|
895 |
+
"grad_norm": 0.42279066677539,
|
896 |
+
"learning_rate": 0.00013463610366787392,
|
897 |
+
"loss": 0.8381,
|
898 |
+
"step": 127
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.4096,
|
902 |
+
"grad_norm": 0.46835688525644387,
|
903 |
+
"learning_rate": 0.00013365837565488064,
|
904 |
+
"loss": 0.8801,
|
905 |
+
"step": 128
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.4128,
|
909 |
+
"grad_norm": 0.38170959986063163,
|
910 |
+
"learning_rate": 0.0001326770053470668,
|
911 |
+
"loss": 0.8177,
|
912 |
+
"step": 129
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.416,
|
916 |
+
"grad_norm": 0.3748992249270009,
|
917 |
+
"learning_rate": 0.0001316920989420703,
|
918 |
+
"loss": 0.7318,
|
919 |
+
"step": 130
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.4192,
|
923 |
+
"grad_norm": 0.32117252663093104,
|
924 |
+
"learning_rate": 0.00013070376302018287,
|
925 |
+
"loss": 0.7911,
|
926 |
+
"step": 131
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.4224,
|
930 |
+
"grad_norm": 0.32643252918835663,
|
931 |
+
"learning_rate": 0.00012971210453281674,
|
932 |
+
"loss": 0.7565,
|
933 |
+
"step": 132
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.4256,
|
937 |
+
"grad_norm": 0.34707082290937796,
|
938 |
+
"learning_rate": 0.000128717230790931,
|
939 |
+
"loss": 0.8179,
|
940 |
+
"step": 133
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.4288,
|
944 |
+
"grad_norm": 0.37747643787259155,
|
945 |
+
"learning_rate": 0.00012771924945341906,
|
946 |
+
"loss": 0.7647,
|
947 |
+
"step": 134
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.432,
|
951 |
+
"grad_norm": 0.4042549495002011,
|
952 |
+
"learning_rate": 0.00012671826851545851,
|
953 |
+
"loss": 0.8638,
|
954 |
+
"step": 135
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.4352,
|
958 |
+
"grad_norm": 0.40682592869078416,
|
959 |
+
"learning_rate": 0.0001257143962968246,
|
960 |
+
"loss": 0.9235,
|
961 |
+
"step": 136
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.4384,
|
965 |
+
"grad_norm": 0.3394181019470551,
|
966 |
+
"learning_rate": 0.00012470774143016853,
|
967 |
+
"loss": 0.8029,
|
968 |
+
"step": 137
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.4416,
|
972 |
+
"grad_norm": 0.37379683333378905,
|
973 |
+
"learning_rate": 0.00012369841284926188,
|
974 |
+
"loss": 0.8183,
|
975 |
+
"step": 138
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.4448,
|
979 |
+
"grad_norm": 0.37760782532653564,
|
980 |
+
"learning_rate": 0.00012268651977720866,
|
981 |
+
"loss": 0.732,
|
982 |
+
"step": 139
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.448,
|
986 |
+
"grad_norm": 0.42926915905667945,
|
987 |
+
"learning_rate": 0.00012167217171462566,
|
988 |
+
"loss": 0.8645,
|
989 |
+
"step": 140
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.4512,
|
993 |
+
"grad_norm": 0.3353578654728262,
|
994 |
+
"learning_rate": 0.0001206554784277931,
|
995 |
+
"loss": 0.7573,
|
996 |
+
"step": 141
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.4544,
|
1000 |
+
"grad_norm": 0.36702732140537525,
|
1001 |
+
"learning_rate": 0.00011963654993677645,
|
1002 |
+
"loss": 0.7855,
|
1003 |
+
"step": 142
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.4576,
|
1007 |
+
"grad_norm": 0.3945829288474564,
|
1008 |
+
"learning_rate": 0.00011861549650352069,
|
1009 |
+
"loss": 0.9015,
|
1010 |
+
"step": 143
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.4608,
|
1014 |
+
"grad_norm": 0.3384991496539021,
|
1015 |
+
"learning_rate": 0.00011759242861991855,
|
1016 |
+
"loss": 0.7672,
|
1017 |
+
"step": 144
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.464,
|
1021 |
+
"grad_norm": 0.36235446616831335,
|
1022 |
+
"learning_rate": 0.00011656745699585371,
|
1023 |
+
"loss": 0.7298,
|
1024 |
+
"step": 145
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.4672,
|
1028 |
+
"grad_norm": 0.37275162386027966,
|
1029 |
+
"learning_rate": 0.00011554069254722051,
|
1030 |
+
"loss": 0.8231,
|
1031 |
+
"step": 146
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.4704,
|
1035 |
+
"grad_norm": 0.3351601172488379,
|
1036 |
+
"learning_rate": 0.00011451224638392129,
|
1037 |
+
"loss": 0.78,
|
1038 |
+
"step": 147
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.4736,
|
1042 |
+
"grad_norm": 0.4177215304468739,
|
1043 |
+
"learning_rate": 0.00011348222979784289,
|
1044 |
+
"loss": 0.8148,
|
1045 |
+
"step": 148
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.4768,
|
1049 |
+
"grad_norm": 0.31882281813401364,
|
1050 |
+
"learning_rate": 0.00011245075425081328,
|
1051 |
+
"loss": 0.8256,
|
1052 |
+
"step": 149
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.48,
|
1056 |
+
"grad_norm": 0.34927524739208016,
|
1057 |
+
"learning_rate": 0.00011141793136253986,
|
1058 |
+
"loss": 0.7206,
|
1059 |
+
"step": 150
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.4832,
|
1063 |
+
"grad_norm": 0.3272143865687314,
|
1064 |
+
"learning_rate": 0.0001103838728985307,
|
1065 |
+
"loss": 0.8422,
|
1066 |
+
"step": 151
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.4864,
|
1070 |
+
"grad_norm": 0.38082561425167516,
|
1071 |
+
"learning_rate": 0.000109348690758,
|
1072 |
+
"loss": 0.801,
|
1073 |
+
"step": 152
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.4896,
|
1077 |
+
"grad_norm": 0.3683618334975772,
|
1078 |
+
"learning_rate": 0.00010831249696175918,
|
1079 |
+
"loss": 0.754,
|
1080 |
+
"step": 153
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.4928,
|
1084 |
+
"grad_norm": 0.3460529686559446,
|
1085 |
+
"learning_rate": 0.0001072754036400944,
|
1086 |
+
"loss": 0.7703,
|
1087 |
+
"step": 154
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.496,
|
1091 |
+
"grad_norm": 0.3906059155144969,
|
1092 |
+
"learning_rate": 0.00010623752302063283,
|
1093 |
+
"loss": 0.8635,
|
1094 |
+
"step": 155
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.4992,
|
1098 |
+
"grad_norm": 0.4498300529094466,
|
1099 |
+
"learning_rate": 0.00010519896741619803,
|
1100 |
+
"loss": 0.887,
|
1101 |
+
"step": 156
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.5024,
|
1105 |
+
"grad_norm": 0.34890284462378934,
|
1106 |
+
"learning_rate": 0.00010415984921265609,
|
1107 |
+
"loss": 0.8415,
|
1108 |
+
"step": 157
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.5056,
|
1112 |
+
"grad_norm": 0.35283761184137985,
|
1113 |
+
"learning_rate": 0.00010312028085675391,
|
1114 |
+
"loss": 0.7897,
|
1115 |
+
"step": 158
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.5088,
|
1119 |
+
"grad_norm": 0.3383893586354043,
|
1120 |
+
"learning_rate": 0.00010208037484395114,
|
1121 |
+
"loss": 0.7699,
|
1122 |
+
"step": 159
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.512,
|
1126 |
+
"grad_norm": 0.37944587951981884,
|
1127 |
+
"learning_rate": 0.00010104024370624644,
|
1128 |
+
"loss": 0.8168,
|
1129 |
+
"step": 160
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.5152,
|
1133 |
+
"grad_norm": 0.35361574612426283,
|
1134 |
+
"learning_rate": 0.0001,
|
1135 |
+
"loss": 0.8237,
|
1136 |
+
"step": 161
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.5184,
|
1140 |
+
"grad_norm": 0.49601519414630724,
|
1141 |
+
"learning_rate": 9.895975629375359e-05,
|
1142 |
+
"loss": 0.8944,
|
1143 |
+
"step": 162
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.5216,
|
1147 |
+
"grad_norm": 0.4329854234900553,
|
1148 |
+
"learning_rate": 9.791962515604887e-05,
|
1149 |
+
"loss": 0.8409,
|
1150 |
+
"step": 163
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.5248,
|
1154 |
+
"grad_norm": 0.33753405466010045,
|
1155 |
+
"learning_rate": 9.687971914324607e-05,
|
1156 |
+
"loss": 0.6823,
|
1157 |
+
"step": 164
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.528,
|
1161 |
+
"grad_norm": 0.3641968735452542,
|
1162 |
+
"learning_rate": 9.584015078734395e-05,
|
1163 |
+
"loss": 0.7999,
|
1164 |
+
"step": 165
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.5312,
|
1168 |
+
"grad_norm": 0.3412252144703986,
|
1169 |
+
"learning_rate": 9.480103258380198e-05,
|
1170 |
+
"loss": 0.7485,
|
1171 |
+
"step": 166
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.5344,
|
1175 |
+
"grad_norm": 0.36557961816073514,
|
1176 |
+
"learning_rate": 9.376247697936719e-05,
|
1177 |
+
"loss": 0.8047,
|
1178 |
+
"step": 167
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.5376,
|
1182 |
+
"grad_norm": 0.3143242351094908,
|
1183 |
+
"learning_rate": 9.272459635990562e-05,
|
1184 |
+
"loss": 0.7417,
|
1185 |
+
"step": 168
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.5408,
|
1189 |
+
"grad_norm": 0.37740381793108085,
|
1190 |
+
"learning_rate": 9.168750303824084e-05,
|
1191 |
+
"loss": 0.7847,
|
1192 |
+
"step": 169
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.544,
|
1196 |
+
"grad_norm": 0.3756001432572221,
|
1197 |
+
"learning_rate": 9.065130924199998e-05,
|
1198 |
+
"loss": 0.8045,
|
1199 |
+
"step": 170
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.5472,
|
1203 |
+
"grad_norm": 0.3886531783560117,
|
1204 |
+
"learning_rate": 8.961612710146934e-05,
|
1205 |
+
"loss": 0.7739,
|
1206 |
+
"step": 171
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.5504,
|
1210 |
+
"grad_norm": 0.35961818663851197,
|
1211 |
+
"learning_rate": 8.858206863746018e-05,
|
1212 |
+
"loss": 0.7682,
|
1213 |
+
"step": 172
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.5536,
|
1217 |
+
"grad_norm": 0.34291998146787495,
|
1218 |
+
"learning_rate": 8.754924574918675e-05,
|
1219 |
+
"loss": 0.8154,
|
1220 |
+
"step": 173
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.5568,
|
1224 |
+
"grad_norm": 0.365623213677616,
|
1225 |
+
"learning_rate": 8.651777020215712e-05,
|
1226 |
+
"loss": 0.7723,
|
1227 |
+
"step": 174
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.56,
|
1231 |
+
"grad_norm": 0.4154693198085254,
|
1232 |
+
"learning_rate": 8.548775361607872e-05,
|
1233 |
+
"loss": 0.8059,
|
1234 |
+
"step": 175
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.5632,
|
1238 |
+
"grad_norm": 0.41853675355977066,
|
1239 |
+
"learning_rate": 8.445930745277953e-05,
|
1240 |
+
"loss": 0.7936,
|
1241 |
+
"step": 176
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.5664,
|
1245 |
+
"grad_norm": 0.3029988536971098,
|
1246 |
+
"learning_rate": 8.343254300414628e-05,
|
1247 |
+
"loss": 0.7538,
|
1248 |
+
"step": 177
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.5696,
|
1252 |
+
"grad_norm": 0.35015284331125907,
|
1253 |
+
"learning_rate": 8.240757138008149e-05,
|
1254 |
+
"loss": 0.784,
|
1255 |
+
"step": 178
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.5728,
|
1259 |
+
"grad_norm": 0.36899119805196373,
|
1260 |
+
"learning_rate": 8.138450349647936e-05,
|
1261 |
+
"loss": 0.7712,
|
1262 |
+
"step": 179
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.576,
|
1266 |
+
"grad_norm": 0.4635909482238713,
|
1267 |
+
"learning_rate": 8.036345006322359e-05,
|
1268 |
+
"loss": 0.8151,
|
1269 |
+
"step": 180
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.5792,
|
1273 |
+
"grad_norm": 0.39324815381124656,
|
1274 |
+
"learning_rate": 7.934452157220694e-05,
|
1275 |
+
"loss": 0.8479,
|
1276 |
+
"step": 181
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.5824,
|
1280 |
+
"grad_norm": 0.3790446343866871,
|
1281 |
+
"learning_rate": 7.832782828537437e-05,
|
1282 |
+
"loss": 0.7849,
|
1283 |
+
"step": 182
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.5856,
|
1287 |
+
"grad_norm": 0.38585647860183997,
|
1288 |
+
"learning_rate": 7.731348022279134e-05,
|
1289 |
+
"loss": 0.8062,
|
1290 |
+
"step": 183
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.5888,
|
1294 |
+
"grad_norm": 0.3851371620272845,
|
1295 |
+
"learning_rate": 7.630158715073813e-05,
|
1296 |
+
"loss": 0.8341,
|
1297 |
+
"step": 184
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.592,
|
1301 |
+
"grad_norm": 0.33007054542330805,
|
1302 |
+
"learning_rate": 7.52922585698315e-05,
|
1303 |
+
"loss": 0.7197,
|
1304 |
+
"step": 185
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.5952,
|
1308 |
+
"grad_norm": 0.42174584227102946,
|
1309 |
+
"learning_rate": 7.428560370317542e-05,
|
1310 |
+
"loss": 0.8017,
|
1311 |
+
"step": 186
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.5984,
|
1315 |
+
"grad_norm": 0.33067818645649066,
|
1316 |
+
"learning_rate": 7.328173148454151e-05,
|
1317 |
+
"loss": 0.7886,
|
1318 |
+
"step": 187
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.6016,
|
1322 |
+
"grad_norm": 0.42265554001971417,
|
1323 |
+
"learning_rate": 7.228075054658096e-05,
|
1324 |
+
"loss": 0.8238,
|
1325 |
+
"step": 188
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.6048,
|
1329 |
+
"grad_norm": 0.34908037692220767,
|
1330 |
+
"learning_rate": 7.1282769209069e-05,
|
1331 |
+
"loss": 0.7928,
|
1332 |
+
"step": 189
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.608,
|
1336 |
+
"grad_norm": 0.3700495310823468,
|
1337 |
+
"learning_rate": 7.028789546718326e-05,
|
1338 |
+
"loss": 0.7491,
|
1339 |
+
"step": 190
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.6112,
|
1343 |
+
"grad_norm": 0.3535456489525407,
|
1344 |
+
"learning_rate": 6.929623697981718e-05,
|
1345 |
+
"loss": 0.8094,
|
1346 |
+
"step": 191
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.6144,
|
1350 |
+
"grad_norm": 0.35992816828147584,
|
1351 |
+
"learning_rate": 6.830790105792973e-05,
|
1352 |
+
"loss": 0.747,
|
1353 |
+
"step": 192
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.6176,
|
1357 |
+
"grad_norm": 0.34357552011988995,
|
1358 |
+
"learning_rate": 6.732299465293322e-05,
|
1359 |
+
"loss": 0.7329,
|
1360 |
+
"step": 193
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.6208,
|
1364 |
+
"grad_norm": 0.4017091190478587,
|
1365 |
+
"learning_rate": 6.63416243451194e-05,
|
1366 |
+
"loss": 0.807,
|
1367 |
+
"step": 194
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.624,
|
1371 |
+
"grad_norm": 0.3092472351482309,
|
1372 |
+
"learning_rate": 6.536389633212609e-05,
|
1373 |
+
"loss": 0.7309,
|
1374 |
+
"step": 195
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.6272,
|
1378 |
+
"grad_norm": 0.3737027071808558,
|
1379 |
+
"learning_rate": 6.43899164174453e-05,
|
1380 |
+
"loss": 0.8191,
|
1381 |
+
"step": 196
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.6304,
|
1385 |
+
"grad_norm": 0.36607238095868505,
|
1386 |
+
"learning_rate": 6.341978999897365e-05,
|
1387 |
+
"loss": 0.7998,
|
1388 |
+
"step": 197
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.6336,
|
1392 |
+
"grad_norm": 0.4042496034179054,
|
1393 |
+
"learning_rate": 6.245362205760704e-05,
|
1394 |
+
"loss": 0.773,
|
1395 |
+
"step": 198
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.6368,
|
1399 |
+
"grad_norm": 0.37764639494375285,
|
1400 |
+
"learning_rate": 6.149151714588009e-05,
|
1401 |
+
"loss": 0.8055,
|
1402 |
+
"step": 199
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.64,
|
1406 |
+
"grad_norm": 0.3348761239337774,
|
1407 |
+
"learning_rate": 6.053357937665237e-05,
|
1408 |
+
"loss": 0.7443,
|
1409 |
+
"step": 200
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.6432,
|
1413 |
+
"grad_norm": 0.37626992146366794,
|
1414 |
+
"learning_rate": 5.957991241184184e-05,
|
1415 |
+
"loss": 0.7826,
|
1416 |
+
"step": 201
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.6464,
|
1420 |
+
"grad_norm": 0.3665662199313722,
|
1421 |
+
"learning_rate": 5.863061945120719e-05,
|
1422 |
+
"loss": 0.8335,
|
1423 |
+
"step": 202
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.6496,
|
1427 |
+
"grad_norm": 0.33785931924439677,
|
1428 |
+
"learning_rate": 5.768580322118034e-05,
|
1429 |
+
"loss": 0.7972,
|
1430 |
+
"step": 203
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.6528,
|
1434 |
+
"grad_norm": 0.3880113242436853,
|
1435 |
+
"learning_rate": 5.6745565963749925e-05,
|
1436 |
+
"loss": 0.8896,
|
1437 |
+
"step": 204
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.656,
|
1441 |
+
"grad_norm": 0.3721493819848271,
|
1442 |
+
"learning_rate": 5.5810009425397294e-05,
|
1443 |
+
"loss": 0.8183,
|
1444 |
+
"step": 205
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 0.6592,
|
1448 |
+
"grad_norm": 0.5198646512996709,
|
1449 |
+
"learning_rate": 5.487923484608629e-05,
|
1450 |
+
"loss": 0.8566,
|
1451 |
+
"step": 206
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 0.6624,
|
1455 |
+
"grad_norm": 0.4065329950447661,
|
1456 |
+
"learning_rate": 5.395334294830765e-05,
|
1457 |
+
"loss": 0.9147,
|
1458 |
+
"step": 207
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 0.6656,
|
1462 |
+
"grad_norm": 0.37859745509497567,
|
1463 |
+
"learning_rate": 5.3032433926179395e-05,
|
1464 |
+
"loss": 0.8042,
|
1465 |
+
"step": 208
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 0.6688,
|
1469 |
+
"grad_norm": 0.3607765773764099,
|
1470 |
+
"learning_rate": 5.211660743460458e-05,
|
1471 |
+
"loss": 0.7938,
|
1472 |
+
"step": 209
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 0.672,
|
1476 |
+
"grad_norm": 0.4092950714526401,
|
1477 |
+
"learning_rate": 5.1205962578487155e-05,
|
1478 |
+
"loss": 0.806,
|
1479 |
+
"step": 210
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 0.6752,
|
1483 |
+
"grad_norm": 0.30432228901089703,
|
1484 |
+
"learning_rate": 5.030059790200756e-05,
|
1485 |
+
"loss": 0.7262,
|
1486 |
+
"step": 211
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 0.6784,
|
1490 |
+
"grad_norm": 0.37088903131596684,
|
1491 |
+
"learning_rate": 4.940061137795876e-05,
|
1492 |
+
"loss": 0.7169,
|
1493 |
+
"step": 212
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.6816,
|
1497 |
+
"grad_norm": 0.3904681317398879,
|
1498 |
+
"learning_rate": 4.850610039714444e-05,
|
1499 |
+
"loss": 0.8104,
|
1500 |
+
"step": 213
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 0.6848,
|
1504 |
+
"grad_norm": 0.46898322895625755,
|
1505 |
+
"learning_rate": 4.761716175783989e-05,
|
1506 |
+
"loss": 0.8688,
|
1507 |
+
"step": 214
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 0.688,
|
1511 |
+
"grad_norm": 0.3006038153490141,
|
1512 |
+
"learning_rate": 4.673389165531714e-05,
|
1513 |
+
"loss": 0.7169,
|
1514 |
+
"step": 215
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 0.6912,
|
1518 |
+
"grad_norm": 0.37810250308131216,
|
1519 |
+
"learning_rate": 4.585638567143529e-05,
|
1520 |
+
"loss": 0.789,
|
1521 |
+
"step": 216
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 0.6944,
|
1525 |
+
"grad_norm": 0.2989111481815409,
|
1526 |
+
"learning_rate": 4.498473876429726e-05,
|
1527 |
+
"loss": 0.7303,
|
1528 |
+
"step": 217
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 0.6976,
|
1532 |
+
"grad_norm": 0.32911143249763547,
|
1533 |
+
"learning_rate": 4.411904525797408e-05,
|
1534 |
+
"loss": 0.7592,
|
1535 |
+
"step": 218
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 0.7008,
|
1539 |
+
"grad_norm": 0.35549746840897417,
|
1540 |
+
"learning_rate": 4.325939883229766e-05,
|
1541 |
+
"loss": 0.8175,
|
1542 |
+
"step": 219
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 0.704,
|
1546 |
+
"grad_norm": 0.31577828848903333,
|
1547 |
+
"learning_rate": 4.240589251272342e-05,
|
1548 |
+
"loss": 0.7581,
|
1549 |
+
"step": 220
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 0.7072,
|
1553 |
+
"grad_norm": 0.38150936709065336,
|
1554 |
+
"learning_rate": 4.155861866026364e-05,
|
1555 |
+
"loss": 0.7383,
|
1556 |
+
"step": 221
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.7104,
|
1560 |
+
"grad_norm": 0.352873952674236,
|
1561 |
+
"learning_rate": 4.071766896149273e-05,
|
1562 |
+
"loss": 0.7899,
|
1563 |
+
"step": 222
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 0.7136,
|
1567 |
+
"grad_norm": 0.3729395002008322,
|
1568 |
+
"learning_rate": 3.988313441862553e-05,
|
1569 |
+
"loss": 0.8267,
|
1570 |
+
"step": 223
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 0.7168,
|
1574 |
+
"grad_norm": 0.38612103435747785,
|
1575 |
+
"learning_rate": 3.9055105339669595e-05,
|
1576 |
+
"loss": 0.8438,
|
1577 |
+
"step": 224
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 0.72,
|
1581 |
+
"grad_norm": 0.3343773300394658,
|
1582 |
+
"learning_rate": 3.823367132865265e-05,
|
1583 |
+
"loss": 0.7766,
|
1584 |
+
"step": 225
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 0.7232,
|
1588 |
+
"grad_norm": 0.34821484213233017,
|
1589 |
+
"learning_rate": 3.741892127592625e-05,
|
1590 |
+
"loss": 0.7869,
|
1591 |
+
"step": 226
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 0.7264,
|
1595 |
+
"grad_norm": 0.34452187805524936,
|
1596 |
+
"learning_rate": 3.6610943348546526e-05,
|
1597 |
+
"loss": 0.8142,
|
1598 |
+
"step": 227
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 0.7296,
|
1602 |
+
"grad_norm": 0.389690433062319,
|
1603 |
+
"learning_rate": 3.580982498073344e-05,
|
1604 |
+
"loss": 0.75,
|
1605 |
+
"step": 228
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 0.7328,
|
1609 |
+
"grad_norm": 0.4645823764551201,
|
1610 |
+
"learning_rate": 3.501565286440914e-05,
|
1611 |
+
"loss": 0.9086,
|
1612 |
+
"step": 229
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 0.736,
|
1616 |
+
"grad_norm": 0.3374398934171241,
|
1617 |
+
"learning_rate": 3.422851293981676e-05,
|
1618 |
+
"loss": 0.7671,
|
1619 |
+
"step": 230
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 0.7392,
|
1623 |
+
"grad_norm": 0.4111564099874274,
|
1624 |
+
"learning_rate": 3.3448490386220355e-05,
|
1625 |
+
"loss": 0.83,
|
1626 |
+
"step": 231
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 0.7424,
|
1630 |
+
"grad_norm": 0.4082259005237418,
|
1631 |
+
"learning_rate": 3.2675669612687565e-05,
|
1632 |
+
"loss": 0.8499,
|
1633 |
+
"step": 232
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 0.7456,
|
1637 |
+
"grad_norm": 0.35262858393716395,
|
1638 |
+
"learning_rate": 3.191013424895536e-05,
|
1639 |
+
"loss": 0.8093,
|
1640 |
+
"step": 233
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 0.7488,
|
1644 |
+
"grad_norm": 0.41355266235382226,
|
1645 |
+
"learning_rate": 3.115196713638e-05,
|
1646 |
+
"loss": 0.8087,
|
1647 |
+
"step": 234
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 0.752,
|
1651 |
+
"grad_norm": 0.41241448924179946,
|
1652 |
+
"learning_rate": 3.040125031897264e-05,
|
1653 |
+
"loss": 0.7911,
|
1654 |
+
"step": 235
|
1655 |
+
},
|
1656 |
+
{
|
1657 |
+
"epoch": 0.7552,
|
1658 |
+
"grad_norm": 0.3279954208792422,
|
1659 |
+
"learning_rate": 2.9658065034520978e-05,
|
1660 |
+
"loss": 0.7616,
|
1661 |
+
"step": 236
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 0.7584,
|
1665 |
+
"grad_norm": 0.3710790783951858,
|
1666 |
+
"learning_rate": 2.892249170579826e-05,
|
1667 |
+
"loss": 0.7417,
|
1668 |
+
"step": 237
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 0.7616,
|
1672 |
+
"grad_norm": 0.3638956295243486,
|
1673 |
+
"learning_rate": 2.8194609931860316e-05,
|
1674 |
+
"loss": 0.7734,
|
1675 |
+
"step": 238
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 0.7648,
|
1679 |
+
"grad_norm": 0.45340861210629174,
|
1680 |
+
"learning_rate": 2.7474498479432087e-05,
|
1681 |
+
"loss": 0.845,
|
1682 |
+
"step": 239
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.768,
|
1686 |
+
"grad_norm": 0.32432848532039027,
|
1687 |
+
"learning_rate": 2.6762235274383772e-05,
|
1688 |
+
"loss": 0.7292,
|
1689 |
+
"step": 240
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 0.7712,
|
1693 |
+
"grad_norm": 0.33285800890422707,
|
1694 |
+
"learning_rate": 2.6057897393298324e-05,
|
1695 |
+
"loss": 0.6951,
|
1696 |
+
"step": 241
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 0.7744,
|
1700 |
+
"grad_norm": 0.36168512959628546,
|
1701 |
+
"learning_rate": 2.536156105513062e-05,
|
1702 |
+
"loss": 0.7777,
|
1703 |
+
"step": 242
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 0.7776,
|
1707 |
+
"grad_norm": 0.3755110018820757,
|
1708 |
+
"learning_rate": 2.4673301612959654e-05,
|
1709 |
+
"loss": 0.7907,
|
1710 |
+
"step": 243
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 0.7808,
|
1714 |
+
"grad_norm": 0.4746969071550545,
|
1715 |
+
"learning_rate": 2.399319354583418e-05,
|
1716 |
+
"loss": 0.8731,
|
1717 |
+
"step": 244
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 0.784,
|
1721 |
+
"grad_norm": 0.3485421714780257,
|
1722 |
+
"learning_rate": 2.3321310450713062e-05,
|
1723 |
+
"loss": 0.7017,
|
1724 |
+
"step": 245
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 0.7872,
|
1728 |
+
"grad_norm": 0.3522895092881672,
|
1729 |
+
"learning_rate": 2.265772503450122e-05,
|
1730 |
+
"loss": 0.7341,
|
1731 |
+
"step": 246
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 0.7904,
|
1735 |
+
"grad_norm": 0.3171714951862727,
|
1736 |
+
"learning_rate": 2.2002509106181624e-05,
|
1737 |
+
"loss": 0.778,
|
1738 |
+
"step": 247
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 0.7936,
|
1742 |
+
"grad_norm": 0.30854770168466894,
|
1743 |
+
"learning_rate": 2.1355733569044635e-05,
|
1744 |
+
"loss": 0.7326,
|
1745 |
+
"step": 248
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 0.7968,
|
1749 |
+
"grad_norm": 0.4491439626115905,
|
1750 |
+
"learning_rate": 2.0717468413015283e-05,
|
1751 |
+
"loss": 0.8296,
|
1752 |
+
"step": 249
|
1753 |
+
},
|
1754 |
+
{
|
1755 |
+
"epoch": 0.8,
|
1756 |
+
"grad_norm": 0.37090037131233355,
|
1757 |
+
"learning_rate": 2.008778270707944e-05,
|
1758 |
+
"loss": 0.7953,
|
1759 |
+
"step": 250
|
1760 |
+
},
|
1761 |
+
{
|
1762 |
+
"epoch": 0.8032,
|
1763 |
+
"grad_norm": 0.3834261067678261,
|
1764 |
+
"learning_rate": 1.946674459180955e-05,
|
1765 |
+
"loss": 0.9055,
|
1766 |
+
"step": 251
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 0.8064,
|
1770 |
+
"grad_norm": 0.3255432432056942,
|
1771 |
+
"learning_rate": 1.8854421271990964e-05,
|
1772 |
+
"loss": 0.7592,
|
1773 |
+
"step": 252
|
1774 |
+
},
|
1775 |
+
{
|
1776 |
+
"epoch": 0.8096,
|
1777 |
+
"grad_norm": 0.38151180353083647,
|
1778 |
+
"learning_rate": 1.8250879009349398e-05,
|
1779 |
+
"loss": 0.7883,
|
1780 |
+
"step": 253
|
1781 |
+
},
|
1782 |
+
{
|
1783 |
+
"epoch": 0.8128,
|
1784 |
+
"grad_norm": 0.43340018294551996,
|
1785 |
+
"learning_rate": 1.7656183115380577e-05,
|
1786 |
+
"loss": 0.9072,
|
1787 |
+
"step": 254
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 0.816,
|
1791 |
+
"grad_norm": 0.4692009486980658,
|
1792 |
+
"learning_rate": 1.707039794428259e-05,
|
1793 |
+
"loss": 0.8892,
|
1794 |
+
"step": 255
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 0.8192,
|
1798 |
+
"grad_norm": 0.34906122041207605,
|
1799 |
+
"learning_rate": 1.649358688599191e-05,
|
1800 |
+
"loss": 0.7177,
|
1801 |
+
"step": 256
|
1802 |
+
},
|
1803 |
+
{
|
1804 |
+
"epoch": 0.8224,
|
1805 |
+
"grad_norm": 0.357075923035099,
|
1806 |
+
"learning_rate": 1.5925812359323745e-05,
|
1807 |
+
"loss": 0.7832,
|
1808 |
+
"step": 257
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 0.8256,
|
1812 |
+
"grad_norm": 0.3853853663460861,
|
1813 |
+
"learning_rate": 1.5367135805217458e-05,
|
1814 |
+
"loss": 0.7875,
|
1815 |
+
"step": 258
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 0.8288,
|
1819 |
+
"grad_norm": 0.3162997323586251,
|
1820 |
+
"learning_rate": 1.4817617680087825e-05,
|
1821 |
+
"loss": 0.7652,
|
1822 |
+
"step": 259
|
1823 |
+
},
|
1824 |
+
{
|
1825 |
+
"epoch": 0.832,
|
1826 |
+
"grad_norm": 0.5058370501510021,
|
1827 |
+
"learning_rate": 1.4277317449282834e-05,
|
1828 |
+
"loss": 0.788,
|
1829 |
+
"step": 260
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 0.8352,
|
1833 |
+
"grad_norm": 0.38934979473108083,
|
1834 |
+
"learning_rate": 1.3746293580648717e-05,
|
1835 |
+
"loss": 0.8643,
|
1836 |
+
"step": 261
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 0.8384,
|
1840 |
+
"grad_norm": 0.36714032393813667,
|
1841 |
+
"learning_rate": 1.3224603538202929e-05,
|
1842 |
+
"loss": 0.7469,
|
1843 |
+
"step": 262
|
1844 |
+
},
|
1845 |
+
{
|
1846 |
+
"epoch": 0.8416,
|
1847 |
+
"grad_norm": 0.3488248102724531,
|
1848 |
+
"learning_rate": 1.2712303775915802e-05,
|
1849 |
+
"loss": 0.7318,
|
1850 |
+
"step": 263
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 0.8448,
|
1854 |
+
"grad_norm": 0.294650221806305,
|
1855 |
+
"learning_rate": 1.220944973160133e-05,
|
1856 |
+
"loss": 0.7344,
|
1857 |
+
"step": 264
|
1858 |
+
},
|
1859 |
+
{
|
1860 |
+
"epoch": 0.848,
|
1861 |
+
"grad_norm": 0.3540730290394561,
|
1862 |
+
"learning_rate": 1.1716095820918216e-05,
|
1863 |
+
"loss": 0.7354,
|
1864 |
+
"step": 265
|
1865 |
+
},
|
1866 |
+
{
|
1867 |
+
"epoch": 0.8512,
|
1868 |
+
"grad_norm": 0.4226203562757057,
|
1869 |
+
"learning_rate": 1.1232295431481222e-05,
|
1870 |
+
"loss": 0.808,
|
1871 |
+
"step": 266
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 0.8544,
|
1875 |
+
"grad_norm": 0.4014616869239377,
|
1876 |
+
"learning_rate": 1.0758100917083991e-05,
|
1877 |
+
"loss": 0.7749,
|
1878 |
+
"step": 267
|
1879 |
+
},
|
1880 |
+
{
|
1881 |
+
"epoch": 0.8576,
|
1882 |
+
"grad_norm": 0.3934215620886321,
|
1883 |
+
"learning_rate": 1.0293563592033595e-05,
|
1884 |
+
"loss": 0.8003,
|
1885 |
+
"step": 268
|
1886 |
+
},
|
1887 |
+
{
|
1888 |
+
"epoch": 0.8608,
|
1889 |
+
"grad_norm": 0.3353659014526165,
|
1890 |
+
"learning_rate": 9.838733725597615e-06,
|
1891 |
+
"loss": 0.7851,
|
1892 |
+
"step": 269
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 0.864,
|
1896 |
+
"grad_norm": 0.3700720016155327,
|
1897 |
+
"learning_rate": 9.393660536564408e-06,
|
1898 |
+
"loss": 0.7941,
|
1899 |
+
"step": 270
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 0.8672,
|
1903 |
+
"grad_norm": 0.31100504311573046,
|
1904 |
+
"learning_rate": 8.958392187916841e-06,
|
1905 |
+
"loss": 0.7298,
|
1906 |
+
"step": 271
|
1907 |
+
},
|
1908 |
+
{
|
1909 |
+
"epoch": 0.8704,
|
1910 |
+
"grad_norm": 0.4172712953645561,
|
1911 |
+
"learning_rate": 8.532975781620512e-06,
|
1912 |
+
"loss": 0.7686,
|
1913 |
+
"step": 272
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 0.8736,
|
1917 |
+
"grad_norm": 0.34667441316708114,
|
1918 |
+
"learning_rate": 8.117457353526625e-06,
|
1919 |
+
"loss": 0.7106,
|
1920 |
+
"step": 273
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 0.8768,
|
1924 |
+
"grad_norm": 0.3004518824396042,
|
1925 |
+
"learning_rate": 7.711881868390291e-06,
|
1926 |
+
"loss": 0.6887,
|
1927 |
+
"step": 274
|
1928 |
+
},
|
1929 |
+
{
|
1930 |
+
"epoch": 0.88,
|
1931 |
+
"grad_norm": 0.38133602184118887,
|
1932 |
+
"learning_rate": 7.3162932150046885e-06,
|
1933 |
+
"loss": 0.8172,
|
1934 |
+
"step": 275
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 0.8832,
|
1938 |
+
"grad_norm": 0.33116090594086295,
|
1939 |
+
"learning_rate": 6.930734201451816e-06,
|
1940 |
+
"loss": 0.7005,
|
1941 |
+
"step": 276
|
1942 |
+
},
|
1943 |
+
{
|
1944 |
+
"epoch": 0.8864,
|
1945 |
+
"grad_norm": 0.3188298094332972,
|
1946 |
+
"learning_rate": 6.555246550469907e-06,
|
1947 |
+
"loss": 0.7534,
|
1948 |
+
"step": 277
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 0.8896,
|
1952 |
+
"grad_norm": 0.4403999231950908,
|
1953 |
+
"learning_rate": 6.189870894938587e-06,
|
1954 |
+
"loss": 0.8213,
|
1955 |
+
"step": 278
|
1956 |
+
},
|
1957 |
+
{
|
1958 |
+
"epoch": 0.8928,
|
1959 |
+
"grad_norm": 0.3608087635662593,
|
1960 |
+
"learning_rate": 5.834646773481811e-06,
|
1961 |
+
"loss": 0.7844,
|
1962 |
+
"step": 279
|
1963 |
+
},
|
1964 |
+
{
|
1965 |
+
"epoch": 0.896,
|
1966 |
+
"grad_norm": 0.40474388045501003,
|
1967 |
+
"learning_rate": 5.489612626189245e-06,
|
1968 |
+
"loss": 0.8244,
|
1969 |
+
"step": 280
|
1970 |
+
},
|
1971 |
+
{
|
1972 |
+
"epoch": 0.8992,
|
1973 |
+
"grad_norm": 0.3389637770265784,
|
1974 |
+
"learning_rate": 5.154805790456485e-06,
|
1975 |
+
"loss": 0.7018,
|
1976 |
+
"step": 281
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 0.9024,
|
1980 |
+
"grad_norm": 0.30643971186231883,
|
1981 |
+
"learning_rate": 4.830262496944693e-06,
|
1982 |
+
"loss": 0.7619,
|
1983 |
+
"step": 282
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 0.9056,
|
1987 |
+
"grad_norm": 0.3661184834317572,
|
1988 |
+
"learning_rate": 4.516017865659949e-06,
|
1989 |
+
"loss": 0.703,
|
1990 |
+
"step": 283
|
1991 |
+
},
|
1992 |
+
{
|
1993 |
+
"epoch": 0.9088,
|
1994 |
+
"grad_norm": 0.3671129154556767,
|
1995 |
+
"learning_rate": 4.21210590215273e-06,
|
1996 |
+
"loss": 0.7824,
|
1997 |
+
"step": 284
|
1998 |
+
},
|
1999 |
+
{
|
2000 |
+
"epoch": 0.912,
|
2001 |
+
"grad_norm": 0.34883381196841734,
|
2002 |
+
"learning_rate": 3.918559493838114e-06,
|
2003 |
+
"loss": 0.8148,
|
2004 |
+
"step": 285
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 0.9152,
|
2008 |
+
"grad_norm": 0.32439361737968153,
|
2009 |
+
"learning_rate": 3.6354104064368566e-06,
|
2010 |
+
"loss": 0.7575,
|
2011 |
+
"step": 286
|
2012 |
+
},
|
2013 |
+
{
|
2014 |
+
"epoch": 0.9184,
|
2015 |
+
"grad_norm": 0.3866839936796524,
|
2016 |
+
"learning_rate": 3.3626892805379562e-06,
|
2017 |
+
"loss": 0.8134,
|
2018 |
+
"step": 287
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 0.9216,
|
2022 |
+
"grad_norm": 0.3825547694903366,
|
2023 |
+
"learning_rate": 3.100425628282899e-06,
|
2024 |
+
"loss": 0.7618,
|
2025 |
+
"step": 288
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 0.9248,
|
2029 |
+
"grad_norm": 0.31827546790247363,
|
2030 |
+
"learning_rate": 2.848647830172024e-06,
|
2031 |
+
"loss": 0.7321,
|
2032 |
+
"step": 289
|
2033 |
+
},
|
2034 |
+
{
|
2035 |
+
"epoch": 0.928,
|
2036 |
+
"grad_norm": 0.29623028045498556,
|
2037 |
+
"learning_rate": 2.607383131993424e-06,
|
2038 |
+
"loss": 0.6688,
|
2039 |
+
"step": 290
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 0.9312,
|
2043 |
+
"grad_norm": 0.40429655628990896,
|
2044 |
+
"learning_rate": 2.3766576418745022e-06,
|
2045 |
+
"loss": 0.8106,
|
2046 |
+
"step": 291
|
2047 |
+
},
|
2048 |
+
{
|
2049 |
+
"epoch": 0.9344,
|
2050 |
+
"grad_norm": 0.3650008962489634,
|
2051 |
+
"learning_rate": 2.1564963274568027e-06,
|
2052 |
+
"loss": 0.7758,
|
2053 |
+
"step": 292
|
2054 |
+
},
|
2055 |
+
{
|
2056 |
+
"epoch": 0.9376,
|
2057 |
+
"grad_norm": 0.34654671660631003,
|
2058 |
+
"learning_rate": 1.9469230131940907e-06,
|
2059 |
+
"loss": 0.7488,
|
2060 |
+
"step": 293
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 0.9408,
|
2064 |
+
"grad_norm": 0.3417601789633415,
|
2065 |
+
"learning_rate": 1.7479603777742938e-06,
|
2066 |
+
"loss": 0.7376,
|
2067 |
+
"step": 294
|
2068 |
+
},
|
2069 |
+
{
|
2070 |
+
"epoch": 0.944,
|
2071 |
+
"grad_norm": 0.4766922892989375,
|
2072 |
+
"learning_rate": 1.559629951665298e-06,
|
2073 |
+
"loss": 0.8448,
|
2074 |
+
"step": 295
|
2075 |
+
},
|
2076 |
+
{
|
2077 |
+
"epoch": 0.9472,
|
2078 |
+
"grad_norm": 0.3764754677513324,
|
2079 |
+
"learning_rate": 1.3819521147851123e-06,
|
2080 |
+
"loss": 0.756,
|
2081 |
+
"step": 296
|
2082 |
+
},
|
2083 |
+
{
|
2084 |
+
"epoch": 0.9504,
|
2085 |
+
"grad_norm": 0.3744144624463219,
|
2086 |
+
"learning_rate": 1.2149460942964098e-06,
|
2087 |
+
"loss": 0.7769,
|
2088 |
+
"step": 297
|
2089 |
+
},
|
2090 |
+
{
|
2091 |
+
"epoch": 0.9536,
|
2092 |
+
"grad_norm": 0.41513981451373816,
|
2093 |
+
"learning_rate": 1.05862996252597e-06,
|
2094 |
+
"loss": 0.7734,
|
2095 |
+
"step": 298
|
2096 |
+
},
|
2097 |
+
{
|
2098 |
+
"epoch": 0.9568,
|
2099 |
+
"grad_norm": 0.36679556444218603,
|
2100 |
+
"learning_rate": 9.130206350089765e-07,
|
2101 |
+
"loss": 0.7664,
|
2102 |
+
"step": 299
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 0.96,
|
2106 |
+
"grad_norm": 0.3859580111548229,
|
2107 |
+
"learning_rate": 7.781338686584927e-07,
|
2108 |
+
"loss": 0.7639,
|
2109 |
+
"step": 300
|
2110 |
+
},
|
2111 |
+
{
|
2112 |
+
"epoch": 0.9632,
|
2113 |
+
"grad_norm": 0.34312029909834285,
|
2114 |
+
"learning_rate": 6.539842600603918e-07,
|
2115 |
+
"loss": 0.8412,
|
2116 |
+
"step": 301
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 0.9664,
|
2120 |
+
"grad_norm": 0.4680442538250724,
|
2121 |
+
"learning_rate": 5.405852438937764e-07,
|
2122 |
+
"loss": 0.8093,
|
2123 |
+
"step": 302
|
2124 |
+
},
|
2125 |
+
{
|
2126 |
+
"epoch": 0.9696,
|
2127 |
+
"grad_norm": 0.3160450323270329,
|
2128 |
+
"learning_rate": 4.3794909147720773e-07,
|
2129 |
+
"loss": 0.709,
|
2130 |
+
"step": 303
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 0.9728,
|
2134 |
+
"grad_norm": 0.45915080458361,
|
2135 |
+
"learning_rate": 3.4608690944071263e-07,
|
2136 |
+
"loss": 0.8064,
|
2137 |
+
"step": 304
|
2138 |
+
},
|
2139 |
+
{
|
2140 |
+
"epoch": 0.976,
|
2141 |
+
"grad_norm": 0.3737969883986542,
|
2142 |
+
"learning_rate": 2.6500863852395584e-07,
|
2143 |
+
"loss": 0.8277,
|
2144 |
+
"step": 305
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 0.9792,
|
2148 |
+
"grad_norm": 0.3789680381062684,
|
2149 |
+
"learning_rate": 1.947230525005006e-07,
|
2150 |
+
"loss": 0.8404,
|
2151 |
+
"step": 306
|
2152 |
+
},
|
2153 |
+
{
|
2154 |
+
"epoch": 0.9824,
|
2155 |
+
"grad_norm": 0.35233074082938165,
|
2156 |
+
"learning_rate": 1.3523775722834587e-07,
|
2157 |
+
"loss": 0.7446,
|
2158 |
+
"step": 307
|
2159 |
+
},
|
2160 |
+
{
|
2161 |
+
"epoch": 0.9856,
|
2162 |
+
"grad_norm": 0.3337846639484774,
|
2163 |
+
"learning_rate": 8.655918982689581e-08,
|
2164 |
+
"loss": 0.7811,
|
2165 |
+
"step": 308
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 0.9888,
|
2169 |
+
"grad_norm": 0.35005314677512966,
|
2170 |
+
"learning_rate": 4.8692617980350406e-08,
|
2171 |
+
"loss": 0.8336,
|
2172 |
+
"step": 309
|
2173 |
+
},
|
2174 |
+
{
|
2175 |
+
"epoch": 0.992,
|
2176 |
+
"grad_norm": 0.38107841486064986,
|
2177 |
+
"learning_rate": 2.164213936770576e-08,
|
2178 |
+
"loss": 0.768,
|
2179 |
+
"step": 310
|
2180 |
+
},
|
2181 |
+
{
|
2182 |
+
"epoch": 0.9952,
|
2183 |
+
"grad_norm": 0.37188047064663365,
|
2184 |
+
"learning_rate": 5.410681219286673e-09,
|
2185 |
+
"loss": 0.7323,
|
2186 |
+
"step": 311
|
2187 |
+
},
|
2188 |
+
{
|
2189 |
+
"epoch": 0.9984,
|
2190 |
+
"grad_norm": 0.3528962453383034,
|
2191 |
+
"learning_rate": 0.0,
|
2192 |
+
"loss": 0.7924,
|
2193 |
+
"step": 312
|
2194 |
+
},
|
2195 |
+
{
|
2196 |
+
"epoch": 0.9984,
|
2197 |
+
"step": 312,
|
2198 |
+
"total_flos": 736497759813632.0,
|
2199 |
+
"train_loss": 0.8298983348485751,
|
2200 |
+
"train_runtime": 9149.0946,
|
2201 |
+
"train_samples_per_second": 1.093,
|
2202 |
+
"train_steps_per_second": 0.034
|
2203 |
+
}
|
2204 |
+
],
|
2205 |
+
"logging_steps": 1.0,
|
2206 |
+
"max_steps": 312,
|
2207 |
+
"num_input_tokens_seen": 0,
|
2208 |
+
"num_train_epochs": 1,
|
2209 |
+
"save_steps": 500,
|
2210 |
+
"stateful_callbacks": {
|
2211 |
+
"TrainerControl": {
|
2212 |
+
"args": {
|
2213 |
+
"should_epoch_stop": false,
|
2214 |
+
"should_evaluate": false,
|
2215 |
+
"should_log": false,
|
2216 |
+
"should_save": false,
|
2217 |
+
"should_training_stop": false
|
2218 |
+
},
|
2219 |
+
"attributes": {}
|
2220 |
+
}
|
2221 |
+
},
|
2222 |
+
"total_flos": 736497759813632.0,
|
2223 |
+
"train_batch_size": 8,
|
2224 |
+
"trial_name": null,
|
2225 |
+
"trial_params": null
|
2226 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: ./weights/Bunny-v1_1-Llama-3-8B-V
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.11.1
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 256,
|
14 |
+
"lora_dropout": 0.1,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 128,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": null,
|
22 |
+
"target_modules": [
|
23 |
+
"q_proj",
|
24 |
+
"up_proj",
|
25 |
+
"k_proj",
|
26 |
+
"down_proj",
|
27 |
+
"o_proj",
|
28 |
+
"gate_proj",
|
29 |
+
"v_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:61659b7548812b91a870a85aa48869803be2f9ca23408f3b764320bcca5a24cc
|
3 |
+
size 671150064
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/config.json
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
3 |
+
"architectures": [
|
4 |
+
"BunnyLlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"auto_map": {
|
9 |
+
"AutoConfig": "configuration_bunny_llama.BunnyLlamaConfig",
|
10 |
+
"AutoModelForCausalLM": "modeling_bunny_llama.BunnyLlamaForCausalLM"
|
11 |
+
},
|
12 |
+
"bos_token_id": 128000,
|
13 |
+
"continuous_training": false,
|
14 |
+
"eos_token_id": 128001,
|
15 |
+
"freeze_mm_mlp_adapter": false,
|
16 |
+
"hidden_act": "silu",
|
17 |
+
"hidden_size": 4096,
|
18 |
+
"image_aspect_ratio": "pad",
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 14336,
|
21 |
+
"max_position_embeddings": 8192,
|
22 |
+
"mm_hidden_size": 3456,
|
23 |
+
"mm_projector_lr": null,
|
24 |
+
"mm_projector_type": "mlp2x_gelu",
|
25 |
+
"mm_vision_tower": "./weights/siglip-so400m-patch14-384",
|
26 |
+
"model_type": "bunny-llama",
|
27 |
+
"num_attention_heads": 32,
|
28 |
+
"num_hidden_layers": 32,
|
29 |
+
"num_key_value_heads": 8,
|
30 |
+
"pretraining_tp": 1,
|
31 |
+
"rms_norm_eps": 1e-05,
|
32 |
+
"rope_scaling": null,
|
33 |
+
"rope_theta": 500000.0,
|
34 |
+
"tie_word_embeddings": false,
|
35 |
+
"tokenizer_model_max_length": 2048,
|
36 |
+
"tokenizer_padding_side": "right",
|
37 |
+
"torch_dtype": "float16",
|
38 |
+
"transformers_version": "4.41.2",
|
39 |
+
"tune_mm_mlp_adapter": false,
|
40 |
+
"unfreeze_vision_tower": true,
|
41 |
+
"use_cache": true,
|
42 |
+
"use_mm_proj": true,
|
43 |
+
"use_s2": true,
|
44 |
+
"vocab_size": 128256
|
45 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/non_lora_trainables.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cf545e27d23f76953d9bdd19703edae1ee5f1f56973e863e6006b8a69c875de6
|
3 |
+
size 918507402
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_2_lora/trainer_state.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: ./weights/Bunny-v1_1-Llama-3-8B-V
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.11.1
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 256,
|
14 |
+
"lora_dropout": 0.1,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 128,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": null,
|
22 |
+
"target_modules": [
|
23 |
+
"up_proj",
|
24 |
+
"q_proj",
|
25 |
+
"o_proj",
|
26 |
+
"v_proj",
|
27 |
+
"down_proj",
|
28 |
+
"gate_proj",
|
29 |
+
"k_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:60e78440e8ad5c216e1bbb491649fb73d8e9423283d6831d816658f013ddea60
|
3 |
+
size 671150064
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/config.json
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
3 |
+
"architectures": [
|
4 |
+
"BunnyLlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"auto_map": {
|
9 |
+
"AutoConfig": "configuration_bunny_llama.BunnyLlamaConfig",
|
10 |
+
"AutoModelForCausalLM": "modeling_bunny_llama.BunnyLlamaForCausalLM"
|
11 |
+
},
|
12 |
+
"bos_token_id": 128000,
|
13 |
+
"continuous_training": false,
|
14 |
+
"eos_token_id": 128001,
|
15 |
+
"freeze_mm_mlp_adapter": false,
|
16 |
+
"hidden_act": "silu",
|
17 |
+
"hidden_size": 4096,
|
18 |
+
"image_aspect_ratio": "pad",
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 14336,
|
21 |
+
"max_position_embeddings": 8192,
|
22 |
+
"mm_hidden_size": 3456,
|
23 |
+
"mm_projector_lr": null,
|
24 |
+
"mm_projector_type": "mlp2x_gelu",
|
25 |
+
"mm_vision_tower": "./weights/siglip-so400m-patch14-384",
|
26 |
+
"model_type": "bunny-llama",
|
27 |
+
"num_attention_heads": 32,
|
28 |
+
"num_hidden_layers": 32,
|
29 |
+
"num_key_value_heads": 8,
|
30 |
+
"pretraining_tp": 1,
|
31 |
+
"rms_norm_eps": 1e-05,
|
32 |
+
"rope_scaling": null,
|
33 |
+
"rope_theta": 500000.0,
|
34 |
+
"tie_word_embeddings": false,
|
35 |
+
"tokenizer_model_max_length": 2048,
|
36 |
+
"tokenizer_padding_side": "right",
|
37 |
+
"torch_dtype": "float16",
|
38 |
+
"transformers_version": "4.41.2",
|
39 |
+
"tune_mm_mlp_adapter": false,
|
40 |
+
"unfreeze_vision_tower": true,
|
41 |
+
"use_cache": true,
|
42 |
+
"use_mm_proj": true,
|
43 |
+
"use_s2": true,
|
44 |
+
"vocab_size": 128256
|
45 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/non_lora_trainables.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7d89c8f41acde27e42e9a8a3b4e1df3f1d80d5199b474eab8c9904c815c8b6d0
|
3 |
+
size 918507402
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_10000_repeat_3_epochs_1_GA_4_lora/trainer_state.json
ADDED
@@ -0,0 +1,2226 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9984,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 312,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0032,
|
13 |
+
"grad_norm": 0.7954520844133007,
|
14 |
+
"learning_rate": 2e-05,
|
15 |
+
"loss": 1.3544,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.0064,
|
20 |
+
"grad_norm": 0.8486736733468345,
|
21 |
+
"learning_rate": 4e-05,
|
22 |
+
"loss": 1.4258,
|
23 |
+
"step": 2
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.0096,
|
27 |
+
"grad_norm": 0.7231129434225938,
|
28 |
+
"learning_rate": 6e-05,
|
29 |
+
"loss": 1.2565,
|
30 |
+
"step": 3
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.0128,
|
34 |
+
"grad_norm": 0.6549525568483539,
|
35 |
+
"learning_rate": 8e-05,
|
36 |
+
"loss": 1.301,
|
37 |
+
"step": 4
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.016,
|
41 |
+
"grad_norm": 0.6713937782391836,
|
42 |
+
"learning_rate": 0.0001,
|
43 |
+
"loss": 1.1783,
|
44 |
+
"step": 5
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.0192,
|
48 |
+
"grad_norm": 0.7775279811022765,
|
49 |
+
"learning_rate": 0.00012,
|
50 |
+
"loss": 1.2018,
|
51 |
+
"step": 6
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.0224,
|
55 |
+
"grad_norm": 0.8210500566974728,
|
56 |
+
"learning_rate": 0.00014,
|
57 |
+
"loss": 1.0322,
|
58 |
+
"step": 7
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.0256,
|
62 |
+
"grad_norm": 0.582072473996548,
|
63 |
+
"learning_rate": 0.00016,
|
64 |
+
"loss": 1.0719,
|
65 |
+
"step": 8
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.0288,
|
69 |
+
"grad_norm": 0.5332524469921555,
|
70 |
+
"learning_rate": 0.00018,
|
71 |
+
"loss": 0.9937,
|
72 |
+
"step": 9
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.032,
|
76 |
+
"grad_norm": 0.4652344399947701,
|
77 |
+
"learning_rate": 0.0002,
|
78 |
+
"loss": 0.9582,
|
79 |
+
"step": 10
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.0352,
|
83 |
+
"grad_norm": 0.528479459724629,
|
84 |
+
"learning_rate": 0.00019999458931878073,
|
85 |
+
"loss": 1.0025,
|
86 |
+
"step": 11
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.0384,
|
90 |
+
"grad_norm": 0.48862298453038183,
|
91 |
+
"learning_rate": 0.0001999783578606323,
|
92 |
+
"loss": 1.0045,
|
93 |
+
"step": 12
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.0416,
|
97 |
+
"grad_norm": 0.4788245246121141,
|
98 |
+
"learning_rate": 0.00019995130738201966,
|
99 |
+
"loss": 0.9681,
|
100 |
+
"step": 13
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.0448,
|
104 |
+
"grad_norm": 0.44036289873008344,
|
105 |
+
"learning_rate": 0.0001999134408101731,
|
106 |
+
"loss": 0.9215,
|
107 |
+
"step": 14
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.048,
|
111 |
+
"grad_norm": 0.5449869090192805,
|
112 |
+
"learning_rate": 0.00019986476224277165,
|
113 |
+
"loss": 0.9593,
|
114 |
+
"step": 15
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.0512,
|
118 |
+
"grad_norm": 0.4237521722332184,
|
119 |
+
"learning_rate": 0.00019980527694749952,
|
120 |
+
"loss": 0.8736,
|
121 |
+
"step": 16
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.0544,
|
125 |
+
"grad_norm": 0.4168713872622836,
|
126 |
+
"learning_rate": 0.00019973499136147606,
|
127 |
+
"loss": 0.8863,
|
128 |
+
"step": 17
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.0576,
|
132 |
+
"grad_norm": 0.45264982727147834,
|
133 |
+
"learning_rate": 0.0001996539130905593,
|
134 |
+
"loss": 0.8696,
|
135 |
+
"step": 18
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.0608,
|
139 |
+
"grad_norm": 0.3929576698684051,
|
140 |
+
"learning_rate": 0.0001995620509085228,
|
141 |
+
"loss": 0.916,
|
142 |
+
"step": 19
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.064,
|
146 |
+
"grad_norm": 0.4415540009570294,
|
147 |
+
"learning_rate": 0.00019945941475610623,
|
148 |
+
"loss": 0.9198,
|
149 |
+
"step": 20
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.0672,
|
153 |
+
"grad_norm": 0.3812265501303348,
|
154 |
+
"learning_rate": 0.0001993460157399396,
|
155 |
+
"loss": 0.8966,
|
156 |
+
"step": 21
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.0704,
|
160 |
+
"grad_norm": 0.4047740643644689,
|
161 |
+
"learning_rate": 0.0001992218661313415,
|
162 |
+
"loss": 0.9509,
|
163 |
+
"step": 22
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.0736,
|
167 |
+
"grad_norm": 0.394670597027155,
|
168 |
+
"learning_rate": 0.00019908697936499103,
|
169 |
+
"loss": 0.8582,
|
170 |
+
"step": 23
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.0768,
|
174 |
+
"grad_norm": 0.40883800080500177,
|
175 |
+
"learning_rate": 0.00019894137003747403,
|
176 |
+
"loss": 0.8873,
|
177 |
+
"step": 24
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.08,
|
181 |
+
"grad_norm": 0.4945237937729557,
|
182 |
+
"learning_rate": 0.00019878505390570362,
|
183 |
+
"loss": 0.9728,
|
184 |
+
"step": 25
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.0832,
|
188 |
+
"grad_norm": 0.4621016693300622,
|
189 |
+
"learning_rate": 0.00019861804788521493,
|
190 |
+
"loss": 0.9232,
|
191 |
+
"step": 26
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.0864,
|
195 |
+
"grad_norm": 0.6732021018661585,
|
196 |
+
"learning_rate": 0.00019844037004833473,
|
197 |
+
"loss": 0.8132,
|
198 |
+
"step": 27
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.0896,
|
202 |
+
"grad_norm": 0.4422468298010891,
|
203 |
+
"learning_rate": 0.00019825203962222572,
|
204 |
+
"loss": 0.9266,
|
205 |
+
"step": 28
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.0928,
|
209 |
+
"grad_norm": 0.40220173065598824,
|
210 |
+
"learning_rate": 0.0001980530769868059,
|
211 |
+
"loss": 0.8905,
|
212 |
+
"step": 29
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.096,
|
216 |
+
"grad_norm": 0.36918169793645084,
|
217 |
+
"learning_rate": 0.00019784350367254322,
|
218 |
+
"loss": 0.8733,
|
219 |
+
"step": 30
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.0992,
|
223 |
+
"grad_norm": 0.5597457386366598,
|
224 |
+
"learning_rate": 0.0001976233423581255,
|
225 |
+
"loss": 0.9705,
|
226 |
+
"step": 31
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.1024,
|
230 |
+
"grad_norm": 0.5595809993867294,
|
231 |
+
"learning_rate": 0.0001973926168680066,
|
232 |
+
"loss": 0.9519,
|
233 |
+
"step": 32
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.1056,
|
237 |
+
"grad_norm": 0.4605324530129531,
|
238 |
+
"learning_rate": 0.00019715135216982798,
|
239 |
+
"loss": 0.895,
|
240 |
+
"step": 33
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.1088,
|
244 |
+
"grad_norm": 0.42815914241954167,
|
245 |
+
"learning_rate": 0.0001968995743717171,
|
246 |
+
"loss": 0.9781,
|
247 |
+
"step": 34
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.112,
|
251 |
+
"grad_norm": 0.4124008245937204,
|
252 |
+
"learning_rate": 0.00019663731071946206,
|
253 |
+
"loss": 0.8654,
|
254 |
+
"step": 35
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.1152,
|
258 |
+
"grad_norm": 0.4388119798381529,
|
259 |
+
"learning_rate": 0.00019636458959356316,
|
260 |
+
"loss": 0.9486,
|
261 |
+
"step": 36
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.1184,
|
265 |
+
"grad_norm": 0.41876874470531383,
|
266 |
+
"learning_rate": 0.0001960814405061619,
|
267 |
+
"loss": 0.8528,
|
268 |
+
"step": 37
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.1216,
|
272 |
+
"grad_norm": 0.4063934253065074,
|
273 |
+
"learning_rate": 0.00019578789409784727,
|
274 |
+
"loss": 0.9519,
|
275 |
+
"step": 38
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.1248,
|
279 |
+
"grad_norm": 0.4281987478958129,
|
280 |
+
"learning_rate": 0.00019548398213434007,
|
281 |
+
"loss": 0.931,
|
282 |
+
"step": 39
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.128,
|
286 |
+
"grad_norm": 0.3995519502824548,
|
287 |
+
"learning_rate": 0.00019516973750305532,
|
288 |
+
"loss": 0.8706,
|
289 |
+
"step": 40
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.1312,
|
293 |
+
"grad_norm": 0.39938546863455104,
|
294 |
+
"learning_rate": 0.00019484519420954354,
|
295 |
+
"loss": 0.9081,
|
296 |
+
"step": 41
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.1344,
|
300 |
+
"grad_norm": 0.41849656692135495,
|
301 |
+
"learning_rate": 0.00019451038737381077,
|
302 |
+
"loss": 0.9316,
|
303 |
+
"step": 42
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.1376,
|
307 |
+
"grad_norm": 0.38704750788751674,
|
308 |
+
"learning_rate": 0.00019416535322651818,
|
309 |
+
"loss": 0.9122,
|
310 |
+
"step": 43
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.1408,
|
314 |
+
"grad_norm": 0.36713908717782284,
|
315 |
+
"learning_rate": 0.00019381012910506146,
|
316 |
+
"loss": 0.8173,
|
317 |
+
"step": 44
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.144,
|
321 |
+
"grad_norm": 0.386332629442072,
|
322 |
+
"learning_rate": 0.00019344475344953012,
|
323 |
+
"loss": 0.8054,
|
324 |
+
"step": 45
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.1472,
|
328 |
+
"grad_norm": 0.40175758635877834,
|
329 |
+
"learning_rate": 0.00019306926579854821,
|
330 |
+
"loss": 0.871,
|
331 |
+
"step": 46
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.1504,
|
335 |
+
"grad_norm": 0.3980262653132001,
|
336 |
+
"learning_rate": 0.00019268370678499533,
|
337 |
+
"loss": 0.864,
|
338 |
+
"step": 47
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.1536,
|
342 |
+
"grad_norm": 0.4340588852915137,
|
343 |
+
"learning_rate": 0.0001922881181316097,
|
344 |
+
"loss": 0.9131,
|
345 |
+
"step": 48
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.1568,
|
349 |
+
"grad_norm": 0.424408064468781,
|
350 |
+
"learning_rate": 0.00019188254264647337,
|
351 |
+
"loss": 0.8575,
|
352 |
+
"step": 49
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.16,
|
356 |
+
"grad_norm": 0.42662287831550505,
|
357 |
+
"learning_rate": 0.0001914670242183795,
|
358 |
+
"loss": 0.8401,
|
359 |
+
"step": 50
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.1632,
|
363 |
+
"grad_norm": 0.4018922149491412,
|
364 |
+
"learning_rate": 0.0001910416078120832,
|
365 |
+
"loss": 0.8477,
|
366 |
+
"step": 51
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.1664,
|
370 |
+
"grad_norm": 0.4547104194477866,
|
371 |
+
"learning_rate": 0.0001906063394634356,
|
372 |
+
"loss": 0.9216,
|
373 |
+
"step": 52
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.1696,
|
377 |
+
"grad_norm": 0.46878889945327834,
|
378 |
+
"learning_rate": 0.00019016126627440237,
|
379 |
+
"loss": 0.8866,
|
380 |
+
"step": 53
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.1728,
|
384 |
+
"grad_norm": 0.3901101154924772,
|
385 |
+
"learning_rate": 0.00018970643640796642,
|
386 |
+
"loss": 0.8071,
|
387 |
+
"step": 54
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.176,
|
391 |
+
"grad_norm": 0.3786177626239962,
|
392 |
+
"learning_rate": 0.000189241899082916,
|
393 |
+
"loss": 0.8155,
|
394 |
+
"step": 55
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.1792,
|
398 |
+
"grad_norm": 0.37399747556550167,
|
399 |
+
"learning_rate": 0.00018876770456851877,
|
400 |
+
"loss": 0.8673,
|
401 |
+
"step": 56
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.1824,
|
405 |
+
"grad_norm": 0.4476713311081534,
|
406 |
+
"learning_rate": 0.0001882839041790818,
|
407 |
+
"loss": 0.8913,
|
408 |
+
"step": 57
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.1856,
|
412 |
+
"grad_norm": 0.3695813169482114,
|
413 |
+
"learning_rate": 0.00018779055026839868,
|
414 |
+
"loss": 0.8532,
|
415 |
+
"step": 58
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.1888,
|
419 |
+
"grad_norm": 0.3740852677452882,
|
420 |
+
"learning_rate": 0.00018728769622408423,
|
421 |
+
"loss": 0.851,
|
422 |
+
"step": 59
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.192,
|
426 |
+
"grad_norm": 0.37805895790025723,
|
427 |
+
"learning_rate": 0.00018677539646179707,
|
428 |
+
"loss": 0.8893,
|
429 |
+
"step": 60
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.1952,
|
433 |
+
"grad_norm": 0.48550367409643636,
|
434 |
+
"learning_rate": 0.00018625370641935129,
|
435 |
+
"loss": 0.9271,
|
436 |
+
"step": 61
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.1984,
|
440 |
+
"grad_norm": 0.4261985087419851,
|
441 |
+
"learning_rate": 0.00018572268255071718,
|
442 |
+
"loss": 0.9485,
|
443 |
+
"step": 62
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.2016,
|
447 |
+
"grad_norm": 0.3506821138816665,
|
448 |
+
"learning_rate": 0.00018518238231991218,
|
449 |
+
"loss": 0.7993,
|
450 |
+
"step": 63
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.2048,
|
454 |
+
"grad_norm": 0.39926954240428475,
|
455 |
+
"learning_rate": 0.00018463286419478255,
|
456 |
+
"loss": 0.9094,
|
457 |
+
"step": 64
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.208,
|
461 |
+
"grad_norm": 0.369474669842929,
|
462 |
+
"learning_rate": 0.00018407418764067627,
|
463 |
+
"loss": 0.8639,
|
464 |
+
"step": 65
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.2112,
|
468 |
+
"grad_norm": 0.38412612507425553,
|
469 |
+
"learning_rate": 0.00018350641311400812,
|
470 |
+
"loss": 0.92,
|
471 |
+
"step": 66
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.2144,
|
475 |
+
"grad_norm": 0.6236483911477445,
|
476 |
+
"learning_rate": 0.0001829296020557174,
|
477 |
+
"loss": 0.9896,
|
478 |
+
"step": 67
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.2176,
|
482 |
+
"grad_norm": 0.40709892702353573,
|
483 |
+
"learning_rate": 0.00018234381688461942,
|
484 |
+
"loss": 0.7737,
|
485 |
+
"step": 68
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.2208,
|
489 |
+
"grad_norm": 0.38384691117683545,
|
490 |
+
"learning_rate": 0.0001817491209906506,
|
491 |
+
"loss": 0.8189,
|
492 |
+
"step": 69
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.224,
|
496 |
+
"grad_norm": 0.419002960060546,
|
497 |
+
"learning_rate": 0.00018114557872800905,
|
498 |
+
"loss": 0.8892,
|
499 |
+
"step": 70
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.2272,
|
503 |
+
"grad_norm": 0.4317304658707876,
|
504 |
+
"learning_rate": 0.00018053325540819045,
|
505 |
+
"loss": 0.9425,
|
506 |
+
"step": 71
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.2304,
|
510 |
+
"grad_norm": 0.3605359976609917,
|
511 |
+
"learning_rate": 0.0001799122172929206,
|
512 |
+
"loss": 0.8432,
|
513 |
+
"step": 72
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.2336,
|
517 |
+
"grad_norm": 0.38387195033721394,
|
518 |
+
"learning_rate": 0.00017928253158698473,
|
519 |
+
"loss": 0.8127,
|
520 |
+
"step": 73
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.2368,
|
524 |
+
"grad_norm": 0.4115773141922566,
|
525 |
+
"learning_rate": 0.0001786442664309554,
|
526 |
+
"loss": 0.835,
|
527 |
+
"step": 74
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.24,
|
531 |
+
"grad_norm": 0.41682944760695084,
|
532 |
+
"learning_rate": 0.0001779974908938184,
|
533 |
+
"loss": 0.8527,
|
534 |
+
"step": 75
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.2432,
|
538 |
+
"grad_norm": 0.44021561768781686,
|
539 |
+
"learning_rate": 0.0001773422749654988,
|
540 |
+
"loss": 0.8634,
|
541 |
+
"step": 76
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.2464,
|
545 |
+
"grad_norm": 0.3788641404846892,
|
546 |
+
"learning_rate": 0.00017667868954928694,
|
547 |
+
"loss": 0.8657,
|
548 |
+
"step": 77
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.2496,
|
552 |
+
"grad_norm": 0.35062726715426834,
|
553 |
+
"learning_rate": 0.00017600680645416583,
|
554 |
+
"loss": 0.776,
|
555 |
+
"step": 78
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.2528,
|
559 |
+
"grad_norm": 0.3640559161150146,
|
560 |
+
"learning_rate": 0.00017532669838704035,
|
561 |
+
"loss": 0.8121,
|
562 |
+
"step": 79
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.256,
|
566 |
+
"grad_norm": 0.35673860083372294,
|
567 |
+
"learning_rate": 0.00017463843894486937,
|
568 |
+
"loss": 0.8013,
|
569 |
+
"step": 80
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.2592,
|
573 |
+
"grad_norm": 0.36719254075722085,
|
574 |
+
"learning_rate": 0.0001739421026067017,
|
575 |
+
"loss": 0.864,
|
576 |
+
"step": 81
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.2624,
|
580 |
+
"grad_norm": 0.39931819574163646,
|
581 |
+
"learning_rate": 0.00017323776472561627,
|
582 |
+
"loss": 0.8932,
|
583 |
+
"step": 82
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.2656,
|
587 |
+
"grad_norm": 0.3982022670626305,
|
588 |
+
"learning_rate": 0.00017252550152056795,
|
589 |
+
"loss": 0.8254,
|
590 |
+
"step": 83
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.2688,
|
594 |
+
"grad_norm": 0.3997476767736355,
|
595 |
+
"learning_rate": 0.0001718053900681397,
|
596 |
+
"loss": 0.8817,
|
597 |
+
"step": 84
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.272,
|
601 |
+
"grad_norm": 0.3463507690011195,
|
602 |
+
"learning_rate": 0.00017107750829420176,
|
603 |
+
"loss": 0.8348,
|
604 |
+
"step": 85
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.2752,
|
608 |
+
"grad_norm": 0.4200196272874818,
|
609 |
+
"learning_rate": 0.00017034193496547902,
|
610 |
+
"loss": 0.9056,
|
611 |
+
"step": 86
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.2784,
|
615 |
+
"grad_norm": 0.3755782720326349,
|
616 |
+
"learning_rate": 0.00016959874968102735,
|
617 |
+
"loss": 0.8419,
|
618 |
+
"step": 87
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.2816,
|
622 |
+
"grad_norm": 0.3981270184947585,
|
623 |
+
"learning_rate": 0.00016884803286362,
|
624 |
+
"loss": 0.8491,
|
625 |
+
"step": 88
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.2848,
|
629 |
+
"grad_norm": 0.37218427069658755,
|
630 |
+
"learning_rate": 0.00016808986575104465,
|
631 |
+
"loss": 0.9352,
|
632 |
+
"step": 89
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.288,
|
636 |
+
"grad_norm": 0.3395579879759133,
|
637 |
+
"learning_rate": 0.00016732433038731242,
|
638 |
+
"loss": 0.7546,
|
639 |
+
"step": 90
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.2912,
|
643 |
+
"grad_norm": 0.3673793454918801,
|
644 |
+
"learning_rate": 0.0001665515096137797,
|
645 |
+
"loss": 0.833,
|
646 |
+
"step": 91
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.2944,
|
650 |
+
"grad_norm": 0.34780686537520417,
|
651 |
+
"learning_rate": 0.00016577148706018328,
|
652 |
+
"loss": 0.8021,
|
653 |
+
"step": 92
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.2976,
|
657 |
+
"grad_norm": 0.38425962561813065,
|
658 |
+
"learning_rate": 0.00016498434713559088,
|
659 |
+
"loss": 0.8679,
|
660 |
+
"step": 93
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.3008,
|
664 |
+
"grad_norm": 0.42936546302655365,
|
665 |
+
"learning_rate": 0.00016419017501926656,
|
666 |
+
"loss": 0.8377,
|
667 |
+
"step": 94
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.304,
|
671 |
+
"grad_norm": 0.4267179181570991,
|
672 |
+
"learning_rate": 0.0001633890566514535,
|
673 |
+
"loss": 0.8357,
|
674 |
+
"step": 95
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.3072,
|
678 |
+
"grad_norm": 0.36916905077663476,
|
679 |
+
"learning_rate": 0.00016258107872407375,
|
680 |
+
"loss": 0.8232,
|
681 |
+
"step": 96
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.3104,
|
685 |
+
"grad_norm": 0.38729223078060027,
|
686 |
+
"learning_rate": 0.0001617663286713474,
|
687 |
+
"loss": 0.8608,
|
688 |
+
"step": 97
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.3136,
|
692 |
+
"grad_norm": 0.4065510182299363,
|
693 |
+
"learning_rate": 0.00016094489466033043,
|
694 |
+
"loss": 0.8215,
|
695 |
+
"step": 98
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.3168,
|
699 |
+
"grad_norm": 0.35357551053351133,
|
700 |
+
"learning_rate": 0.00016011686558137448,
|
701 |
+
"loss": 0.8115,
|
702 |
+
"step": 99
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.32,
|
706 |
+
"grad_norm": 1.0001838046966431,
|
707 |
+
"learning_rate": 0.0001592823310385073,
|
708 |
+
"loss": 0.8702,
|
709 |
+
"step": 100
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.3232,
|
713 |
+
"grad_norm": 0.40193375582947716,
|
714 |
+
"learning_rate": 0.0001584413813397364,
|
715 |
+
"loss": 0.8128,
|
716 |
+
"step": 101
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.3264,
|
720 |
+
"grad_norm": 0.3935235182421146,
|
721 |
+
"learning_rate": 0.00015759410748727662,
|
722 |
+
"loss": 0.7955,
|
723 |
+
"step": 102
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.3296,
|
727 |
+
"grad_norm": 0.6013285238613454,
|
728 |
+
"learning_rate": 0.00015674060116770236,
|
729 |
+
"loss": 0.8225,
|
730 |
+
"step": 103
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.3328,
|
734 |
+
"grad_norm": 0.3787137979507313,
|
735 |
+
"learning_rate": 0.00015588095474202595,
|
736 |
+
"loss": 0.8257,
|
737 |
+
"step": 104
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.336,
|
741 |
+
"grad_norm": 0.38425572159674526,
|
742 |
+
"learning_rate": 0.00015501526123570277,
|
743 |
+
"loss": 0.8475,
|
744 |
+
"step": 105
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.3392,
|
748 |
+
"grad_norm": 0.4006328412662351,
|
749 |
+
"learning_rate": 0.00015414361432856475,
|
750 |
+
"loss": 0.8648,
|
751 |
+
"step": 106
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.3424,
|
755 |
+
"grad_norm": 0.4086678609977253,
|
756 |
+
"learning_rate": 0.0001532661083446829,
|
757 |
+
"loss": 0.8863,
|
758 |
+
"step": 107
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.3456,
|
762 |
+
"grad_norm": 0.4881485864286447,
|
763 |
+
"learning_rate": 0.00015238283824216015,
|
764 |
+
"loss": 0.9356,
|
765 |
+
"step": 108
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.3488,
|
769 |
+
"grad_norm": 0.4025668507825546,
|
770 |
+
"learning_rate": 0.00015149389960285558,
|
771 |
+
"loss": 0.8127,
|
772 |
+
"step": 109
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.352,
|
776 |
+
"grad_norm": 0.3710129691466552,
|
777 |
+
"learning_rate": 0.00015059938862204127,
|
778 |
+
"loss": 0.8428,
|
779 |
+
"step": 110
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.3552,
|
783 |
+
"grad_norm": 0.4003824576144442,
|
784 |
+
"learning_rate": 0.00014969940209799248,
|
785 |
+
"loss": 0.8425,
|
786 |
+
"step": 111
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.3584,
|
790 |
+
"grad_norm": 0.5018583001311245,
|
791 |
+
"learning_rate": 0.00014879403742151283,
|
792 |
+
"loss": 0.8779,
|
793 |
+
"step": 112
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.3616,
|
797 |
+
"grad_norm": 0.36121263099811374,
|
798 |
+
"learning_rate": 0.00014788339256539544,
|
799 |
+
"loss": 0.7713,
|
800 |
+
"step": 113
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.3648,
|
804 |
+
"grad_norm": 0.4251687064188239,
|
805 |
+
"learning_rate": 0.0001469675660738206,
|
806 |
+
"loss": 0.7802,
|
807 |
+
"step": 114
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.368,
|
811 |
+
"grad_norm": 0.3569667148399426,
|
812 |
+
"learning_rate": 0.00014604665705169237,
|
813 |
+
"loss": 0.8449,
|
814 |
+
"step": 115
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.3712,
|
818 |
+
"grad_norm": 0.4066427039698594,
|
819 |
+
"learning_rate": 0.00014512076515391375,
|
820 |
+
"loss": 0.8426,
|
821 |
+
"step": 116
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.3744,
|
825 |
+
"grad_norm": 0.44110171219803324,
|
826 |
+
"learning_rate": 0.00014418999057460276,
|
827 |
+
"loss": 0.8606,
|
828 |
+
"step": 117
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.3776,
|
832 |
+
"grad_norm": 0.37113662406191406,
|
833 |
+
"learning_rate": 0.0001432544340362501,
|
834 |
+
"loss": 0.8036,
|
835 |
+
"step": 118
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.3808,
|
839 |
+
"grad_norm": 0.36927548404036525,
|
840 |
+
"learning_rate": 0.00014231419677881966,
|
841 |
+
"loss": 0.7892,
|
842 |
+
"step": 119
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.384,
|
846 |
+
"grad_norm": 0.3628801931909292,
|
847 |
+
"learning_rate": 0.00014136938054879283,
|
848 |
+
"loss": 0.8945,
|
849 |
+
"step": 120
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.3872,
|
853 |
+
"grad_norm": 0.3881852305317355,
|
854 |
+
"learning_rate": 0.00014042008758815818,
|
855 |
+
"loss": 0.8457,
|
856 |
+
"step": 121
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.3904,
|
860 |
+
"grad_norm": 0.3493063856843603,
|
861 |
+
"learning_rate": 0.00013946642062334766,
|
862 |
+
"loss": 0.8096,
|
863 |
+
"step": 122
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.3936,
|
867 |
+
"grad_norm": 0.4238637282608415,
|
868 |
+
"learning_rate": 0.00013850848285411994,
|
869 |
+
"loss": 0.8239,
|
870 |
+
"step": 123
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.3968,
|
874 |
+
"grad_norm": 0.35268751443414337,
|
875 |
+
"learning_rate": 0.000137546377942393,
|
876 |
+
"loss": 0.8348,
|
877 |
+
"step": 124
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.4,
|
881 |
+
"grad_norm": 0.39291285524686237,
|
882 |
+
"learning_rate": 0.00013658021000102636,
|
883 |
+
"loss": 0.8241,
|
884 |
+
"step": 125
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.4032,
|
888 |
+
"grad_norm": 0.34506213306561334,
|
889 |
+
"learning_rate": 0.00013561008358255468,
|
890 |
+
"loss": 0.7677,
|
891 |
+
"step": 126
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.4064,
|
895 |
+
"grad_norm": 0.4203851880192952,
|
896 |
+
"learning_rate": 0.00013463610366787392,
|
897 |
+
"loss": 0.8461,
|
898 |
+
"step": 127
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.4096,
|
902 |
+
"grad_norm": 0.4474038709406084,
|
903 |
+
"learning_rate": 0.00013365837565488064,
|
904 |
+
"loss": 0.8768,
|
905 |
+
"step": 128
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.4128,
|
909 |
+
"grad_norm": 0.3778914426775528,
|
910 |
+
"learning_rate": 0.0001326770053470668,
|
911 |
+
"loss": 0.8762,
|
912 |
+
"step": 129
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.416,
|
916 |
+
"grad_norm": 0.3583604676991779,
|
917 |
+
"learning_rate": 0.0001316920989420703,
|
918 |
+
"loss": 0.749,
|
919 |
+
"step": 130
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.4192,
|
923 |
+
"grad_norm": 0.3201977553502979,
|
924 |
+
"learning_rate": 0.00013070376302018287,
|
925 |
+
"loss": 0.771,
|
926 |
+
"step": 131
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.4224,
|
930 |
+
"grad_norm": 0.327874635613636,
|
931 |
+
"learning_rate": 0.00012971210453281674,
|
932 |
+
"loss": 0.7247,
|
933 |
+
"step": 132
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.4256,
|
937 |
+
"grad_norm": 0.36654127787678314,
|
938 |
+
"learning_rate": 0.000128717230790931,
|
939 |
+
"loss": 0.8057,
|
940 |
+
"step": 133
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.4288,
|
944 |
+
"grad_norm": 0.37665903879585244,
|
945 |
+
"learning_rate": 0.00012771924945341906,
|
946 |
+
"loss": 0.7767,
|
947 |
+
"step": 134
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.432,
|
951 |
+
"grad_norm": 0.4034764628456416,
|
952 |
+
"learning_rate": 0.00012671826851545851,
|
953 |
+
"loss": 0.8348,
|
954 |
+
"step": 135
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.4352,
|
958 |
+
"grad_norm": 0.3907539802126158,
|
959 |
+
"learning_rate": 0.0001257143962968246,
|
960 |
+
"loss": 0.8872,
|
961 |
+
"step": 136
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.4384,
|
965 |
+
"grad_norm": 0.3569846944905879,
|
966 |
+
"learning_rate": 0.00012470774143016853,
|
967 |
+
"loss": 0.8112,
|
968 |
+
"step": 137
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.4416,
|
972 |
+
"grad_norm": 0.36188126942655424,
|
973 |
+
"learning_rate": 0.00012369841284926188,
|
974 |
+
"loss": 0.8019,
|
975 |
+
"step": 138
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.4448,
|
979 |
+
"grad_norm": 0.379019802148045,
|
980 |
+
"learning_rate": 0.00012268651977720866,
|
981 |
+
"loss": 0.7994,
|
982 |
+
"step": 139
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.448,
|
986 |
+
"grad_norm": 0.41234904215035534,
|
987 |
+
"learning_rate": 0.00012167217171462566,
|
988 |
+
"loss": 0.8809,
|
989 |
+
"step": 140
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.4512,
|
993 |
+
"grad_norm": 0.38868434007296393,
|
994 |
+
"learning_rate": 0.0001206554784277931,
|
995 |
+
"loss": 0.7578,
|
996 |
+
"step": 141
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.4544,
|
1000 |
+
"grad_norm": 0.350346834606368,
|
1001 |
+
"learning_rate": 0.00011963654993677645,
|
1002 |
+
"loss": 0.8003,
|
1003 |
+
"step": 142
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.4576,
|
1007 |
+
"grad_norm": 0.4135182041369236,
|
1008 |
+
"learning_rate": 0.00011861549650352069,
|
1009 |
+
"loss": 0.7891,
|
1010 |
+
"step": 143
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.4608,
|
1014 |
+
"grad_norm": 0.3447428610535347,
|
1015 |
+
"learning_rate": 0.00011759242861991855,
|
1016 |
+
"loss": 0.7747,
|
1017 |
+
"step": 144
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.464,
|
1021 |
+
"grad_norm": 0.3659043229276162,
|
1022 |
+
"learning_rate": 0.00011656745699585371,
|
1023 |
+
"loss": 0.7847,
|
1024 |
+
"step": 145
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.4672,
|
1028 |
+
"grad_norm": 0.3627598928666497,
|
1029 |
+
"learning_rate": 0.00011554069254722051,
|
1030 |
+
"loss": 0.8729,
|
1031 |
+
"step": 146
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.4704,
|
1035 |
+
"grad_norm": 0.3316910349747776,
|
1036 |
+
"learning_rate": 0.00011451224638392129,
|
1037 |
+
"loss": 0.7812,
|
1038 |
+
"step": 147
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.4736,
|
1042 |
+
"grad_norm": 0.3946570922891007,
|
1043 |
+
"learning_rate": 0.00011348222979784289,
|
1044 |
+
"loss": 0.7865,
|
1045 |
+
"step": 148
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.4768,
|
1049 |
+
"grad_norm": 0.3359636799382807,
|
1050 |
+
"learning_rate": 0.00011245075425081328,
|
1051 |
+
"loss": 0.7965,
|
1052 |
+
"step": 149
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.48,
|
1056 |
+
"grad_norm": 0.37137439376427284,
|
1057 |
+
"learning_rate": 0.00011141793136253986,
|
1058 |
+
"loss": 0.849,
|
1059 |
+
"step": 150
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.4832,
|
1063 |
+
"grad_norm": 0.3574031642165924,
|
1064 |
+
"learning_rate": 0.0001103838728985307,
|
1065 |
+
"loss": 0.825,
|
1066 |
+
"step": 151
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.4864,
|
1070 |
+
"grad_norm": 0.38556543326751924,
|
1071 |
+
"learning_rate": 0.000109348690758,
|
1072 |
+
"loss": 0.8379,
|
1073 |
+
"step": 152
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.4896,
|
1077 |
+
"grad_norm": 0.38805213476282924,
|
1078 |
+
"learning_rate": 0.00010831249696175918,
|
1079 |
+
"loss": 0.8097,
|
1080 |
+
"step": 153
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.4928,
|
1084 |
+
"grad_norm": 0.3574217944877886,
|
1085 |
+
"learning_rate": 0.0001072754036400944,
|
1086 |
+
"loss": 0.7886,
|
1087 |
+
"step": 154
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.496,
|
1091 |
+
"grad_norm": 0.39271223057544774,
|
1092 |
+
"learning_rate": 0.00010623752302063283,
|
1093 |
+
"loss": 0.8542,
|
1094 |
+
"step": 155
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.4992,
|
1098 |
+
"grad_norm": 0.4324299524551468,
|
1099 |
+
"learning_rate": 0.00010519896741619803,
|
1100 |
+
"loss": 0.8882,
|
1101 |
+
"step": 156
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.5024,
|
1105 |
+
"grad_norm": 0.32450275667877587,
|
1106 |
+
"learning_rate": 0.00010415984921265609,
|
1107 |
+
"loss": 0.7915,
|
1108 |
+
"step": 157
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.5056,
|
1112 |
+
"grad_norm": 0.37732713298950105,
|
1113 |
+
"learning_rate": 0.00010312028085675391,
|
1114 |
+
"loss": 0.7915,
|
1115 |
+
"step": 158
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.5088,
|
1119 |
+
"grad_norm": 0.3406928749347663,
|
1120 |
+
"learning_rate": 0.00010208037484395114,
|
1121 |
+
"loss": 0.817,
|
1122 |
+
"step": 159
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.512,
|
1126 |
+
"grad_norm": 0.3751715626743408,
|
1127 |
+
"learning_rate": 0.00010104024370624644,
|
1128 |
+
"loss": 0.8437,
|
1129 |
+
"step": 160
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.5152,
|
1133 |
+
"grad_norm": 0.33757211373718954,
|
1134 |
+
"learning_rate": 0.0001,
|
1135 |
+
"loss": 0.7485,
|
1136 |
+
"step": 161
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.5184,
|
1140 |
+
"grad_norm": 0.46079705496359785,
|
1141 |
+
"learning_rate": 9.895975629375359e-05,
|
1142 |
+
"loss": 0.8788,
|
1143 |
+
"step": 162
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.5216,
|
1147 |
+
"grad_norm": 0.4271610568124271,
|
1148 |
+
"learning_rate": 9.791962515604887e-05,
|
1149 |
+
"loss": 0.8616,
|
1150 |
+
"step": 163
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.5248,
|
1154 |
+
"grad_norm": 0.3730244116708226,
|
1155 |
+
"learning_rate": 9.687971914324607e-05,
|
1156 |
+
"loss": 0.713,
|
1157 |
+
"step": 164
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.528,
|
1161 |
+
"grad_norm": 0.32767683574358164,
|
1162 |
+
"learning_rate": 9.584015078734395e-05,
|
1163 |
+
"loss": 0.7503,
|
1164 |
+
"step": 165
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.5312,
|
1168 |
+
"grad_norm": 0.32244567576195554,
|
1169 |
+
"learning_rate": 9.480103258380198e-05,
|
1170 |
+
"loss": 0.7239,
|
1171 |
+
"step": 166
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.5344,
|
1175 |
+
"grad_norm": 0.3397126044271277,
|
1176 |
+
"learning_rate": 9.376247697936719e-05,
|
1177 |
+
"loss": 0.7984,
|
1178 |
+
"step": 167
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.5376,
|
1182 |
+
"grad_norm": 0.3012960591552935,
|
1183 |
+
"learning_rate": 9.272459635990562e-05,
|
1184 |
+
"loss": 0.7531,
|
1185 |
+
"step": 168
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.5408,
|
1189 |
+
"grad_norm": 0.37938450909301,
|
1190 |
+
"learning_rate": 9.168750303824084e-05,
|
1191 |
+
"loss": 0.7915,
|
1192 |
+
"step": 169
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.544,
|
1196 |
+
"grad_norm": 0.38701381132864615,
|
1197 |
+
"learning_rate": 9.065130924199998e-05,
|
1198 |
+
"loss": 0.7544,
|
1199 |
+
"step": 170
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.5472,
|
1203 |
+
"grad_norm": 0.38690509598881456,
|
1204 |
+
"learning_rate": 8.961612710146934e-05,
|
1205 |
+
"loss": 0.8261,
|
1206 |
+
"step": 171
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.5504,
|
1210 |
+
"grad_norm": 0.3662178194553636,
|
1211 |
+
"learning_rate": 8.858206863746018e-05,
|
1212 |
+
"loss": 0.7712,
|
1213 |
+
"step": 172
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.5536,
|
1217 |
+
"grad_norm": 0.325465438950959,
|
1218 |
+
"learning_rate": 8.754924574918675e-05,
|
1219 |
+
"loss": 0.7461,
|
1220 |
+
"step": 173
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.5568,
|
1224 |
+
"grad_norm": 0.36372519207365633,
|
1225 |
+
"learning_rate": 8.651777020215712e-05,
|
1226 |
+
"loss": 0.7577,
|
1227 |
+
"step": 174
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.56,
|
1231 |
+
"grad_norm": 0.39791618344605173,
|
1232 |
+
"learning_rate": 8.548775361607872e-05,
|
1233 |
+
"loss": 0.8952,
|
1234 |
+
"step": 175
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.5632,
|
1238 |
+
"grad_norm": 0.39106877210397273,
|
1239 |
+
"learning_rate": 8.445930745277953e-05,
|
1240 |
+
"loss": 0.8067,
|
1241 |
+
"step": 176
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.5664,
|
1245 |
+
"grad_norm": 0.32165292969442727,
|
1246 |
+
"learning_rate": 8.343254300414628e-05,
|
1247 |
+
"loss": 0.704,
|
1248 |
+
"step": 177
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.5696,
|
1252 |
+
"grad_norm": 0.36154222910864736,
|
1253 |
+
"learning_rate": 8.240757138008149e-05,
|
1254 |
+
"loss": 0.8184,
|
1255 |
+
"step": 178
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.5728,
|
1259 |
+
"grad_norm": 0.42719982445728705,
|
1260 |
+
"learning_rate": 8.138450349647936e-05,
|
1261 |
+
"loss": 0.8165,
|
1262 |
+
"step": 179
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.576,
|
1266 |
+
"grad_norm": 0.5246656865774878,
|
1267 |
+
"learning_rate": 8.036345006322359e-05,
|
1268 |
+
"loss": 0.9235,
|
1269 |
+
"step": 180
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.5792,
|
1273 |
+
"grad_norm": 0.3889988384699533,
|
1274 |
+
"learning_rate": 7.934452157220694e-05,
|
1275 |
+
"loss": 0.8564,
|
1276 |
+
"step": 181
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.5824,
|
1280 |
+
"grad_norm": 0.34293279991160774,
|
1281 |
+
"learning_rate": 7.832782828537437e-05,
|
1282 |
+
"loss": 0.7917,
|
1283 |
+
"step": 182
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.5856,
|
1287 |
+
"grad_norm": 0.43448804859625373,
|
1288 |
+
"learning_rate": 7.731348022279134e-05,
|
1289 |
+
"loss": 0.8397,
|
1290 |
+
"step": 183
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.5888,
|
1294 |
+
"grad_norm": 0.37332421671048194,
|
1295 |
+
"learning_rate": 7.630158715073813e-05,
|
1296 |
+
"loss": 0.847,
|
1297 |
+
"step": 184
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.592,
|
1301 |
+
"grad_norm": 0.2847408908726539,
|
1302 |
+
"learning_rate": 7.52922585698315e-05,
|
1303 |
+
"loss": 0.7008,
|
1304 |
+
"step": 185
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.5952,
|
1308 |
+
"grad_norm": 0.3793630586511894,
|
1309 |
+
"learning_rate": 7.428560370317542e-05,
|
1310 |
+
"loss": 0.8073,
|
1311 |
+
"step": 186
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.5984,
|
1315 |
+
"grad_norm": 0.3314978495730383,
|
1316 |
+
"learning_rate": 7.328173148454151e-05,
|
1317 |
+
"loss": 0.7501,
|
1318 |
+
"step": 187
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.6016,
|
1322 |
+
"grad_norm": 0.38192244839668066,
|
1323 |
+
"learning_rate": 7.228075054658096e-05,
|
1324 |
+
"loss": 0.8806,
|
1325 |
+
"step": 188
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.6048,
|
1329 |
+
"grad_norm": 0.35144673379944147,
|
1330 |
+
"learning_rate": 7.1282769209069e-05,
|
1331 |
+
"loss": 0.7983,
|
1332 |
+
"step": 189
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.608,
|
1336 |
+
"grad_norm": 0.3320272570958794,
|
1337 |
+
"learning_rate": 7.028789546718326e-05,
|
1338 |
+
"loss": 0.7778,
|
1339 |
+
"step": 190
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.6112,
|
1343 |
+
"grad_norm": 0.3245100936187699,
|
1344 |
+
"learning_rate": 6.929623697981718e-05,
|
1345 |
+
"loss": 0.7638,
|
1346 |
+
"step": 191
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.6144,
|
1350 |
+
"grad_norm": 0.3487129086320174,
|
1351 |
+
"learning_rate": 6.830790105792973e-05,
|
1352 |
+
"loss": 0.771,
|
1353 |
+
"step": 192
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.6176,
|
1357 |
+
"grad_norm": 0.35931606259571913,
|
1358 |
+
"learning_rate": 6.732299465293322e-05,
|
1359 |
+
"loss": 0.7663,
|
1360 |
+
"step": 193
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.6208,
|
1364 |
+
"grad_norm": 0.3367390991318867,
|
1365 |
+
"learning_rate": 6.63416243451194e-05,
|
1366 |
+
"loss": 0.7966,
|
1367 |
+
"step": 194
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.624,
|
1371 |
+
"grad_norm": 0.33183255910992066,
|
1372 |
+
"learning_rate": 6.536389633212609e-05,
|
1373 |
+
"loss": 0.78,
|
1374 |
+
"step": 195
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.6272,
|
1378 |
+
"grad_norm": 0.3264528270630262,
|
1379 |
+
"learning_rate": 6.43899164174453e-05,
|
1380 |
+
"loss": 0.7772,
|
1381 |
+
"step": 196
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.6304,
|
1385 |
+
"grad_norm": 0.3517378205790495,
|
1386 |
+
"learning_rate": 6.341978999897365e-05,
|
1387 |
+
"loss": 0.7642,
|
1388 |
+
"step": 197
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.6336,
|
1392 |
+
"grad_norm": 0.431338962220006,
|
1393 |
+
"learning_rate": 6.245362205760704e-05,
|
1394 |
+
"loss": 0.7927,
|
1395 |
+
"step": 198
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.6368,
|
1399 |
+
"grad_norm": 0.3731631592093385,
|
1400 |
+
"learning_rate": 6.149151714588009e-05,
|
1401 |
+
"loss": 0.8503,
|
1402 |
+
"step": 199
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.64,
|
1406 |
+
"grad_norm": 0.33909105312590193,
|
1407 |
+
"learning_rate": 6.053357937665237e-05,
|
1408 |
+
"loss": 0.7154,
|
1409 |
+
"step": 200
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.6432,
|
1413 |
+
"grad_norm": 0.36200420260594113,
|
1414 |
+
"learning_rate": 5.957991241184184e-05,
|
1415 |
+
"loss": 0.7569,
|
1416 |
+
"step": 201
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.6464,
|
1420 |
+
"grad_norm": 0.4149004455820632,
|
1421 |
+
"learning_rate": 5.863061945120719e-05,
|
1422 |
+
"loss": 0.8286,
|
1423 |
+
"step": 202
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.6496,
|
1427 |
+
"grad_norm": 0.32292850370856624,
|
1428 |
+
"learning_rate": 5.768580322118034e-05,
|
1429 |
+
"loss": 0.8015,
|
1430 |
+
"step": 203
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.6528,
|
1434 |
+
"grad_norm": 0.3771935747682809,
|
1435 |
+
"learning_rate": 5.6745565963749925e-05,
|
1436 |
+
"loss": 0.803,
|
1437 |
+
"step": 204
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.656,
|
1441 |
+
"grad_norm": 0.3541848583253109,
|
1442 |
+
"learning_rate": 5.5810009425397294e-05,
|
1443 |
+
"loss": 0.8545,
|
1444 |
+
"step": 205
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 0.6592,
|
1448 |
+
"grad_norm": 0.44832715565151354,
|
1449 |
+
"learning_rate": 5.487923484608629e-05,
|
1450 |
+
"loss": 0.8812,
|
1451 |
+
"step": 206
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 0.6624,
|
1455 |
+
"grad_norm": 0.37472326339826867,
|
1456 |
+
"learning_rate": 5.395334294830765e-05,
|
1457 |
+
"loss": 0.9129,
|
1458 |
+
"step": 207
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 0.6656,
|
1462 |
+
"grad_norm": 0.35228666839882417,
|
1463 |
+
"learning_rate": 5.3032433926179395e-05,
|
1464 |
+
"loss": 0.7832,
|
1465 |
+
"step": 208
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 0.6688,
|
1469 |
+
"grad_norm": 0.3350939709150189,
|
1470 |
+
"learning_rate": 5.211660743460458e-05,
|
1471 |
+
"loss": 0.7491,
|
1472 |
+
"step": 209
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 0.672,
|
1476 |
+
"grad_norm": 0.38948109383274765,
|
1477 |
+
"learning_rate": 5.1205962578487155e-05,
|
1478 |
+
"loss": 0.7651,
|
1479 |
+
"step": 210
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 0.6752,
|
1483 |
+
"grad_norm": 0.32094235443298985,
|
1484 |
+
"learning_rate": 5.030059790200756e-05,
|
1485 |
+
"loss": 0.769,
|
1486 |
+
"step": 211
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 0.6784,
|
1490 |
+
"grad_norm": 0.30333546694103175,
|
1491 |
+
"learning_rate": 4.940061137795876e-05,
|
1492 |
+
"loss": 0.6676,
|
1493 |
+
"step": 212
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.6816,
|
1497 |
+
"grad_norm": 0.36011644742045107,
|
1498 |
+
"learning_rate": 4.850610039714444e-05,
|
1499 |
+
"loss": 0.7948,
|
1500 |
+
"step": 213
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 0.6848,
|
1504 |
+
"grad_norm": 0.4501270149702274,
|
1505 |
+
"learning_rate": 4.761716175783989e-05,
|
1506 |
+
"loss": 0.8075,
|
1507 |
+
"step": 214
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 0.688,
|
1511 |
+
"grad_norm": 0.32443270170477734,
|
1512 |
+
"learning_rate": 4.673389165531714e-05,
|
1513 |
+
"loss": 0.7327,
|
1514 |
+
"step": 215
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 0.6912,
|
1518 |
+
"grad_norm": 0.35437103043306173,
|
1519 |
+
"learning_rate": 4.585638567143529e-05,
|
1520 |
+
"loss": 0.7306,
|
1521 |
+
"step": 216
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 0.6944,
|
1525 |
+
"grad_norm": 0.32026538320410597,
|
1526 |
+
"learning_rate": 4.498473876429726e-05,
|
1527 |
+
"loss": 0.7387,
|
1528 |
+
"step": 217
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 0.6976,
|
1532 |
+
"grad_norm": 0.3565566202307532,
|
1533 |
+
"learning_rate": 4.411904525797408e-05,
|
1534 |
+
"loss": 0.7722,
|
1535 |
+
"step": 218
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 0.7008,
|
1539 |
+
"grad_norm": 0.3770279902011318,
|
1540 |
+
"learning_rate": 4.325939883229766e-05,
|
1541 |
+
"loss": 0.8262,
|
1542 |
+
"step": 219
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 0.704,
|
1546 |
+
"grad_norm": 0.3225427774272655,
|
1547 |
+
"learning_rate": 4.240589251272342e-05,
|
1548 |
+
"loss": 0.7808,
|
1549 |
+
"step": 220
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 0.7072,
|
1553 |
+
"grad_norm": 0.3643305874741898,
|
1554 |
+
"learning_rate": 4.155861866026364e-05,
|
1555 |
+
"loss": 0.6751,
|
1556 |
+
"step": 221
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.7104,
|
1560 |
+
"grad_norm": 0.3495365472341216,
|
1561 |
+
"learning_rate": 4.071766896149273e-05,
|
1562 |
+
"loss": 0.7634,
|
1563 |
+
"step": 222
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 0.7136,
|
1567 |
+
"grad_norm": 0.3865725013932627,
|
1568 |
+
"learning_rate": 3.988313441862553e-05,
|
1569 |
+
"loss": 0.8201,
|
1570 |
+
"step": 223
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 0.7168,
|
1574 |
+
"grad_norm": 0.39371326689707414,
|
1575 |
+
"learning_rate": 3.9055105339669595e-05,
|
1576 |
+
"loss": 0.783,
|
1577 |
+
"step": 224
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 0.72,
|
1581 |
+
"grad_norm": 0.3343399455415386,
|
1582 |
+
"learning_rate": 3.823367132865265e-05,
|
1583 |
+
"loss": 0.7558,
|
1584 |
+
"step": 225
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 0.7232,
|
1588 |
+
"grad_norm": 0.37018368726419704,
|
1589 |
+
"learning_rate": 3.741892127592625e-05,
|
1590 |
+
"loss": 0.8034,
|
1591 |
+
"step": 226
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 0.7264,
|
1595 |
+
"grad_norm": 0.2917671574400929,
|
1596 |
+
"learning_rate": 3.6610943348546526e-05,
|
1597 |
+
"loss": 0.7608,
|
1598 |
+
"step": 227
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 0.7296,
|
1602 |
+
"grad_norm": 0.3149055123541856,
|
1603 |
+
"learning_rate": 3.580982498073344e-05,
|
1604 |
+
"loss": 0.7376,
|
1605 |
+
"step": 228
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 0.7328,
|
1609 |
+
"grad_norm": 0.48194732886134795,
|
1610 |
+
"learning_rate": 3.501565286440914e-05,
|
1611 |
+
"loss": 0.8543,
|
1612 |
+
"step": 229
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 0.736,
|
1616 |
+
"grad_norm": 0.3201762925229535,
|
1617 |
+
"learning_rate": 3.422851293981676e-05,
|
1618 |
+
"loss": 0.7311,
|
1619 |
+
"step": 230
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 0.7392,
|
1623 |
+
"grad_norm": 0.3832969139432672,
|
1624 |
+
"learning_rate": 3.3448490386220355e-05,
|
1625 |
+
"loss": 0.8361,
|
1626 |
+
"step": 231
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 0.7424,
|
1630 |
+
"grad_norm": 0.3591778802980467,
|
1631 |
+
"learning_rate": 3.2675669612687565e-05,
|
1632 |
+
"loss": 0.8246,
|
1633 |
+
"step": 232
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 0.7456,
|
1637 |
+
"grad_norm": 0.33711687357568193,
|
1638 |
+
"learning_rate": 3.191013424895536e-05,
|
1639 |
+
"loss": 0.7911,
|
1640 |
+
"step": 233
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 0.7488,
|
1644 |
+
"grad_norm": 0.3992050715809478,
|
1645 |
+
"learning_rate": 3.115196713638e-05,
|
1646 |
+
"loss": 0.784,
|
1647 |
+
"step": 234
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 0.752,
|
1651 |
+
"grad_norm": 0.3717791655175765,
|
1652 |
+
"learning_rate": 3.040125031897264e-05,
|
1653 |
+
"loss": 0.8544,
|
1654 |
+
"step": 235
|
1655 |
+
},
|
1656 |
+
{
|
1657 |
+
"epoch": 0.7552,
|
1658 |
+
"grad_norm": 0.3266902771814692,
|
1659 |
+
"learning_rate": 2.9658065034520978e-05,
|
1660 |
+
"loss": 0.758,
|
1661 |
+
"step": 236
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 0.7584,
|
1665 |
+
"grad_norm": 0.35563021204215095,
|
1666 |
+
"learning_rate": 2.892249170579826e-05,
|
1667 |
+
"loss": 0.7699,
|
1668 |
+
"step": 237
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 0.7616,
|
1672 |
+
"grad_norm": 0.379714101588036,
|
1673 |
+
"learning_rate": 2.8194609931860316e-05,
|
1674 |
+
"loss": 0.8409,
|
1675 |
+
"step": 238
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 0.7648,
|
1679 |
+
"grad_norm": 0.47451149824579747,
|
1680 |
+
"learning_rate": 2.7474498479432087e-05,
|
1681 |
+
"loss": 0.8289,
|
1682 |
+
"step": 239
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.768,
|
1686 |
+
"grad_norm": 0.3188599450071592,
|
1687 |
+
"learning_rate": 2.6762235274383772e-05,
|
1688 |
+
"loss": 0.7431,
|
1689 |
+
"step": 240
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 0.7712,
|
1693 |
+
"grad_norm": 0.3537473485993502,
|
1694 |
+
"learning_rate": 2.6057897393298324e-05,
|
1695 |
+
"loss": 0.7519,
|
1696 |
+
"step": 241
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 0.7744,
|
1700 |
+
"grad_norm": 0.35260123842359997,
|
1701 |
+
"learning_rate": 2.536156105513062e-05,
|
1702 |
+
"loss": 0.8052,
|
1703 |
+
"step": 242
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 0.7776,
|
1707 |
+
"grad_norm": 0.3782134918689213,
|
1708 |
+
"learning_rate": 2.4673301612959654e-05,
|
1709 |
+
"loss": 0.8498,
|
1710 |
+
"step": 243
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 0.7808,
|
1714 |
+
"grad_norm": 0.4240797008066036,
|
1715 |
+
"learning_rate": 2.399319354583418e-05,
|
1716 |
+
"loss": 0.8504,
|
1717 |
+
"step": 244
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 0.784,
|
1721 |
+
"grad_norm": 0.3493922027357962,
|
1722 |
+
"learning_rate": 2.3321310450713062e-05,
|
1723 |
+
"loss": 0.748,
|
1724 |
+
"step": 245
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 0.7872,
|
1728 |
+
"grad_norm": 0.3109975196467063,
|
1729 |
+
"learning_rate": 2.265772503450122e-05,
|
1730 |
+
"loss": 0.7332,
|
1731 |
+
"step": 246
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 0.7904,
|
1735 |
+
"grad_norm": 0.3490386685549296,
|
1736 |
+
"learning_rate": 2.2002509106181624e-05,
|
1737 |
+
"loss": 0.7745,
|
1738 |
+
"step": 247
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 0.7936,
|
1742 |
+
"grad_norm": 0.36175624431520176,
|
1743 |
+
"learning_rate": 2.1355733569044635e-05,
|
1744 |
+
"loss": 0.7388,
|
1745 |
+
"step": 248
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 0.7968,
|
1749 |
+
"grad_norm": 0.4367153475876239,
|
1750 |
+
"learning_rate": 2.0717468413015283e-05,
|
1751 |
+
"loss": 0.8179,
|
1752 |
+
"step": 249
|
1753 |
+
},
|
1754 |
+
{
|
1755 |
+
"epoch": 0.8,
|
1756 |
+
"grad_norm": 0.39966660383957925,
|
1757 |
+
"learning_rate": 2.008778270707944e-05,
|
1758 |
+
"loss": 0.7862,
|
1759 |
+
"step": 250
|
1760 |
+
},
|
1761 |
+
{
|
1762 |
+
"epoch": 0.8032,
|
1763 |
+
"grad_norm": 0.39454280119505963,
|
1764 |
+
"learning_rate": 1.946674459180955e-05,
|
1765 |
+
"loss": 0.7943,
|
1766 |
+
"step": 251
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 0.8064,
|
1770 |
+
"grad_norm": 0.34240338202982135,
|
1771 |
+
"learning_rate": 1.8854421271990964e-05,
|
1772 |
+
"loss": 0.7764,
|
1773 |
+
"step": 252
|
1774 |
+
},
|
1775 |
+
{
|
1776 |
+
"epoch": 0.8096,
|
1777 |
+
"grad_norm": 0.3836219267710939,
|
1778 |
+
"learning_rate": 1.8250879009349398e-05,
|
1779 |
+
"loss": 0.7249,
|
1780 |
+
"step": 253
|
1781 |
+
},
|
1782 |
+
{
|
1783 |
+
"epoch": 0.8128,
|
1784 |
+
"grad_norm": 0.3545049008632808,
|
1785 |
+
"learning_rate": 1.7656183115380577e-05,
|
1786 |
+
"loss": 0.745,
|
1787 |
+
"step": 254
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 0.816,
|
1791 |
+
"grad_norm": 0.4782319117565856,
|
1792 |
+
"learning_rate": 1.707039794428259e-05,
|
1793 |
+
"loss": 0.9546,
|
1794 |
+
"step": 255
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 0.8192,
|
1798 |
+
"grad_norm": 0.30912276688779844,
|
1799 |
+
"learning_rate": 1.649358688599191e-05,
|
1800 |
+
"loss": 0.7098,
|
1801 |
+
"step": 256
|
1802 |
+
},
|
1803 |
+
{
|
1804 |
+
"epoch": 0.8224,
|
1805 |
+
"grad_norm": 0.3564793276918867,
|
1806 |
+
"learning_rate": 1.5925812359323745e-05,
|
1807 |
+
"loss": 0.7553,
|
1808 |
+
"step": 257
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 0.8256,
|
1812 |
+
"grad_norm": 0.34525121870000597,
|
1813 |
+
"learning_rate": 1.5367135805217458e-05,
|
1814 |
+
"loss": 0.7499,
|
1815 |
+
"step": 258
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 0.8288,
|
1819 |
+
"grad_norm": 0.3349420082155695,
|
1820 |
+
"learning_rate": 1.4817617680087825e-05,
|
1821 |
+
"loss": 0.805,
|
1822 |
+
"step": 259
|
1823 |
+
},
|
1824 |
+
{
|
1825 |
+
"epoch": 0.832,
|
1826 |
+
"grad_norm": 0.433540567668089,
|
1827 |
+
"learning_rate": 1.4277317449282834e-05,
|
1828 |
+
"loss": 0.877,
|
1829 |
+
"step": 260
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 0.8352,
|
1833 |
+
"grad_norm": 0.3801150070687427,
|
1834 |
+
"learning_rate": 1.3746293580648717e-05,
|
1835 |
+
"loss": 0.8694,
|
1836 |
+
"step": 261
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 0.8384,
|
1840 |
+
"grad_norm": 0.3738768456908641,
|
1841 |
+
"learning_rate": 1.3224603538202929e-05,
|
1842 |
+
"loss": 0.7737,
|
1843 |
+
"step": 262
|
1844 |
+
},
|
1845 |
+
{
|
1846 |
+
"epoch": 0.8416,
|
1847 |
+
"grad_norm": 0.3801506817792511,
|
1848 |
+
"learning_rate": 1.2712303775915802e-05,
|
1849 |
+
"loss": 0.7256,
|
1850 |
+
"step": 263
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 0.8448,
|
1854 |
+
"grad_norm": 0.29294064483194054,
|
1855 |
+
"learning_rate": 1.220944973160133e-05,
|
1856 |
+
"loss": 0.7278,
|
1857 |
+
"step": 264
|
1858 |
+
},
|
1859 |
+
{
|
1860 |
+
"epoch": 0.848,
|
1861 |
+
"grad_norm": 0.3574084890505441,
|
1862 |
+
"learning_rate": 1.1716095820918216e-05,
|
1863 |
+
"loss": 0.8017,
|
1864 |
+
"step": 265
|
1865 |
+
},
|
1866 |
+
{
|
1867 |
+
"epoch": 0.8512,
|
1868 |
+
"grad_norm": 0.44325322418927066,
|
1869 |
+
"learning_rate": 1.1232295431481222e-05,
|
1870 |
+
"loss": 0.8188,
|
1871 |
+
"step": 266
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 0.8544,
|
1875 |
+
"grad_norm": 0.33465511889302607,
|
1876 |
+
"learning_rate": 1.0758100917083991e-05,
|
1877 |
+
"loss": 0.7404,
|
1878 |
+
"step": 267
|
1879 |
+
},
|
1880 |
+
{
|
1881 |
+
"epoch": 0.8576,
|
1882 |
+
"grad_norm": 0.35582143701746155,
|
1883 |
+
"learning_rate": 1.0293563592033595e-05,
|
1884 |
+
"loss": 0.828,
|
1885 |
+
"step": 268
|
1886 |
+
},
|
1887 |
+
{
|
1888 |
+
"epoch": 0.8608,
|
1889 |
+
"grad_norm": 0.32908366694518915,
|
1890 |
+
"learning_rate": 9.838733725597615e-06,
|
1891 |
+
"loss": 0.7496,
|
1892 |
+
"step": 269
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 0.864,
|
1896 |
+
"grad_norm": 0.3923286422615176,
|
1897 |
+
"learning_rate": 9.393660536564408e-06,
|
1898 |
+
"loss": 0.7605,
|
1899 |
+
"step": 270
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 0.8672,
|
1903 |
+
"grad_norm": 0.32871379992022076,
|
1904 |
+
"learning_rate": 8.958392187916841e-06,
|
1905 |
+
"loss": 0.7252,
|
1906 |
+
"step": 271
|
1907 |
+
},
|
1908 |
+
{
|
1909 |
+
"epoch": 0.8704,
|
1910 |
+
"grad_norm": 0.4464878317171815,
|
1911 |
+
"learning_rate": 8.532975781620512e-06,
|
1912 |
+
"loss": 0.7487,
|
1913 |
+
"step": 272
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 0.8736,
|
1917 |
+
"grad_norm": 0.37431391710968503,
|
1918 |
+
"learning_rate": 8.117457353526625e-06,
|
1919 |
+
"loss": 0.7393,
|
1920 |
+
"step": 273
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 0.8768,
|
1924 |
+
"grad_norm": 0.31383965433569194,
|
1925 |
+
"learning_rate": 7.711881868390291e-06,
|
1926 |
+
"loss": 0.7417,
|
1927 |
+
"step": 274
|
1928 |
+
},
|
1929 |
+
{
|
1930 |
+
"epoch": 0.88,
|
1931 |
+
"grad_norm": 0.40622594368739234,
|
1932 |
+
"learning_rate": 7.3162932150046885e-06,
|
1933 |
+
"loss": 0.8212,
|
1934 |
+
"step": 275
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 0.8832,
|
1938 |
+
"grad_norm": 0.312688075077189,
|
1939 |
+
"learning_rate": 6.930734201451816e-06,
|
1940 |
+
"loss": 0.6931,
|
1941 |
+
"step": 276
|
1942 |
+
},
|
1943 |
+
{
|
1944 |
+
"epoch": 0.8864,
|
1945 |
+
"grad_norm": 0.2951795837647385,
|
1946 |
+
"learning_rate": 6.555246550469907e-06,
|
1947 |
+
"loss": 0.7062,
|
1948 |
+
"step": 277
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 0.8896,
|
1952 |
+
"grad_norm": 0.3906241449752177,
|
1953 |
+
"learning_rate": 6.189870894938587e-06,
|
1954 |
+
"loss": 0.8502,
|
1955 |
+
"step": 278
|
1956 |
+
},
|
1957 |
+
{
|
1958 |
+
"epoch": 0.8928,
|
1959 |
+
"grad_norm": 0.33976393181898756,
|
1960 |
+
"learning_rate": 5.834646773481811e-06,
|
1961 |
+
"loss": 0.7466,
|
1962 |
+
"step": 279
|
1963 |
+
},
|
1964 |
+
{
|
1965 |
+
"epoch": 0.896,
|
1966 |
+
"grad_norm": 0.3645631566569671,
|
1967 |
+
"learning_rate": 5.489612626189245e-06,
|
1968 |
+
"loss": 0.8331,
|
1969 |
+
"step": 280
|
1970 |
+
},
|
1971 |
+
{
|
1972 |
+
"epoch": 0.8992,
|
1973 |
+
"grad_norm": 0.32171725161543063,
|
1974 |
+
"learning_rate": 5.154805790456485e-06,
|
1975 |
+
"loss": 0.7301,
|
1976 |
+
"step": 281
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 0.9024,
|
1980 |
+
"grad_norm": 0.3215222867159588,
|
1981 |
+
"learning_rate": 4.830262496944693e-06,
|
1982 |
+
"loss": 0.7803,
|
1983 |
+
"step": 282
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 0.9056,
|
1987 |
+
"grad_norm": 0.3930936220455362,
|
1988 |
+
"learning_rate": 4.516017865659949e-06,
|
1989 |
+
"loss": 0.8015,
|
1990 |
+
"step": 283
|
1991 |
+
},
|
1992 |
+
{
|
1993 |
+
"epoch": 0.9088,
|
1994 |
+
"grad_norm": 0.4009757739317676,
|
1995 |
+
"learning_rate": 4.21210590215273e-06,
|
1996 |
+
"loss": 0.7702,
|
1997 |
+
"step": 284
|
1998 |
+
},
|
1999 |
+
{
|
2000 |
+
"epoch": 0.912,
|
2001 |
+
"grad_norm": 0.3509902652942518,
|
2002 |
+
"learning_rate": 3.918559493838114e-06,
|
2003 |
+
"loss": 0.8313,
|
2004 |
+
"step": 285
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 0.9152,
|
2008 |
+
"grad_norm": 0.4823527769540213,
|
2009 |
+
"learning_rate": 3.6354104064368566e-06,
|
2010 |
+
"loss": 0.7737,
|
2011 |
+
"step": 286
|
2012 |
+
},
|
2013 |
+
{
|
2014 |
+
"epoch": 0.9184,
|
2015 |
+
"grad_norm": 0.37189019955946206,
|
2016 |
+
"learning_rate": 3.3626892805379562e-06,
|
2017 |
+
"loss": 0.7485,
|
2018 |
+
"step": 287
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 0.9216,
|
2022 |
+
"grad_norm": 0.3929380712488443,
|
2023 |
+
"learning_rate": 3.100425628282899e-06,
|
2024 |
+
"loss": 0.7541,
|
2025 |
+
"step": 288
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 0.9248,
|
2029 |
+
"grad_norm": 0.3230172315650001,
|
2030 |
+
"learning_rate": 2.848647830172024e-06,
|
2031 |
+
"loss": 0.7514,
|
2032 |
+
"step": 289
|
2033 |
+
},
|
2034 |
+
{
|
2035 |
+
"epoch": 0.928,
|
2036 |
+
"grad_norm": 0.331298596116422,
|
2037 |
+
"learning_rate": 2.607383131993424e-06,
|
2038 |
+
"loss": 0.7476,
|
2039 |
+
"step": 290
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 0.9312,
|
2043 |
+
"grad_norm": 0.41208820254007095,
|
2044 |
+
"learning_rate": 2.3766576418745022e-06,
|
2045 |
+
"loss": 0.8253,
|
2046 |
+
"step": 291
|
2047 |
+
},
|
2048 |
+
{
|
2049 |
+
"epoch": 0.9344,
|
2050 |
+
"grad_norm": 0.3540797011234827,
|
2051 |
+
"learning_rate": 2.1564963274568027e-06,
|
2052 |
+
"loss": 0.7754,
|
2053 |
+
"step": 292
|
2054 |
+
},
|
2055 |
+
{
|
2056 |
+
"epoch": 0.9376,
|
2057 |
+
"grad_norm": 0.40367747859127645,
|
2058 |
+
"learning_rate": 1.9469230131940907e-06,
|
2059 |
+
"loss": 0.8121,
|
2060 |
+
"step": 293
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 0.9408,
|
2064 |
+
"grad_norm": 0.36985043876259127,
|
2065 |
+
"learning_rate": 1.7479603777742938e-06,
|
2066 |
+
"loss": 0.7875,
|
2067 |
+
"step": 294
|
2068 |
+
},
|
2069 |
+
{
|
2070 |
+
"epoch": 0.944,
|
2071 |
+
"grad_norm": 0.3757213402313114,
|
2072 |
+
"learning_rate": 1.559629951665298e-06,
|
2073 |
+
"loss": 0.7626,
|
2074 |
+
"step": 295
|
2075 |
+
},
|
2076 |
+
{
|
2077 |
+
"epoch": 0.9472,
|
2078 |
+
"grad_norm": 0.3519319526157233,
|
2079 |
+
"learning_rate": 1.3819521147851123e-06,
|
2080 |
+
"loss": 0.8127,
|
2081 |
+
"step": 296
|
2082 |
+
},
|
2083 |
+
{
|
2084 |
+
"epoch": 0.9504,
|
2085 |
+
"grad_norm": 0.3733908580196815,
|
2086 |
+
"learning_rate": 1.2149460942964098e-06,
|
2087 |
+
"loss": 0.7547,
|
2088 |
+
"step": 297
|
2089 |
+
},
|
2090 |
+
{
|
2091 |
+
"epoch": 0.9536,
|
2092 |
+
"grad_norm": 0.36965768774397506,
|
2093 |
+
"learning_rate": 1.05862996252597e-06,
|
2094 |
+
"loss": 0.7968,
|
2095 |
+
"step": 298
|
2096 |
+
},
|
2097 |
+
{
|
2098 |
+
"epoch": 0.9568,
|
2099 |
+
"grad_norm": 0.392672840473129,
|
2100 |
+
"learning_rate": 9.130206350089765e-07,
|
2101 |
+
"loss": 0.7971,
|
2102 |
+
"step": 299
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 0.96,
|
2106 |
+
"grad_norm": 0.35145171210243764,
|
2107 |
+
"learning_rate": 7.781338686584927e-07,
|
2108 |
+
"loss": 0.7712,
|
2109 |
+
"step": 300
|
2110 |
+
},
|
2111 |
+
{
|
2112 |
+
"epoch": 0.9632,
|
2113 |
+
"grad_norm": 0.3557470819008264,
|
2114 |
+
"learning_rate": 6.539842600603918e-07,
|
2115 |
+
"loss": 0.8122,
|
2116 |
+
"step": 301
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 0.9664,
|
2120 |
+
"grad_norm": 0.39875493221618996,
|
2121 |
+
"learning_rate": 5.405852438937764e-07,
|
2122 |
+
"loss": 0.7222,
|
2123 |
+
"step": 302
|
2124 |
+
},
|
2125 |
+
{
|
2126 |
+
"epoch": 0.9696,
|
2127 |
+
"grad_norm": 0.3398025671483795,
|
2128 |
+
"learning_rate": 4.3794909147720773e-07,
|
2129 |
+
"loss": 0.7241,
|
2130 |
+
"step": 303
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 0.9728,
|
2134 |
+
"grad_norm": 0.44734223609044316,
|
2135 |
+
"learning_rate": 3.4608690944071263e-07,
|
2136 |
+
"loss": 0.8064,
|
2137 |
+
"step": 304
|
2138 |
+
},
|
2139 |
+
{
|
2140 |
+
"epoch": 0.976,
|
2141 |
+
"grad_norm": 0.36261141044117756,
|
2142 |
+
"learning_rate": 2.6500863852395584e-07,
|
2143 |
+
"loss": 0.7815,
|
2144 |
+
"step": 305
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 0.9792,
|
2148 |
+
"grad_norm": 1.1171234591773465,
|
2149 |
+
"learning_rate": 1.947230525005006e-07,
|
2150 |
+
"loss": 0.7992,
|
2151 |
+
"step": 306
|
2152 |
+
},
|
2153 |
+
{
|
2154 |
+
"epoch": 0.9824,
|
2155 |
+
"grad_norm": 0.3255664161802146,
|
2156 |
+
"learning_rate": 1.3523775722834587e-07,
|
2157 |
+
"loss": 0.7498,
|
2158 |
+
"step": 307
|
2159 |
+
},
|
2160 |
+
{
|
2161 |
+
"epoch": 0.9856,
|
2162 |
+
"grad_norm": 0.34989009087457607,
|
2163 |
+
"learning_rate": 8.655918982689581e-08,
|
2164 |
+
"loss": 0.8037,
|
2165 |
+
"step": 308
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 0.9888,
|
2169 |
+
"grad_norm": 0.33217965843871977,
|
2170 |
+
"learning_rate": 4.8692617980350406e-08,
|
2171 |
+
"loss": 0.8156,
|
2172 |
+
"step": 309
|
2173 |
+
},
|
2174 |
+
{
|
2175 |
+
"epoch": 0.992,
|
2176 |
+
"grad_norm": 0.3603261858382364,
|
2177 |
+
"learning_rate": 2.164213936770576e-08,
|
2178 |
+
"loss": 0.7071,
|
2179 |
+
"step": 310
|
2180 |
+
},
|
2181 |
+
{
|
2182 |
+
"epoch": 0.9952,
|
2183 |
+
"grad_norm": 0.3558631704357258,
|
2184 |
+
"learning_rate": 5.410681219286673e-09,
|
2185 |
+
"loss": 0.742,
|
2186 |
+
"step": 311
|
2187 |
+
},
|
2188 |
+
{
|
2189 |
+
"epoch": 0.9984,
|
2190 |
+
"grad_norm": 0.3200726372631764,
|
2191 |
+
"learning_rate": 0.0,
|
2192 |
+
"loss": 0.7306,
|
2193 |
+
"step": 312
|
2194 |
+
},
|
2195 |
+
{
|
2196 |
+
"epoch": 0.9984,
|
2197 |
+
"step": 312,
|
2198 |
+
"total_flos": 739882583326720.0,
|
2199 |
+
"train_loss": 0.8309416449986972,
|
2200 |
+
"train_runtime": 9163.8195,
|
2201 |
+
"train_samples_per_second": 1.091,
|
2202 |
+
"train_steps_per_second": 0.034
|
2203 |
+
}
|
2204 |
+
],
|
2205 |
+
"logging_steps": 1.0,
|
2206 |
+
"max_steps": 312,
|
2207 |
+
"num_input_tokens_seen": 0,
|
2208 |
+
"num_train_epochs": 1,
|
2209 |
+
"save_steps": 500,
|
2210 |
+
"stateful_callbacks": {
|
2211 |
+
"TrainerControl": {
|
2212 |
+
"args": {
|
2213 |
+
"should_epoch_stop": false,
|
2214 |
+
"should_evaluate": false,
|
2215 |
+
"should_log": false,
|
2216 |
+
"should_save": false,
|
2217 |
+
"should_training_stop": false
|
2218 |
+
},
|
2219 |
+
"attributes": {}
|
2220 |
+
}
|
2221 |
+
},
|
2222 |
+
"total_flos": 739882583326720.0,
|
2223 |
+
"train_batch_size": 8,
|
2224 |
+
"trial_name": null,
|
2225 |
+
"trial_params": null
|
2226 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: ./weights/Bunny-v1_1-Llama-3-8B-V
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.11.1
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 256,
|
14 |
+
"lora_dropout": 0.1,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 128,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": null,
|
22 |
+
"target_modules": [
|
23 |
+
"q_proj",
|
24 |
+
"v_proj",
|
25 |
+
"gate_proj",
|
26 |
+
"k_proj",
|
27 |
+
"o_proj",
|
28 |
+
"down_proj",
|
29 |
+
"up_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:807df09410a8bac8441b10edd939efedcbdf49e2206c5d73aed6963a3087c65e
|
3 |
+
size 671150064
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/config.json
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
3 |
+
"architectures": [
|
4 |
+
"BunnyLlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"auto_map": {
|
9 |
+
"AutoConfig": "configuration_bunny_llama.BunnyLlamaConfig",
|
10 |
+
"AutoModelForCausalLM": "modeling_bunny_llama.BunnyLlamaForCausalLM"
|
11 |
+
},
|
12 |
+
"bos_token_id": 128000,
|
13 |
+
"continuous_training": false,
|
14 |
+
"eos_token_id": 128001,
|
15 |
+
"freeze_mm_mlp_adapter": false,
|
16 |
+
"hidden_act": "silu",
|
17 |
+
"hidden_size": 4096,
|
18 |
+
"image_aspect_ratio": "pad",
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 14336,
|
21 |
+
"max_position_embeddings": 8192,
|
22 |
+
"mm_hidden_size": 3456,
|
23 |
+
"mm_projector_lr": null,
|
24 |
+
"mm_projector_type": "mlp2x_gelu",
|
25 |
+
"mm_vision_tower": "./weights/siglip-so400m-patch14-384",
|
26 |
+
"model_type": "bunny-llama",
|
27 |
+
"num_attention_heads": 32,
|
28 |
+
"num_hidden_layers": 32,
|
29 |
+
"num_key_value_heads": 8,
|
30 |
+
"pretraining_tp": 1,
|
31 |
+
"rms_norm_eps": 1e-05,
|
32 |
+
"rope_scaling": null,
|
33 |
+
"rope_theta": 500000.0,
|
34 |
+
"tie_word_embeddings": false,
|
35 |
+
"tokenizer_model_max_length": 2048,
|
36 |
+
"tokenizer_padding_side": "right",
|
37 |
+
"torch_dtype": "float16",
|
38 |
+
"transformers_version": "4.41.2",
|
39 |
+
"tune_mm_mlp_adapter": false,
|
40 |
+
"unfreeze_vision_tower": true,
|
41 |
+
"use_cache": true,
|
42 |
+
"use_mm_proj": true,
|
43 |
+
"use_s2": true,
|
44 |
+
"vocab_size": 128256
|
45 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/non_lora_trainables.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4a957a55853852e70c4e4559e96f956dcefc8993421eea81bbc558164c8b426a
|
3 |
+
size 918507402
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_1_epochs_1_GA_2_lora/trainer_state.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: ./weights/Bunny-v1_1-Llama-3-8B-V
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.11.1
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 256,
|
14 |
+
"lora_dropout": 0.1,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 128,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": null,
|
22 |
+
"target_modules": [
|
23 |
+
"k_proj",
|
24 |
+
"q_proj",
|
25 |
+
"down_proj",
|
26 |
+
"up_proj",
|
27 |
+
"o_proj",
|
28 |
+
"v_proj",
|
29 |
+
"gate_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:222d6020a814fe4a9d38c7187c52bcc5c8fab80c418dd58f37f19789d5d3258e
|
3 |
+
size 671150064
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/config.json
ADDED
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
3 |
+
"architectures": [
|
4 |
+
"BunnyLlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"auto_map": {
|
9 |
+
"AutoConfig": "configuration_bunny_llama.BunnyLlamaConfig",
|
10 |
+
"AutoModelForCausalLM": "modeling_bunny_llama.BunnyLlamaForCausalLM"
|
11 |
+
},
|
12 |
+
"bos_token_id": 128000,
|
13 |
+
"continuous_training": false,
|
14 |
+
"eos_token_id": 128001,
|
15 |
+
"freeze_mm_mlp_adapter": false,
|
16 |
+
"hidden_act": "silu",
|
17 |
+
"hidden_size": 4096,
|
18 |
+
"image_aspect_ratio": "pad",
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 14336,
|
21 |
+
"max_position_embeddings": 8192,
|
22 |
+
"mm_hidden_size": 3456,
|
23 |
+
"mm_projector_lr": null,
|
24 |
+
"mm_projector_type": "mlp2x_gelu",
|
25 |
+
"mm_vision_tower": "./weights/siglip-so400m-patch14-384",
|
26 |
+
"model_type": "bunny-llama",
|
27 |
+
"num_attention_heads": 32,
|
28 |
+
"num_hidden_layers": 32,
|
29 |
+
"num_key_value_heads": 8,
|
30 |
+
"pretraining_tp": 1,
|
31 |
+
"rms_norm_eps": 1e-05,
|
32 |
+
"rope_scaling": null,
|
33 |
+
"rope_theta": 500000.0,
|
34 |
+
"tie_word_embeddings": false,
|
35 |
+
"tokenizer_model_max_length": 2048,
|
36 |
+
"tokenizer_padding_side": "right",
|
37 |
+
"torch_dtype": "float16",
|
38 |
+
"transformers_version": "4.41.2",
|
39 |
+
"tune_mm_mlp_adapter": false,
|
40 |
+
"unfreeze_vision_tower": true,
|
41 |
+
"use_cache": true,
|
42 |
+
"use_mm_proj": true,
|
43 |
+
"use_s2": true,
|
44 |
+
"vocab_size": 128256
|
45 |
+
}
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/non_lora_trainables.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:48d577f56281a55d9ebce8783da6a9f998306862aa87ce27b012361b7d4c3a07
|
3 |
+
size 918507402
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_2_epochs_1_GA_2_lora/trainer_state.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_3_epochs_1_GA_2_lora/README.md
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
base_model: ./weights/Bunny-v1_1-Llama-3-8B-V
|
4 |
+
---
|
5 |
+
|
6 |
+
# Model Card for Model ID
|
7 |
+
|
8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
9 |
+
|
10 |
+
|
11 |
+
|
12 |
+
## Model Details
|
13 |
+
|
14 |
+
### Model Description
|
15 |
+
|
16 |
+
<!-- Provide a longer summary of what this model is. -->
|
17 |
+
|
18 |
+
|
19 |
+
|
20 |
+
- **Developed by:** [More Information Needed]
|
21 |
+
- **Funded by [optional]:** [More Information Needed]
|
22 |
+
- **Shared by [optional]:** [More Information Needed]
|
23 |
+
- **Model type:** [More Information Needed]
|
24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
25 |
+
- **License:** [More Information Needed]
|
26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
27 |
+
|
28 |
+
### Model Sources [optional]
|
29 |
+
|
30 |
+
<!-- Provide the basic links for the model. -->
|
31 |
+
|
32 |
+
- **Repository:** [More Information Needed]
|
33 |
+
- **Paper [optional]:** [More Information Needed]
|
34 |
+
- **Demo [optional]:** [More Information Needed]
|
35 |
+
|
36 |
+
## Uses
|
37 |
+
|
38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
39 |
+
|
40 |
+
### Direct Use
|
41 |
+
|
42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
43 |
+
|
44 |
+
[More Information Needed]
|
45 |
+
|
46 |
+
### Downstream Use [optional]
|
47 |
+
|
48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
49 |
+
|
50 |
+
[More Information Needed]
|
51 |
+
|
52 |
+
### Out-of-Scope Use
|
53 |
+
|
54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
55 |
+
|
56 |
+
[More Information Needed]
|
57 |
+
|
58 |
+
## Bias, Risks, and Limitations
|
59 |
+
|
60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
61 |
+
|
62 |
+
[More Information Needed]
|
63 |
+
|
64 |
+
### Recommendations
|
65 |
+
|
66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
67 |
+
|
68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
69 |
+
|
70 |
+
## How to Get Started with the Model
|
71 |
+
|
72 |
+
Use the code below to get started with the model.
|
73 |
+
|
74 |
+
[More Information Needed]
|
75 |
+
|
76 |
+
## Training Details
|
77 |
+
|
78 |
+
### Training Data
|
79 |
+
|
80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
81 |
+
|
82 |
+
[More Information Needed]
|
83 |
+
|
84 |
+
### Training Procedure
|
85 |
+
|
86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
87 |
+
|
88 |
+
#### Preprocessing [optional]
|
89 |
+
|
90 |
+
[More Information Needed]
|
91 |
+
|
92 |
+
|
93 |
+
#### Training Hyperparameters
|
94 |
+
|
95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
96 |
+
|
97 |
+
#### Speeds, Sizes, Times [optional]
|
98 |
+
|
99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
100 |
+
|
101 |
+
[More Information Needed]
|
102 |
+
|
103 |
+
## Evaluation
|
104 |
+
|
105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
106 |
+
|
107 |
+
### Testing Data, Factors & Metrics
|
108 |
+
|
109 |
+
#### Testing Data
|
110 |
+
|
111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
112 |
+
|
113 |
+
[More Information Needed]
|
114 |
+
|
115 |
+
#### Factors
|
116 |
+
|
117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
118 |
+
|
119 |
+
[More Information Needed]
|
120 |
+
|
121 |
+
#### Metrics
|
122 |
+
|
123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
124 |
+
|
125 |
+
[More Information Needed]
|
126 |
+
|
127 |
+
### Results
|
128 |
+
|
129 |
+
[More Information Needed]
|
130 |
+
|
131 |
+
#### Summary
|
132 |
+
|
133 |
+
|
134 |
+
|
135 |
+
## Model Examination [optional]
|
136 |
+
|
137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
138 |
+
|
139 |
+
[More Information Needed]
|
140 |
+
|
141 |
+
## Environmental Impact
|
142 |
+
|
143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
144 |
+
|
145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
146 |
+
|
147 |
+
- **Hardware Type:** [More Information Needed]
|
148 |
+
- **Hours used:** [More Information Needed]
|
149 |
+
- **Cloud Provider:** [More Information Needed]
|
150 |
+
- **Compute Region:** [More Information Needed]
|
151 |
+
- **Carbon Emitted:** [More Information Needed]
|
152 |
+
|
153 |
+
## Technical Specifications [optional]
|
154 |
+
|
155 |
+
### Model Architecture and Objective
|
156 |
+
|
157 |
+
[More Information Needed]
|
158 |
+
|
159 |
+
### Compute Infrastructure
|
160 |
+
|
161 |
+
[More Information Needed]
|
162 |
+
|
163 |
+
#### Hardware
|
164 |
+
|
165 |
+
[More Information Needed]
|
166 |
+
|
167 |
+
#### Software
|
168 |
+
|
169 |
+
[More Information Needed]
|
170 |
+
|
171 |
+
## Citation [optional]
|
172 |
+
|
173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
174 |
+
|
175 |
+
**BibTeX:**
|
176 |
+
|
177 |
+
[More Information Needed]
|
178 |
+
|
179 |
+
**APA:**
|
180 |
+
|
181 |
+
[More Information Needed]
|
182 |
+
|
183 |
+
## Glossary [optional]
|
184 |
+
|
185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
186 |
+
|
187 |
+
[More Information Needed]
|
188 |
+
|
189 |
+
## More Information [optional]
|
190 |
+
|
191 |
+
[More Information Needed]
|
192 |
+
|
193 |
+
## Model Card Authors [optional]
|
194 |
+
|
195 |
+
[More Information Needed]
|
196 |
+
|
197 |
+
## Model Card Contact
|
198 |
+
|
199 |
+
[More Information Needed]
|
200 |
+
### Framework versions
|
201 |
+
|
202 |
+
- PEFT 0.11.1
|
mixing_strategies/Equal/bugsBunny-v1_1-Llama-3-8B-V-Equal_dataset_20000_repeat_3_epochs_1_GA_2_lora/adapter_config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "./weights/Bunny-v1_1-Llama-3-8B-V",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layer_replication": null,
|
10 |
+
"layers_pattern": null,
|
11 |
+
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
+
"lora_alpha": 256,
|
14 |
+
"lora_dropout": 0.1,
|
15 |
+
"megatron_config": null,
|
16 |
+
"megatron_core": "megatron.core",
|
17 |
+
"modules_to_save": null,
|
18 |
+
"peft_type": "LORA",
|
19 |
+
"r": 128,
|
20 |
+
"rank_pattern": {},
|
21 |
+
"revision": null,
|
22 |
+
"target_modules": [
|
23 |
+
"k_proj",
|
24 |
+
"down_proj",
|
25 |
+
"up_proj",
|
26 |
+
"v_proj",
|
27 |
+
"q_proj",
|
28 |
+
"o_proj",
|
29 |
+
"gate_proj"
|
30 |
+
],
|
31 |
+
"task_type": "CAUSAL_LM",
|
32 |
+
"use_dora": false,
|
33 |
+
"use_rslora": false
|
34 |
+
}
|