RMHalak commited on
Commit
b93f7fb
1 Parent(s): 4c613d8

Task: SequenceClassification

Browse files
README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: RMHalak/opt-6.7b-GPTQ-4bits-128g-wikitext2
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 0.10.0
adapter_config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "RMHalak/opt-6.7b-GPTQ-4bits-128g-wikitext2",
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": 32,
14
+ "lora_dropout": 0.1,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": [
18
+ "score"
19
+ ],
20
+ "peft_type": "LORA",
21
+ "r": 8,
22
+ "rank_pattern": {},
23
+ "revision": null,
24
+ "target_modules": [
25
+ "q_proj",
26
+ "fc1",
27
+ "fc2",
28
+ "v_proj",
29
+ "k_proj",
30
+ "out_proj"
31
+ ],
32
+ "task_type": "SEQ_CLS",
33
+ "use_dora": false,
34
+ "use_rslora": false
35
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67c8390209e0de19acf0615d7cd5cdf97230a08bb49ef5976b199198e7eac6d1
3
+ size 75566824
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "</s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "</s>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "1": {
6
+ "content": "<pad>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "2": {
14
+ "content": "</s>",
15
+ "lstrip": false,
16
+ "normalized": true,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ }
21
+ },
22
+ "bos_token": "</s>",
23
+ "clean_up_tokenization_spaces": true,
24
+ "eos_token": "</s>",
25
+ "errors": "replace",
26
+ "model_max_length": 1000000000000000019884624838656,
27
+ "pad_token": "</s>",
28
+ "tokenizer_class": "GPT2Tokenizer",
29
+ "unk_token": "</s>"
30
+ }
trainer_state-opt-gptq-QLORA-super_glue-multirc-sequence_classification.json ADDED
@@ -0,0 +1,2262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.984,
5
+ "eval_steps": 1,
6
+ "global_step": 124,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.016,
13
+ "grad_norm": 26.392181396484375,
14
+ "learning_rate": 2.5e-05,
15
+ "loss": 1.2683,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.016,
20
+ "eval_exact_match": 0.14285714285714285,
21
+ "eval_f1_a": 0.544,
22
+ "eval_f1_m": 0.5331374974232117,
23
+ "eval_loss": 0.9992128610610962,
24
+ "eval_runtime": 12.4039,
25
+ "eval_samples_per_second": 20.155,
26
+ "eval_steps_per_second": 2.58,
27
+ "step": 1
28
+ },
29
+ {
30
+ "epoch": 0.032,
31
+ "grad_norm": 19.146682739257812,
32
+ "learning_rate": 5e-05,
33
+ "loss": 1.0276,
34
+ "step": 2
35
+ },
36
+ {
37
+ "epoch": 0.032,
38
+ "eval_exact_match": 0.12244897959183673,
39
+ "eval_f1_a": 0.5344129554655871,
40
+ "eval_f1_m": 0.5208925994640281,
41
+ "eval_loss": 0.9813457131385803,
42
+ "eval_runtime": 12.2968,
43
+ "eval_samples_per_second": 20.331,
44
+ "eval_steps_per_second": 2.602,
45
+ "step": 2
46
+ },
47
+ {
48
+ "epoch": 0.048,
49
+ "grad_norm": 25.024534225463867,
50
+ "learning_rate": 4.959016393442623e-05,
51
+ "loss": 1.1935,
52
+ "step": 3
53
+ },
54
+ {
55
+ "epoch": 0.048,
56
+ "eval_exact_match": 0.04081632653061224,
57
+ "eval_f1_a": 0.4979253112033195,
58
+ "eval_f1_m": 0.4637011514562535,
59
+ "eval_loss": 0.9183418154716492,
60
+ "eval_runtime": 12.5058,
61
+ "eval_samples_per_second": 19.991,
62
+ "eval_steps_per_second": 2.559,
63
+ "step": 3
64
+ },
65
+ {
66
+ "epoch": 0.064,
67
+ "grad_norm": 18.022754669189453,
68
+ "learning_rate": 4.918032786885246e-05,
69
+ "loss": 0.9304,
70
+ "step": 4
71
+ },
72
+ {
73
+ "epoch": 0.064,
74
+ "eval_exact_match": 0.04081632653061224,
75
+ "eval_f1_a": 0.43243243243243246,
76
+ "eval_f1_m": 0.42084400859911064,
77
+ "eval_loss": 0.8643535375595093,
78
+ "eval_runtime": 12.5036,
79
+ "eval_samples_per_second": 19.994,
80
+ "eval_steps_per_second": 2.559,
81
+ "step": 4
82
+ },
83
+ {
84
+ "epoch": 0.08,
85
+ "grad_norm": 17.011953353881836,
86
+ "learning_rate": 4.8770491803278687e-05,
87
+ "loss": 1.0187,
88
+ "step": 5
89
+ },
90
+ {
91
+ "epoch": 0.08,
92
+ "eval_exact_match": 0.04081632653061224,
93
+ "eval_f1_a": 0.44131455399061037,
94
+ "eval_f1_m": 0.435791589363018,
95
+ "eval_loss": 0.8290175795555115,
96
+ "eval_runtime": 12.501,
97
+ "eval_samples_per_second": 19.998,
98
+ "eval_steps_per_second": 2.56,
99
+ "step": 5
100
+ },
101
+ {
102
+ "epoch": 0.096,
103
+ "grad_norm": 9.931246757507324,
104
+ "learning_rate": 4.836065573770492e-05,
105
+ "loss": 0.781,
106
+ "step": 6
107
+ },
108
+ {
109
+ "epoch": 0.096,
110
+ "eval_exact_match": 0.04081632653061224,
111
+ "eval_f1_a": 0.44761904761904764,
112
+ "eval_f1_m": 0.44164457990988604,
113
+ "eval_loss": 0.803396463394165,
114
+ "eval_runtime": 12.603,
115
+ "eval_samples_per_second": 19.837,
116
+ "eval_steps_per_second": 2.539,
117
+ "step": 6
118
+ },
119
+ {
120
+ "epoch": 0.112,
121
+ "grad_norm": 34.31171417236328,
122
+ "learning_rate": 4.795081967213115e-05,
123
+ "loss": 1.2274,
124
+ "step": 7
125
+ },
126
+ {
127
+ "epoch": 0.112,
128
+ "eval_exact_match": 0.04081632653061224,
129
+ "eval_f1_a": 0.416243654822335,
130
+ "eval_f1_m": 0.42698010606173864,
131
+ "eval_loss": 0.7892948985099792,
132
+ "eval_runtime": 12.5071,
133
+ "eval_samples_per_second": 19.989,
134
+ "eval_steps_per_second": 2.559,
135
+ "step": 7
136
+ },
137
+ {
138
+ "epoch": 0.128,
139
+ "grad_norm": 19.665807723999023,
140
+ "learning_rate": 4.754098360655738e-05,
141
+ "loss": 1.0631,
142
+ "step": 8
143
+ },
144
+ {
145
+ "epoch": 0.128,
146
+ "eval_exact_match": 0.04081632653061224,
147
+ "eval_f1_a": 0.3734939759036145,
148
+ "eval_f1_m": 0.44151661996800046,
149
+ "eval_loss": 0.779171884059906,
150
+ "eval_runtime": 12.401,
151
+ "eval_samples_per_second": 20.16,
152
+ "eval_steps_per_second": 2.58,
153
+ "step": 8
154
+ },
155
+ {
156
+ "epoch": 0.144,
157
+ "grad_norm": 6.531087875366211,
158
+ "learning_rate": 4.713114754098361e-05,
159
+ "loss": 0.6823,
160
+ "step": 9
161
+ },
162
+ {
163
+ "epoch": 0.144,
164
+ "eval_exact_match": 0.0,
165
+ "eval_f1_a": 0.24657534246575338,
166
+ "eval_f1_m": 0.38393596199718655,
167
+ "eval_loss": 0.7732793092727661,
168
+ "eval_runtime": 12.3963,
169
+ "eval_samples_per_second": 20.167,
170
+ "eval_steps_per_second": 2.581,
171
+ "step": 9
172
+ },
173
+ {
174
+ "epoch": 0.16,
175
+ "grad_norm": 9.769821166992188,
176
+ "learning_rate": 4.672131147540984e-05,
177
+ "loss": 0.6971,
178
+ "step": 10
179
+ },
180
+ {
181
+ "epoch": 0.16,
182
+ "eval_exact_match": 0.0,
183
+ "eval_f1_a": 0.24827586206896554,
184
+ "eval_f1_m": 0.38636550524305635,
185
+ "eval_loss": 0.7666972875595093,
186
+ "eval_runtime": 12.3966,
187
+ "eval_samples_per_second": 20.167,
188
+ "eval_steps_per_second": 2.581,
189
+ "step": 10
190
+ },
191
+ {
192
+ "epoch": 0.176,
193
+ "grad_norm": 18.875186920166016,
194
+ "learning_rate": 4.631147540983607e-05,
195
+ "loss": 0.8232,
196
+ "step": 11
197
+ },
198
+ {
199
+ "epoch": 0.176,
200
+ "eval_exact_match": 0.0,
201
+ "eval_f1_a": 0.2028985507246377,
202
+ "eval_f1_m": 0.37477658396025737,
203
+ "eval_loss": 0.7616797089576721,
204
+ "eval_runtime": 12.293,
205
+ "eval_samples_per_second": 20.337,
206
+ "eval_steps_per_second": 2.603,
207
+ "step": 11
208
+ },
209
+ {
210
+ "epoch": 0.192,
211
+ "grad_norm": 12.508312225341797,
212
+ "learning_rate": 4.59016393442623e-05,
213
+ "loss": 0.4358,
214
+ "step": 12
215
+ },
216
+ {
217
+ "epoch": 0.192,
218
+ "eval_exact_match": 0.0,
219
+ "eval_f1_a": 0.2142857142857143,
220
+ "eval_f1_m": 0.37786431255819003,
221
+ "eval_loss": 0.7548164129257202,
222
+ "eval_runtime": 12.2924,
223
+ "eval_samples_per_second": 20.338,
224
+ "eval_steps_per_second": 2.603,
225
+ "step": 12
226
+ },
227
+ {
228
+ "epoch": 0.208,
229
+ "grad_norm": 13.73599624633789,
230
+ "learning_rate": 4.549180327868853e-05,
231
+ "loss": 0.7875,
232
+ "step": 13
233
+ },
234
+ {
235
+ "epoch": 0.208,
236
+ "eval_exact_match": 0.0,
237
+ "eval_f1_a": 0.21582733812949637,
238
+ "eval_f1_m": 0.37958707958707955,
239
+ "eval_loss": 0.7478398680686951,
240
+ "eval_runtime": 12.293,
241
+ "eval_samples_per_second": 20.337,
242
+ "eval_steps_per_second": 2.603,
243
+ "step": 13
244
+ },
245
+ {
246
+ "epoch": 0.224,
247
+ "grad_norm": 17.087984085083008,
248
+ "learning_rate": 4.508196721311476e-05,
249
+ "loss": 0.9012,
250
+ "step": 14
251
+ },
252
+ {
253
+ "epoch": 0.224,
254
+ "eval_exact_match": 0.0,
255
+ "eval_f1_a": 0.2377622377622378,
256
+ "eval_f1_m": 0.38485918843061695,
257
+ "eval_loss": 0.7398672103881836,
258
+ "eval_runtime": 12.294,
259
+ "eval_samples_per_second": 20.335,
260
+ "eval_steps_per_second": 2.603,
261
+ "step": 14
262
+ },
263
+ {
264
+ "epoch": 0.24,
265
+ "grad_norm": 17.19772720336914,
266
+ "learning_rate": 4.467213114754098e-05,
267
+ "loss": 0.8344,
268
+ "step": 15
269
+ },
270
+ {
271
+ "epoch": 0.24,
272
+ "eval_exact_match": 0.0,
273
+ "eval_f1_a": 0.24657534246575338,
274
+ "eval_f1_m": 0.38393596199718655,
275
+ "eval_loss": 0.7301445603370667,
276
+ "eval_runtime": 12.3925,
277
+ "eval_samples_per_second": 20.173,
278
+ "eval_steps_per_second": 2.582,
279
+ "step": 15
280
+ },
281
+ {
282
+ "epoch": 0.256,
283
+ "grad_norm": 9.059144973754883,
284
+ "learning_rate": 4.426229508196721e-05,
285
+ "loss": 0.6957,
286
+ "step": 16
287
+ },
288
+ {
289
+ "epoch": 0.256,
290
+ "eval_exact_match": 0.02040816326530612,
291
+ "eval_f1_a": 0.3184713375796179,
292
+ "eval_f1_m": 0.4192099850263115,
293
+ "eval_loss": 0.7225546836853027,
294
+ "eval_runtime": 12.2951,
295
+ "eval_samples_per_second": 20.333,
296
+ "eval_steps_per_second": 2.603,
297
+ "step": 16
298
+ },
299
+ {
300
+ "epoch": 0.272,
301
+ "grad_norm": 4.891639709472656,
302
+ "learning_rate": 4.3852459016393444e-05,
303
+ "loss": 0.6085,
304
+ "step": 17
305
+ },
306
+ {
307
+ "epoch": 0.272,
308
+ "eval_exact_match": 0.061224489795918366,
309
+ "eval_f1_a": 0.3878787878787879,
310
+ "eval_f1_m": 0.4543832318142042,
311
+ "eval_loss": 0.7196054458618164,
312
+ "eval_runtime": 12.2946,
313
+ "eval_samples_per_second": 20.334,
314
+ "eval_steps_per_second": 2.603,
315
+ "step": 17
316
+ },
317
+ {
318
+ "epoch": 0.288,
319
+ "grad_norm": 11.897939682006836,
320
+ "learning_rate": 4.3442622950819674e-05,
321
+ "loss": 0.7777,
322
+ "step": 18
323
+ },
324
+ {
325
+ "epoch": 0.288,
326
+ "eval_exact_match": 0.04081632653061224,
327
+ "eval_f1_a": 0.3976608187134503,
328
+ "eval_f1_m": 0.4537870592792561,
329
+ "eval_loss": 0.7144648432731628,
330
+ "eval_runtime": 12.3977,
331
+ "eval_samples_per_second": 20.165,
332
+ "eval_steps_per_second": 2.581,
333
+ "step": 18
334
+ },
335
+ {
336
+ "epoch": 0.304,
337
+ "grad_norm": 11.019088745117188,
338
+ "learning_rate": 4.3032786885245904e-05,
339
+ "loss": 0.6732,
340
+ "step": 19
341
+ },
342
+ {
343
+ "epoch": 0.304,
344
+ "eval_exact_match": 0.04081632653061224,
345
+ "eval_f1_a": 0.39285714285714285,
346
+ "eval_f1_m": 0.4570021548412905,
347
+ "eval_loss": 0.7114218473434448,
348
+ "eval_runtime": 12.3988,
349
+ "eval_samples_per_second": 20.163,
350
+ "eval_steps_per_second": 2.581,
351
+ "step": 19
352
+ },
353
+ {
354
+ "epoch": 0.32,
355
+ "grad_norm": 9.549395561218262,
356
+ "learning_rate": 4.262295081967213e-05,
357
+ "loss": 0.7477,
358
+ "step": 20
359
+ },
360
+ {
361
+ "epoch": 0.32,
362
+ "eval_exact_match": 0.04081632653061224,
363
+ "eval_f1_a": 0.4046242774566474,
364
+ "eval_f1_m": 0.4577005477515681,
365
+ "eval_loss": 0.709179699420929,
366
+ "eval_runtime": 12.4018,
367
+ "eval_samples_per_second": 20.158,
368
+ "eval_steps_per_second": 2.58,
369
+ "step": 20
370
+ },
371
+ {
372
+ "epoch": 0.336,
373
+ "grad_norm": 6.315933704376221,
374
+ "learning_rate": 4.2213114754098365e-05,
375
+ "loss": 0.6196,
376
+ "step": 21
377
+ },
378
+ {
379
+ "epoch": 0.336,
380
+ "eval_exact_match": 0.04081632653061224,
381
+ "eval_f1_a": 0.4069767441860465,
382
+ "eval_f1_m": 0.4574949710153791,
383
+ "eval_loss": 0.7092265486717224,
384
+ "eval_runtime": 12.2969,
385
+ "eval_samples_per_second": 20.33,
386
+ "eval_steps_per_second": 2.602,
387
+ "step": 21
388
+ },
389
+ {
390
+ "epoch": 0.352,
391
+ "grad_norm": 13.465363502502441,
392
+ "learning_rate": 4.1803278688524595e-05,
393
+ "loss": 0.8407,
394
+ "step": 22
395
+ },
396
+ {
397
+ "epoch": 0.352,
398
+ "eval_exact_match": 0.04081632653061224,
399
+ "eval_f1_a": 0.4333333333333333,
400
+ "eval_f1_m": 0.4608648720893619,
401
+ "eval_loss": 0.7104062438011169,
402
+ "eval_runtime": 12.3998,
403
+ "eval_samples_per_second": 20.162,
404
+ "eval_steps_per_second": 2.581,
405
+ "step": 22
406
+ },
407
+ {
408
+ "epoch": 0.368,
409
+ "grad_norm": 5.6240763664245605,
410
+ "learning_rate": 4.1393442622950826e-05,
411
+ "loss": 0.6945,
412
+ "step": 23
413
+ },
414
+ {
415
+ "epoch": 0.368,
416
+ "eval_exact_match": 0.04081632653061224,
417
+ "eval_f1_a": 0.4175824175824176,
418
+ "eval_f1_m": 0.4523820850351463,
419
+ "eval_loss": 0.7099843621253967,
420
+ "eval_runtime": 12.5009,
421
+ "eval_samples_per_second": 19.999,
422
+ "eval_steps_per_second": 2.56,
423
+ "step": 23
424
+ },
425
+ {
426
+ "epoch": 0.384,
427
+ "grad_norm": 11.571455001831055,
428
+ "learning_rate": 4.098360655737705e-05,
429
+ "loss": 0.7534,
430
+ "step": 24
431
+ },
432
+ {
433
+ "epoch": 0.384,
434
+ "eval_exact_match": 0.04081632653061224,
435
+ "eval_f1_a": 0.42774566473988435,
436
+ "eval_f1_m": 0.4625007645415809,
437
+ "eval_loss": 0.7112500071525574,
438
+ "eval_runtime": 12.505,
439
+ "eval_samples_per_second": 19.992,
440
+ "eval_steps_per_second": 2.559,
441
+ "step": 24
442
+ },
443
+ {
444
+ "epoch": 0.4,
445
+ "grad_norm": 9.693904876708984,
446
+ "learning_rate": 4.057377049180328e-05,
447
+ "loss": 0.6428,
448
+ "step": 25
449
+ },
450
+ {
451
+ "epoch": 0.4,
452
+ "eval_exact_match": 0.02040816326530612,
453
+ "eval_f1_a": 0.32258064516129026,
454
+ "eval_f1_m": 0.42786660784860053,
455
+ "eval_loss": 0.7106679677963257,
456
+ "eval_runtime": 12.5005,
457
+ "eval_samples_per_second": 19.999,
458
+ "eval_steps_per_second": 2.56,
459
+ "step": 25
460
+ },
461
+ {
462
+ "epoch": 0.416,
463
+ "grad_norm": 5.713263034820557,
464
+ "learning_rate": 4.016393442622951e-05,
465
+ "loss": 0.649,
466
+ "step": 26
467
+ },
468
+ {
469
+ "epoch": 0.416,
470
+ "eval_exact_match": 0.0,
471
+ "eval_f1_a": 0.2638888888888889,
472
+ "eval_f1_m": 0.4046327255510929,
473
+ "eval_loss": 0.7118203043937683,
474
+ "eval_runtime": 12.4034,
475
+ "eval_samples_per_second": 20.156,
476
+ "eval_steps_per_second": 2.58,
477
+ "step": 26
478
+ },
479
+ {
480
+ "epoch": 0.432,
481
+ "grad_norm": 7.6466965675354,
482
+ "learning_rate": 3.975409836065574e-05,
483
+ "loss": 0.7344,
484
+ "step": 27
485
+ },
486
+ {
487
+ "epoch": 0.432,
488
+ "eval_exact_match": 0.0,
489
+ "eval_f1_a": 0.18045112781954886,
490
+ "eval_f1_m": 0.3727819799248371,
491
+ "eval_loss": 0.7153554558753967,
492
+ "eval_runtime": 12.4059,
493
+ "eval_samples_per_second": 20.152,
494
+ "eval_steps_per_second": 2.579,
495
+ "step": 27
496
+ },
497
+ {
498
+ "epoch": 0.448,
499
+ "grad_norm": 12.54140567779541,
500
+ "learning_rate": 3.934426229508197e-05,
501
+ "loss": 0.5762,
502
+ "step": 28
503
+ },
504
+ {
505
+ "epoch": 0.448,
506
+ "eval_exact_match": 0.0,
507
+ "eval_f1_a": 0.13846153846153844,
508
+ "eval_f1_m": 0.3607012601910561,
509
+ "eval_loss": 0.7164726853370667,
510
+ "eval_runtime": 12.4987,
511
+ "eval_samples_per_second": 20.002,
512
+ "eval_steps_per_second": 2.56,
513
+ "step": 28
514
+ },
515
+ {
516
+ "epoch": 0.464,
517
+ "grad_norm": 7.025119781494141,
518
+ "learning_rate": 3.89344262295082e-05,
519
+ "loss": 0.6606,
520
+ "step": 29
521
+ },
522
+ {
523
+ "epoch": 0.464,
524
+ "eval_exact_match": 0.0,
525
+ "eval_f1_a": 0.11111111111111112,
526
+ "eval_f1_m": 0.35377706194032715,
527
+ "eval_loss": 0.719851553440094,
528
+ "eval_runtime": 12.599,
529
+ "eval_samples_per_second": 19.843,
530
+ "eval_steps_per_second": 2.54,
531
+ "step": 29
532
+ },
533
+ {
534
+ "epoch": 0.48,
535
+ "grad_norm": 16.650728225708008,
536
+ "learning_rate": 3.8524590163934424e-05,
537
+ "loss": 0.6886,
538
+ "step": 30
539
+ },
540
+ {
541
+ "epoch": 0.48,
542
+ "eval_exact_match": 0.0,
543
+ "eval_f1_a": 0.05454545454545454,
544
+ "eval_f1_m": 0.3549857455469701,
545
+ "eval_loss": 0.7235468626022339,
546
+ "eval_runtime": 12.5049,
547
+ "eval_samples_per_second": 19.992,
548
+ "eval_steps_per_second": 2.559,
549
+ "step": 30
550
+ },
551
+ {
552
+ "epoch": 0.496,
553
+ "grad_norm": 9.604567527770996,
554
+ "learning_rate": 3.8114754098360655e-05,
555
+ "loss": 0.6693,
556
+ "step": 31
557
+ },
558
+ {
559
+ "epoch": 0.496,
560
+ "eval_exact_match": 0.0,
561
+ "eval_f1_a": 0.03773584905660377,
562
+ "eval_f1_m": 0.3570682452135034,
563
+ "eval_loss": 0.730707049369812,
564
+ "eval_runtime": 12.6051,
565
+ "eval_samples_per_second": 19.833,
566
+ "eval_steps_per_second": 2.539,
567
+ "step": 31
568
+ },
569
+ {
570
+ "epoch": 0.512,
571
+ "grad_norm": 10.662591934204102,
572
+ "learning_rate": 3.7704918032786885e-05,
573
+ "loss": 0.8203,
574
+ "step": 32
575
+ },
576
+ {
577
+ "epoch": 0.512,
578
+ "eval_exact_match": 0.0,
579
+ "eval_f1_a": 0.019230769230769232,
580
+ "eval_f1_m": 0.35394517780472173,
581
+ "eval_loss": 0.7325780987739563,
582
+ "eval_runtime": 12.5102,
583
+ "eval_samples_per_second": 19.984,
584
+ "eval_steps_per_second": 2.558,
585
+ "step": 32
586
+ },
587
+ {
588
+ "epoch": 0.528,
589
+ "grad_norm": 5.810702323913574,
590
+ "learning_rate": 3.729508196721312e-05,
591
+ "loss": 0.6831,
592
+ "step": 33
593
+ },
594
+ {
595
+ "epoch": 0.528,
596
+ "eval_exact_match": 0.0,
597
+ "eval_f1_a": 0.019230769230769232,
598
+ "eval_f1_m": 0.35394517780472173,
599
+ "eval_loss": 0.7296562790870667,
600
+ "eval_runtime": 12.5052,
601
+ "eval_samples_per_second": 19.992,
602
+ "eval_steps_per_second": 2.559,
603
+ "step": 33
604
+ },
605
+ {
606
+ "epoch": 0.544,
607
+ "grad_norm": 6.527573585510254,
608
+ "learning_rate": 3.6885245901639346e-05,
609
+ "loss": 0.6942,
610
+ "step": 34
611
+ },
612
+ {
613
+ "epoch": 0.544,
614
+ "eval_exact_match": 0.0,
615
+ "eval_f1_a": 0.019230769230769232,
616
+ "eval_f1_m": 0.35394517780472173,
617
+ "eval_loss": 0.7321445345878601,
618
+ "eval_runtime": 12.6062,
619
+ "eval_samples_per_second": 19.831,
620
+ "eval_steps_per_second": 2.538,
621
+ "step": 34
622
+ },
623
+ {
624
+ "epoch": 0.56,
625
+ "grad_norm": 15.23520565032959,
626
+ "learning_rate": 3.6475409836065576e-05,
627
+ "loss": 0.7819,
628
+ "step": 35
629
+ },
630
+ {
631
+ "epoch": 0.56,
632
+ "eval_exact_match": 0.0,
633
+ "eval_f1_a": 0.019417475728155338,
634
+ "eval_f1_m": 0.3550274288869728,
635
+ "eval_loss": 0.7335781455039978,
636
+ "eval_runtime": 12.6066,
637
+ "eval_samples_per_second": 19.831,
638
+ "eval_steps_per_second": 2.538,
639
+ "step": 35
640
+ },
641
+ {
642
+ "epoch": 0.576,
643
+ "grad_norm": 12.970118522644043,
644
+ "learning_rate": 3.6065573770491806e-05,
645
+ "loss": 0.5959,
646
+ "step": 36
647
+ },
648
+ {
649
+ "epoch": 0.576,
650
+ "eval_exact_match": 0.0,
651
+ "eval_f1_a": 0.0,
652
+ "eval_f1_m": 0.3489535707722984,
653
+ "eval_loss": 0.7367480397224426,
654
+ "eval_runtime": 12.5014,
655
+ "eval_samples_per_second": 19.998,
656
+ "eval_steps_per_second": 2.56,
657
+ "step": 36
658
+ },
659
+ {
660
+ "epoch": 0.592,
661
+ "grad_norm": 9.984509468078613,
662
+ "learning_rate": 3.5655737704918037e-05,
663
+ "loss": 0.5692,
664
+ "step": 37
665
+ },
666
+ {
667
+ "epoch": 0.592,
668
+ "eval_exact_match": 0.0,
669
+ "eval_f1_a": 0.0,
670
+ "eval_f1_m": 0.3489535707722984,
671
+ "eval_loss": 0.7453652620315552,
672
+ "eval_runtime": 12.4005,
673
+ "eval_samples_per_second": 20.16,
674
+ "eval_steps_per_second": 2.581,
675
+ "step": 37
676
+ },
677
+ {
678
+ "epoch": 0.608,
679
+ "grad_norm": 9.310087203979492,
680
+ "learning_rate": 3.524590163934427e-05,
681
+ "loss": 0.8538,
682
+ "step": 38
683
+ },
684
+ {
685
+ "epoch": 0.608,
686
+ "eval_exact_match": 0.0,
687
+ "eval_f1_a": 0.0,
688
+ "eval_f1_m": 0.35077572820670067,
689
+ "eval_loss": 0.7506445050239563,
690
+ "eval_runtime": 12.5041,
691
+ "eval_samples_per_second": 19.993,
692
+ "eval_steps_per_second": 2.559,
693
+ "step": 38
694
+ },
695
+ {
696
+ "epoch": 0.624,
697
+ "grad_norm": 17.32353401184082,
698
+ "learning_rate": 3.483606557377049e-05,
699
+ "loss": 0.7333,
700
+ "step": 39
701
+ },
702
+ {
703
+ "epoch": 0.624,
704
+ "eval_exact_match": 0.0,
705
+ "eval_f1_a": 0.0,
706
+ "eval_f1_m": 0.35077572820670067,
707
+ "eval_loss": 0.7485976815223694,
708
+ "eval_runtime": 12.605,
709
+ "eval_samples_per_second": 19.833,
710
+ "eval_steps_per_second": 2.539,
711
+ "step": 39
712
+ },
713
+ {
714
+ "epoch": 0.64,
715
+ "grad_norm": 9.058735847473145,
716
+ "learning_rate": 3.442622950819672e-05,
717
+ "loss": 0.4784,
718
+ "step": 40
719
+ },
720
+ {
721
+ "epoch": 0.64,
722
+ "eval_exact_match": 0.0,
723
+ "eval_f1_a": 0.0,
724
+ "eval_f1_m": 0.35077572820670067,
725
+ "eval_loss": 0.7543515563011169,
726
+ "eval_runtime": 12.4985,
727
+ "eval_samples_per_second": 20.002,
728
+ "eval_steps_per_second": 2.56,
729
+ "step": 40
730
+ },
731
+ {
732
+ "epoch": 0.656,
733
+ "grad_norm": 15.80337142944336,
734
+ "learning_rate": 3.401639344262295e-05,
735
+ "loss": 0.8348,
736
+ "step": 41
737
+ },
738
+ {
739
+ "epoch": 0.656,
740
+ "eval_exact_match": 0.0,
741
+ "eval_f1_a": 0.0,
742
+ "eval_f1_m": 0.35077572820670067,
743
+ "eval_loss": 0.7519648671150208,
744
+ "eval_runtime": 12.4002,
745
+ "eval_samples_per_second": 20.161,
746
+ "eval_steps_per_second": 2.581,
747
+ "step": 41
748
+ },
749
+ {
750
+ "epoch": 0.672,
751
+ "grad_norm": 18.595117568969727,
752
+ "learning_rate": 3.360655737704918e-05,
753
+ "loss": 0.8212,
754
+ "step": 42
755
+ },
756
+ {
757
+ "epoch": 0.672,
758
+ "eval_exact_match": 0.0,
759
+ "eval_f1_a": 0.0,
760
+ "eval_f1_m": 0.35077572820670067,
761
+ "eval_loss": 0.7456523180007935,
762
+ "eval_runtime": 12.5047,
763
+ "eval_samples_per_second": 19.992,
764
+ "eval_steps_per_second": 2.559,
765
+ "step": 42
766
+ },
767
+ {
768
+ "epoch": 0.688,
769
+ "grad_norm": 9.790151596069336,
770
+ "learning_rate": 3.319672131147541e-05,
771
+ "loss": 0.9515,
772
+ "step": 43
773
+ },
774
+ {
775
+ "epoch": 0.688,
776
+ "eval_exact_match": 0.0,
777
+ "eval_f1_a": 0.0,
778
+ "eval_f1_m": 0.3489535707722984,
779
+ "eval_loss": 0.7356074452400208,
780
+ "eval_runtime": 12.5029,
781
+ "eval_samples_per_second": 19.995,
782
+ "eval_steps_per_second": 2.559,
783
+ "step": 43
784
+ },
785
+ {
786
+ "epoch": 0.704,
787
+ "grad_norm": 19.172956466674805,
788
+ "learning_rate": 3.2786885245901635e-05,
789
+ "loss": 0.8488,
790
+ "step": 44
791
+ },
792
+ {
793
+ "epoch": 0.704,
794
+ "eval_exact_match": 0.0,
795
+ "eval_f1_a": 0.019230769230769232,
796
+ "eval_f1_m": 0.35394517780472173,
797
+ "eval_loss": 0.7283183336257935,
798
+ "eval_runtime": 12.6008,
799
+ "eval_samples_per_second": 19.84,
800
+ "eval_steps_per_second": 2.54,
801
+ "step": 44
802
+ },
803
+ {
804
+ "epoch": 0.72,
805
+ "grad_norm": 10.353903770446777,
806
+ "learning_rate": 3.237704918032787e-05,
807
+ "loss": 0.6007,
808
+ "step": 45
809
+ },
810
+ {
811
+ "epoch": 0.72,
812
+ "eval_exact_match": 0.0,
813
+ "eval_f1_a": 0.019230769230769232,
814
+ "eval_f1_m": 0.35394517780472173,
815
+ "eval_loss": 0.7258906364440918,
816
+ "eval_runtime": 12.4066,
817
+ "eval_samples_per_second": 20.15,
818
+ "eval_steps_per_second": 2.579,
819
+ "step": 45
820
+ },
821
+ {
822
+ "epoch": 0.736,
823
+ "grad_norm": 10.669143676757812,
824
+ "learning_rate": 3.19672131147541e-05,
825
+ "loss": 0.8167,
826
+ "step": 46
827
+ },
828
+ {
829
+ "epoch": 0.736,
830
+ "eval_exact_match": 0.0,
831
+ "eval_f1_a": 0.019230769230769232,
832
+ "eval_f1_m": 0.35394517780472173,
833
+ "eval_loss": 0.7255585789680481,
834
+ "eval_runtime": 12.6073,
835
+ "eval_samples_per_second": 19.83,
836
+ "eval_steps_per_second": 2.538,
837
+ "step": 46
838
+ },
839
+ {
840
+ "epoch": 0.752,
841
+ "grad_norm": 12.651082992553711,
842
+ "learning_rate": 3.155737704918033e-05,
843
+ "loss": 0.7909,
844
+ "step": 47
845
+ },
846
+ {
847
+ "epoch": 0.752,
848
+ "eval_exact_match": 0.0,
849
+ "eval_f1_a": 0.019047619047619046,
850
+ "eval_f1_m": 0.35264647650602043,
851
+ "eval_loss": 0.7224531173706055,
852
+ "eval_runtime": 12.5046,
853
+ "eval_samples_per_second": 19.993,
854
+ "eval_steps_per_second": 2.559,
855
+ "step": 47
856
+ },
857
+ {
858
+ "epoch": 0.768,
859
+ "grad_norm": 16.77728843688965,
860
+ "learning_rate": 3.114754098360656e-05,
861
+ "loss": 0.7277,
862
+ "step": 48
863
+ },
864
+ {
865
+ "epoch": 0.768,
866
+ "eval_exact_match": 0.0,
867
+ "eval_f1_a": 0.037383177570093455,
868
+ "eval_f1_m": 0.3553675649413946,
869
+ "eval_loss": 0.7222617268562317,
870
+ "eval_runtime": 12.5048,
871
+ "eval_samples_per_second": 19.992,
872
+ "eval_steps_per_second": 2.559,
873
+ "step": 48
874
+ },
875
+ {
876
+ "epoch": 0.784,
877
+ "grad_norm": 9.990218162536621,
878
+ "learning_rate": 3.073770491803279e-05,
879
+ "loss": 0.6484,
880
+ "step": 49
881
+ },
882
+ {
883
+ "epoch": 0.784,
884
+ "eval_exact_match": 0.0,
885
+ "eval_f1_a": 0.018867924528301886,
886
+ "eval_f1_m": 0.35094579623391153,
887
+ "eval_loss": 0.7233593463897705,
888
+ "eval_runtime": 12.3979,
889
+ "eval_samples_per_second": 20.165,
890
+ "eval_steps_per_second": 2.581,
891
+ "step": 49
892
+ },
893
+ {
894
+ "epoch": 0.8,
895
+ "grad_norm": 19.805315017700195,
896
+ "learning_rate": 3.0327868852459017e-05,
897
+ "loss": 0.7366,
898
+ "step": 50
899
+ },
900
+ {
901
+ "epoch": 0.8,
902
+ "eval_exact_match": 0.0,
903
+ "eval_f1_a": 0.05454545454545454,
904
+ "eval_f1_m": 0.35851045213290117,
905
+ "eval_loss": 0.7192890644073486,
906
+ "eval_runtime": 12.394,
907
+ "eval_samples_per_second": 20.171,
908
+ "eval_steps_per_second": 2.582,
909
+ "step": 50
910
+ },
911
+ {
912
+ "epoch": 0.816,
913
+ "grad_norm": 6.260676860809326,
914
+ "learning_rate": 2.9918032786885248e-05,
915
+ "loss": 0.6544,
916
+ "step": 51
917
+ },
918
+ {
919
+ "epoch": 0.816,
920
+ "eval_exact_match": 0.0,
921
+ "eval_f1_a": 0.07207207207207207,
922
+ "eval_f1_m": 0.3631490108530924,
923
+ "eval_loss": 0.7178593873977661,
924
+ "eval_runtime": 12.399,
925
+ "eval_samples_per_second": 20.163,
926
+ "eval_steps_per_second": 2.581,
927
+ "step": 51
928
+ },
929
+ {
930
+ "epoch": 0.832,
931
+ "grad_norm": 6.211192607879639,
932
+ "learning_rate": 2.9508196721311478e-05,
933
+ "loss": 0.5797,
934
+ "step": 52
935
+ },
936
+ {
937
+ "epoch": 0.832,
938
+ "eval_exact_match": 0.0,
939
+ "eval_f1_a": 0.07142857142857144,
940
+ "eval_f1_m": 0.36225194873154054,
941
+ "eval_loss": 0.7176406383514404,
942
+ "eval_runtime": 12.2942,
943
+ "eval_samples_per_second": 20.335,
944
+ "eval_steps_per_second": 2.603,
945
+ "step": 52
946
+ },
947
+ {
948
+ "epoch": 0.848,
949
+ "grad_norm": 19.603347778320312,
950
+ "learning_rate": 2.9098360655737705e-05,
951
+ "loss": 0.613,
952
+ "step": 53
953
+ },
954
+ {
955
+ "epoch": 0.848,
956
+ "eval_exact_match": 0.0,
957
+ "eval_f1_a": 0.08928571428571429,
958
+ "eval_f1_m": 0.36680547343812653,
959
+ "eval_loss": 0.719406247138977,
960
+ "eval_runtime": 12.2965,
961
+ "eval_samples_per_second": 20.331,
962
+ "eval_steps_per_second": 2.602,
963
+ "step": 53
964
+ },
965
+ {
966
+ "epoch": 0.864,
967
+ "grad_norm": 7.0200042724609375,
968
+ "learning_rate": 2.8688524590163935e-05,
969
+ "loss": 0.7246,
970
+ "step": 54
971
+ },
972
+ {
973
+ "epoch": 0.864,
974
+ "eval_exact_match": 0.0,
975
+ "eval_f1_a": 0.03773584905660377,
976
+ "eval_f1_m": 0.35666626624009584,
977
+ "eval_loss": 0.7230820059776306,
978
+ "eval_runtime": 12.3942,
979
+ "eval_samples_per_second": 20.171,
980
+ "eval_steps_per_second": 2.582,
981
+ "step": 54
982
+ },
983
+ {
984
+ "epoch": 0.88,
985
+ "grad_norm": 7.6928791999816895,
986
+ "learning_rate": 2.8278688524590162e-05,
987
+ "loss": 0.723,
988
+ "step": 55
989
+ },
990
+ {
991
+ "epoch": 0.88,
992
+ "eval_exact_match": 0.0,
993
+ "eval_f1_a": 0.019230769230769232,
994
+ "eval_f1_m": 0.35394517780472173,
995
+ "eval_loss": 0.7283515334129333,
996
+ "eval_runtime": 12.4051,
997
+ "eval_samples_per_second": 20.153,
998
+ "eval_steps_per_second": 2.58,
999
+ "step": 55
1000
+ },
1001
+ {
1002
+ "epoch": 0.896,
1003
+ "grad_norm": 9.09825611114502,
1004
+ "learning_rate": 2.7868852459016392e-05,
1005
+ "loss": 0.526,
1006
+ "step": 56
1007
+ },
1008
+ {
1009
+ "epoch": 0.896,
1010
+ "eval_exact_match": 0.0,
1011
+ "eval_f1_a": 0.019230769230769232,
1012
+ "eval_f1_m": 0.35394517780472173,
1013
+ "eval_loss": 0.736648440361023,
1014
+ "eval_runtime": 12.5063,
1015
+ "eval_samples_per_second": 19.99,
1016
+ "eval_steps_per_second": 2.559,
1017
+ "step": 56
1018
+ },
1019
+ {
1020
+ "epoch": 0.912,
1021
+ "grad_norm": 12.663421630859375,
1022
+ "learning_rate": 2.7459016393442626e-05,
1023
+ "loss": 0.7038,
1024
+ "step": 57
1025
+ },
1026
+ {
1027
+ "epoch": 0.912,
1028
+ "eval_exact_match": 0.0,
1029
+ "eval_f1_a": 0.0,
1030
+ "eval_f1_m": 0.3489535707722984,
1031
+ "eval_loss": 0.7481816411018372,
1032
+ "eval_runtime": 12.4036,
1033
+ "eval_samples_per_second": 20.155,
1034
+ "eval_steps_per_second": 2.58,
1035
+ "step": 57
1036
+ },
1037
+ {
1038
+ "epoch": 0.928,
1039
+ "grad_norm": 15.922131538391113,
1040
+ "learning_rate": 2.7049180327868856e-05,
1041
+ "loss": 0.7805,
1042
+ "step": 58
1043
+ },
1044
+ {
1045
+ "epoch": 0.928,
1046
+ "eval_exact_match": 0.0,
1047
+ "eval_f1_a": 0.0,
1048
+ "eval_f1_m": 0.35077572820670067,
1049
+ "eval_loss": 0.7556718587875366,
1050
+ "eval_runtime": 12.4053,
1051
+ "eval_samples_per_second": 20.153,
1052
+ "eval_steps_per_second": 2.58,
1053
+ "step": 58
1054
+ },
1055
+ {
1056
+ "epoch": 0.944,
1057
+ "grad_norm": 15.116573333740234,
1058
+ "learning_rate": 2.6639344262295087e-05,
1059
+ "loss": 0.7645,
1060
+ "step": 59
1061
+ },
1062
+ {
1063
+ "epoch": 0.944,
1064
+ "eval_exact_match": 0.0,
1065
+ "eval_f1_a": 0.0,
1066
+ "eval_f1_m": 0.35077572820670067,
1067
+ "eval_loss": 0.7615898251533508,
1068
+ "eval_runtime": 12.301,
1069
+ "eval_samples_per_second": 20.324,
1070
+ "eval_steps_per_second": 2.601,
1071
+ "step": 59
1072
+ },
1073
+ {
1074
+ "epoch": 0.96,
1075
+ "grad_norm": 16.11196517944336,
1076
+ "learning_rate": 2.6229508196721314e-05,
1077
+ "loss": 0.802,
1078
+ "step": 60
1079
+ },
1080
+ {
1081
+ "epoch": 0.96,
1082
+ "eval_exact_match": 0.0,
1083
+ "eval_f1_a": 0.0,
1084
+ "eval_f1_m": 0.35077572820670067,
1085
+ "eval_loss": 0.762087881565094,
1086
+ "eval_runtime": 12.4045,
1087
+ "eval_samples_per_second": 20.154,
1088
+ "eval_steps_per_second": 2.58,
1089
+ "step": 60
1090
+ },
1091
+ {
1092
+ "epoch": 0.976,
1093
+ "grad_norm": 11.784616470336914,
1094
+ "learning_rate": 2.5819672131147544e-05,
1095
+ "loss": 0.7266,
1096
+ "step": 61
1097
+ },
1098
+ {
1099
+ "epoch": 0.976,
1100
+ "eval_exact_match": 0.0,
1101
+ "eval_f1_a": 0.0,
1102
+ "eval_f1_m": 0.35077572820670067,
1103
+ "eval_loss": 0.7605761885643005,
1104
+ "eval_runtime": 12.5067,
1105
+ "eval_samples_per_second": 19.989,
1106
+ "eval_steps_per_second": 2.559,
1107
+ "step": 61
1108
+ },
1109
+ {
1110
+ "epoch": 0.992,
1111
+ "grad_norm": 5.855625152587891,
1112
+ "learning_rate": 2.540983606557377e-05,
1113
+ "loss": 0.6895,
1114
+ "step": 62
1115
+ },
1116
+ {
1117
+ "epoch": 0.992,
1118
+ "eval_exact_match": 0.0,
1119
+ "eval_f1_a": 0.0,
1120
+ "eval_f1_m": 0.35077572820670067,
1121
+ "eval_loss": 0.7616230249404907,
1122
+ "eval_runtime": 12.4021,
1123
+ "eval_samples_per_second": 20.158,
1124
+ "eval_steps_per_second": 2.58,
1125
+ "step": 62
1126
+ },
1127
+ {
1128
+ "epoch": 1.008,
1129
+ "grad_norm": 12.561187744140625,
1130
+ "learning_rate": 2.5e-05,
1131
+ "loss": 0.8057,
1132
+ "step": 63
1133
+ },
1134
+ {
1135
+ "epoch": 1.008,
1136
+ "eval_exact_match": 0.0,
1137
+ "eval_f1_a": 0.0,
1138
+ "eval_f1_m": 0.35077572820670067,
1139
+ "eval_loss": 0.7573671936988831,
1140
+ "eval_runtime": 12.5096,
1141
+ "eval_samples_per_second": 19.985,
1142
+ "eval_steps_per_second": 2.558,
1143
+ "step": 63
1144
+ },
1145
+ {
1146
+ "epoch": 1.024,
1147
+ "grad_norm": 10.385027885437012,
1148
+ "learning_rate": 2.459016393442623e-05,
1149
+ "loss": 0.7454,
1150
+ "step": 64
1151
+ },
1152
+ {
1153
+ "epoch": 1.024,
1154
+ "eval_exact_match": 0.0,
1155
+ "eval_f1_a": 0.0,
1156
+ "eval_f1_m": 0.3489535707722984,
1157
+ "eval_loss": 0.7491288781166077,
1158
+ "eval_runtime": 12.4029,
1159
+ "eval_samples_per_second": 20.157,
1160
+ "eval_steps_per_second": 2.58,
1161
+ "step": 64
1162
+ },
1163
+ {
1164
+ "epoch": 1.04,
1165
+ "grad_norm": 15.67496395111084,
1166
+ "learning_rate": 2.418032786885246e-05,
1167
+ "loss": 0.877,
1168
+ "step": 65
1169
+ },
1170
+ {
1171
+ "epoch": 1.04,
1172
+ "eval_exact_match": 0.0,
1173
+ "eval_f1_a": 0.019417475728155338,
1174
+ "eval_f1_m": 0.3550274288869728,
1175
+ "eval_loss": 0.7429726719856262,
1176
+ "eval_runtime": 12.4019,
1177
+ "eval_samples_per_second": 20.158,
1178
+ "eval_steps_per_second": 2.58,
1179
+ "step": 65
1180
+ },
1181
+ {
1182
+ "epoch": 1.056,
1183
+ "grad_norm": 14.67371654510498,
1184
+ "learning_rate": 2.377049180327869e-05,
1185
+ "loss": 0.7684,
1186
+ "step": 66
1187
+ },
1188
+ {
1189
+ "epoch": 1.056,
1190
+ "eval_exact_match": 0.0,
1191
+ "eval_f1_a": 0.019230769230769232,
1192
+ "eval_f1_m": 0.35394517780472173,
1193
+ "eval_loss": 0.7341777086257935,
1194
+ "eval_runtime": 12.3064,
1195
+ "eval_samples_per_second": 20.315,
1196
+ "eval_steps_per_second": 2.6,
1197
+ "step": 66
1198
+ },
1199
+ {
1200
+ "epoch": 1.072,
1201
+ "grad_norm": 11.17091178894043,
1202
+ "learning_rate": 2.336065573770492e-05,
1203
+ "loss": 0.6712,
1204
+ "step": 67
1205
+ },
1206
+ {
1207
+ "epoch": 1.072,
1208
+ "eval_exact_match": 0.0,
1209
+ "eval_f1_a": 0.03669724770642201,
1210
+ "eval_f1_m": 0.34935849464660995,
1211
+ "eval_loss": 0.724734365940094,
1212
+ "eval_runtime": 12.4001,
1213
+ "eval_samples_per_second": 20.161,
1214
+ "eval_steps_per_second": 2.581,
1215
+ "step": 67
1216
+ },
1217
+ {
1218
+ "epoch": 1.088,
1219
+ "grad_norm": 7.933443546295166,
1220
+ "learning_rate": 2.295081967213115e-05,
1221
+ "loss": 0.545,
1222
+ "step": 68
1223
+ },
1224
+ {
1225
+ "epoch": 1.088,
1226
+ "eval_exact_match": 0.0,
1227
+ "eval_f1_a": 0.1,
1228
+ "eval_f1_m": 0.35936518130395684,
1229
+ "eval_loss": 0.7180312275886536,
1230
+ "eval_runtime": 12.6064,
1231
+ "eval_samples_per_second": 19.831,
1232
+ "eval_steps_per_second": 2.538,
1233
+ "step": 68
1234
+ },
1235
+ {
1236
+ "epoch": 1.104,
1237
+ "grad_norm": 6.20879602432251,
1238
+ "learning_rate": 2.254098360655738e-05,
1239
+ "loss": 0.5563,
1240
+ "step": 69
1241
+ },
1242
+ {
1243
+ "epoch": 1.104,
1244
+ "eval_exact_match": 0.0,
1245
+ "eval_f1_a": 0.15503875968992248,
1246
+ "eval_f1_m": 0.3728975672853224,
1247
+ "eval_loss": 0.7145351767539978,
1248
+ "eval_runtime": 12.3015,
1249
+ "eval_samples_per_second": 20.323,
1250
+ "eval_steps_per_second": 2.601,
1251
+ "step": 69
1252
+ },
1253
+ {
1254
+ "epoch": 1.12,
1255
+ "grad_norm": 5.216189861297607,
1256
+ "learning_rate": 2.2131147540983607e-05,
1257
+ "loss": 0.5797,
1258
+ "step": 70
1259
+ },
1260
+ {
1261
+ "epoch": 1.12,
1262
+ "eval_exact_match": 0.0,
1263
+ "eval_f1_a": 0.2222222222222222,
1264
+ "eval_f1_m": 0.3975035168912719,
1265
+ "eval_loss": 0.7122148275375366,
1266
+ "eval_runtime": 12.2962,
1267
+ "eval_samples_per_second": 20.331,
1268
+ "eval_steps_per_second": 2.602,
1269
+ "step": 70
1270
+ },
1271
+ {
1272
+ "epoch": 1.1360000000000001,
1273
+ "grad_norm": 4.51043701171875,
1274
+ "learning_rate": 2.1721311475409837e-05,
1275
+ "loss": 0.7785,
1276
+ "step": 71
1277
+ },
1278
+ {
1279
+ "epoch": 1.1360000000000001,
1280
+ "eval_exact_match": 0.0,
1281
+ "eval_f1_a": 0.2571428571428571,
1282
+ "eval_f1_m": 0.4030671709243137,
1283
+ "eval_loss": 0.7087773680686951,
1284
+ "eval_runtime": 12.2926,
1285
+ "eval_samples_per_second": 20.337,
1286
+ "eval_steps_per_second": 2.603,
1287
+ "step": 71
1288
+ },
1289
+ {
1290
+ "epoch": 1.152,
1291
+ "grad_norm": 9.325611114501953,
1292
+ "learning_rate": 2.1311475409836064e-05,
1293
+ "loss": 0.757,
1294
+ "step": 72
1295
+ },
1296
+ {
1297
+ "epoch": 1.152,
1298
+ "eval_exact_match": 0.0,
1299
+ "eval_f1_a": 0.3013698630136986,
1300
+ "eval_f1_m": 0.4158932903830863,
1301
+ "eval_loss": 0.7072968482971191,
1302
+ "eval_runtime": 12.3042,
1303
+ "eval_samples_per_second": 20.318,
1304
+ "eval_steps_per_second": 2.601,
1305
+ "step": 72
1306
+ },
1307
+ {
1308
+ "epoch": 1.168,
1309
+ "grad_norm": 15.549028396606445,
1310
+ "learning_rate": 2.0901639344262298e-05,
1311
+ "loss": 0.85,
1312
+ "step": 73
1313
+ },
1314
+ {
1315
+ "epoch": 1.168,
1316
+ "eval_exact_match": 0.0,
1317
+ "eval_f1_a": 0.3013698630136986,
1318
+ "eval_f1_m": 0.4126730412444698,
1319
+ "eval_loss": 0.7068906426429749,
1320
+ "eval_runtime": 12.3976,
1321
+ "eval_samples_per_second": 20.165,
1322
+ "eval_steps_per_second": 2.581,
1323
+ "step": 73
1324
+ },
1325
+ {
1326
+ "epoch": 1.184,
1327
+ "grad_norm": 7.413804531097412,
1328
+ "learning_rate": 2.0491803278688525e-05,
1329
+ "loss": 0.6694,
1330
+ "step": 74
1331
+ },
1332
+ {
1333
+ "epoch": 1.184,
1334
+ "eval_exact_match": 0.02040816326530612,
1335
+ "eval_f1_a": 0.32258064516129026,
1336
+ "eval_f1_m": 0.41783029895274804,
1337
+ "eval_loss": 0.7051249742507935,
1338
+ "eval_runtime": 12.3988,
1339
+ "eval_samples_per_second": 20.163,
1340
+ "eval_steps_per_second": 2.581,
1341
+ "step": 74
1342
+ },
1343
+ {
1344
+ "epoch": 1.2,
1345
+ "grad_norm": 13.059216499328613,
1346
+ "learning_rate": 2.0081967213114755e-05,
1347
+ "loss": 0.7654,
1348
+ "step": 75
1349
+ },
1350
+ {
1351
+ "epoch": 1.2,
1352
+ "eval_exact_match": 0.02040816326530612,
1353
+ "eval_f1_a": 0.3658536585365854,
1354
+ "eval_f1_m": 0.4367546285913633,
1355
+ "eval_loss": 0.7052109241485596,
1356
+ "eval_runtime": 12.3021,
1357
+ "eval_samples_per_second": 20.322,
1358
+ "eval_steps_per_second": 2.601,
1359
+ "step": 75
1360
+ },
1361
+ {
1362
+ "epoch": 1.216,
1363
+ "grad_norm": 18.173431396484375,
1364
+ "learning_rate": 1.9672131147540985e-05,
1365
+ "loss": 0.7874,
1366
+ "step": 76
1367
+ },
1368
+ {
1369
+ "epoch": 1.216,
1370
+ "eval_exact_match": 0.02040816326530612,
1371
+ "eval_f1_a": 0.393063583815029,
1372
+ "eval_f1_m": 0.4441113308460247,
1373
+ "eval_loss": 0.7044414281845093,
1374
+ "eval_runtime": 12.3005,
1375
+ "eval_samples_per_second": 20.324,
1376
+ "eval_steps_per_second": 2.602,
1377
+ "step": 76
1378
+ },
1379
+ {
1380
+ "epoch": 1.232,
1381
+ "grad_norm": 5.066444396972656,
1382
+ "learning_rate": 1.9262295081967212e-05,
1383
+ "loss": 0.585,
1384
+ "step": 77
1385
+ },
1386
+ {
1387
+ "epoch": 1.232,
1388
+ "eval_exact_match": 0.02040816326530612,
1389
+ "eval_f1_a": 0.4067796610169491,
1390
+ "eval_f1_m": 0.44523701921661113,
1391
+ "eval_loss": 0.7045234441757202,
1392
+ "eval_runtime": 12.4024,
1393
+ "eval_samples_per_second": 20.157,
1394
+ "eval_steps_per_second": 2.58,
1395
+ "step": 77
1396
+ },
1397
+ {
1398
+ "epoch": 1.248,
1399
+ "grad_norm": 15.81064224243164,
1400
+ "learning_rate": 1.8852459016393442e-05,
1401
+ "loss": 0.7125,
1402
+ "step": 78
1403
+ },
1404
+ {
1405
+ "epoch": 1.248,
1406
+ "eval_exact_match": 0.02040816326530612,
1407
+ "eval_f1_a": 0.4347826086956522,
1408
+ "eval_f1_m": 0.45324236421175196,
1409
+ "eval_loss": 0.7049570083618164,
1410
+ "eval_runtime": 12.2997,
1411
+ "eval_samples_per_second": 20.326,
1412
+ "eval_steps_per_second": 2.602,
1413
+ "step": 78
1414
+ },
1415
+ {
1416
+ "epoch": 1.264,
1417
+ "grad_norm": 15.786576271057129,
1418
+ "learning_rate": 1.8442622950819673e-05,
1419
+ "loss": 0.674,
1420
+ "step": 79
1421
+ },
1422
+ {
1423
+ "epoch": 1.264,
1424
+ "eval_exact_match": 0.04081632653061224,
1425
+ "eval_f1_a": 0.45263157894736844,
1426
+ "eval_f1_m": 0.45977191742497864,
1427
+ "eval_loss": 0.7056093811988831,
1428
+ "eval_runtime": 12.297,
1429
+ "eval_samples_per_second": 20.33,
1430
+ "eval_steps_per_second": 2.602,
1431
+ "step": 79
1432
+ },
1433
+ {
1434
+ "epoch": 1.28,
1435
+ "grad_norm": 20.205602645874023,
1436
+ "learning_rate": 1.8032786885245903e-05,
1437
+ "loss": 0.7886,
1438
+ "step": 80
1439
+ },
1440
+ {
1441
+ "epoch": 1.28,
1442
+ "eval_exact_match": 0.04081632653061224,
1443
+ "eval_f1_a": 0.450261780104712,
1444
+ "eval_f1_m": 0.45734237417910883,
1445
+ "eval_loss": 0.706125020980835,
1446
+ "eval_runtime": 12.2957,
1447
+ "eval_samples_per_second": 20.332,
1448
+ "eval_steps_per_second": 2.603,
1449
+ "step": 80
1450
+ },
1451
+ {
1452
+ "epoch": 1.296,
1453
+ "grad_norm": 7.352344036102295,
1454
+ "learning_rate": 1.7622950819672133e-05,
1455
+ "loss": 0.521,
1456
+ "step": 81
1457
+ },
1458
+ {
1459
+ "epoch": 1.296,
1460
+ "eval_exact_match": 0.04081632653061224,
1461
+ "eval_f1_a": 0.44329896907216493,
1462
+ "eval_f1_m": 0.44956970467174556,
1463
+ "eval_loss": 0.7077500224113464,
1464
+ "eval_runtime": 12.4027,
1465
+ "eval_samples_per_second": 20.157,
1466
+ "eval_steps_per_second": 2.58,
1467
+ "step": 81
1468
+ },
1469
+ {
1470
+ "epoch": 1.312,
1471
+ "grad_norm": 9.654190063476562,
1472
+ "learning_rate": 1.721311475409836e-05,
1473
+ "loss": 0.7754,
1474
+ "step": 82
1475
+ },
1476
+ {
1477
+ "epoch": 1.312,
1478
+ "eval_exact_match": 0.04081632653061224,
1479
+ "eval_f1_a": 0.45226130653266333,
1480
+ "eval_f1_m": 0.45042921930677027,
1481
+ "eval_loss": 0.7080312371253967,
1482
+ "eval_runtime": 12.2984,
1483
+ "eval_samples_per_second": 20.328,
1484
+ "eval_steps_per_second": 2.602,
1485
+ "step": 82
1486
+ },
1487
+ {
1488
+ "epoch": 1.328,
1489
+ "grad_norm": 19.156660079956055,
1490
+ "learning_rate": 1.680327868852459e-05,
1491
+ "loss": 0.7493,
1492
+ "step": 83
1493
+ },
1494
+ {
1495
+ "epoch": 1.328,
1496
+ "eval_exact_match": 0.04081632653061224,
1497
+ "eval_f1_a": 0.4577114427860697,
1498
+ "eval_f1_m": 0.4507745089377742,
1499
+ "eval_loss": 0.7095780968666077,
1500
+ "eval_runtime": 12.2981,
1501
+ "eval_samples_per_second": 20.328,
1502
+ "eval_steps_per_second": 2.602,
1503
+ "step": 83
1504
+ },
1505
+ {
1506
+ "epoch": 1.3439999999999999,
1507
+ "grad_norm": 4.014058589935303,
1508
+ "learning_rate": 1.6393442622950818e-05,
1509
+ "loss": 0.7076,
1510
+ "step": 84
1511
+ },
1512
+ {
1513
+ "epoch": 1.3439999999999999,
1514
+ "eval_exact_match": 0.04081632653061224,
1515
+ "eval_f1_a": 0.45999999999999996,
1516
+ "eval_f1_m": 0.45553641369967895,
1517
+ "eval_loss": 0.7096328139305115,
1518
+ "eval_runtime": 12.298,
1519
+ "eval_samples_per_second": 20.329,
1520
+ "eval_steps_per_second": 2.602,
1521
+ "step": 84
1522
+ },
1523
+ {
1524
+ "epoch": 1.3599999999999999,
1525
+ "grad_norm": 12.296324729919434,
1526
+ "learning_rate": 1.598360655737705e-05,
1527
+ "loss": 0.6649,
1528
+ "step": 85
1529
+ },
1530
+ {
1531
+ "epoch": 1.3599999999999999,
1532
+ "eval_exact_match": 0.04081632653061224,
1533
+ "eval_f1_a": 0.4577114427860697,
1534
+ "eval_f1_m": 0.4507745089377742,
1535
+ "eval_loss": 0.7094843983650208,
1536
+ "eval_runtime": 12.2932,
1537
+ "eval_samples_per_second": 20.337,
1538
+ "eval_steps_per_second": 2.603,
1539
+ "step": 85
1540
+ },
1541
+ {
1542
+ "epoch": 1.376,
1543
+ "grad_norm": 8.357748985290527,
1544
+ "learning_rate": 1.557377049180328e-05,
1545
+ "loss": 0.6199,
1546
+ "step": 86
1547
+ },
1548
+ {
1549
+ "epoch": 1.376,
1550
+ "eval_exact_match": 0.04081632653061224,
1551
+ "eval_f1_a": 0.45544554455445546,
1552
+ "eval_f1_m": 0.44890376063845455,
1553
+ "eval_loss": 0.7101484537124634,
1554
+ "eval_runtime": 12.2925,
1555
+ "eval_samples_per_second": 20.338,
1556
+ "eval_steps_per_second": 2.603,
1557
+ "step": 86
1558
+ },
1559
+ {
1560
+ "epoch": 1.392,
1561
+ "grad_norm": 5.569490909576416,
1562
+ "learning_rate": 1.5163934426229509e-05,
1563
+ "loss": 0.7666,
1564
+ "step": 87
1565
+ },
1566
+ {
1567
+ "epoch": 1.392,
1568
+ "eval_exact_match": 0.04081632653061224,
1569
+ "eval_f1_a": 0.4577114427860697,
1570
+ "eval_f1_m": 0.4507745089377742,
1571
+ "eval_loss": 0.7110859155654907,
1572
+ "eval_runtime": 12.3978,
1573
+ "eval_samples_per_second": 20.165,
1574
+ "eval_steps_per_second": 2.581,
1575
+ "step": 87
1576
+ },
1577
+ {
1578
+ "epoch": 1.408,
1579
+ "grad_norm": 20.458703994750977,
1580
+ "learning_rate": 1.4754098360655739e-05,
1581
+ "loss": 0.7505,
1582
+ "step": 88
1583
+ },
1584
+ {
1585
+ "epoch": 1.408,
1586
+ "eval_exact_match": 0.04081632653061224,
1587
+ "eval_f1_a": 0.4577114427860697,
1588
+ "eval_f1_m": 0.4507745089377742,
1589
+ "eval_loss": 0.7112030982971191,
1590
+ "eval_runtime": 12.4088,
1591
+ "eval_samples_per_second": 20.147,
1592
+ "eval_steps_per_second": 2.579,
1593
+ "step": 88
1594
+ },
1595
+ {
1596
+ "epoch": 1.424,
1597
+ "grad_norm": 17.410215377807617,
1598
+ "learning_rate": 1.4344262295081968e-05,
1599
+ "loss": 0.7426,
1600
+ "step": 89
1601
+ },
1602
+ {
1603
+ "epoch": 1.424,
1604
+ "eval_exact_match": 0.04081632653061224,
1605
+ "eval_f1_a": 0.45,
1606
+ "eval_f1_m": 0.4473113054745707,
1607
+ "eval_loss": 0.7114999890327454,
1608
+ "eval_runtime": 12.4068,
1609
+ "eval_samples_per_second": 20.15,
1610
+ "eval_steps_per_second": 2.579,
1611
+ "step": 89
1612
+ },
1613
+ {
1614
+ "epoch": 1.44,
1615
+ "grad_norm": 12.000615119934082,
1616
+ "learning_rate": 1.3934426229508196e-05,
1617
+ "loss": 0.8439,
1618
+ "step": 90
1619
+ },
1620
+ {
1621
+ "epoch": 1.44,
1622
+ "eval_exact_match": 0.04081632653061224,
1623
+ "eval_f1_a": 0.45,
1624
+ "eval_f1_m": 0.4473113054745707,
1625
+ "eval_loss": 0.7099843621253967,
1626
+ "eval_runtime": 12.3004,
1627
+ "eval_samples_per_second": 20.325,
1628
+ "eval_steps_per_second": 2.602,
1629
+ "step": 90
1630
+ },
1631
+ {
1632
+ "epoch": 1.456,
1633
+ "grad_norm": 9.254725456237793,
1634
+ "learning_rate": 1.3524590163934428e-05,
1635
+ "loss": 0.7011,
1636
+ "step": 91
1637
+ },
1638
+ {
1639
+ "epoch": 1.456,
1640
+ "eval_exact_match": 0.04081632653061224,
1641
+ "eval_f1_a": 0.44102564102564107,
1642
+ "eval_f1_m": 0.44650038170446327,
1643
+ "eval_loss": 0.7092187404632568,
1644
+ "eval_runtime": 12.6142,
1645
+ "eval_samples_per_second": 19.819,
1646
+ "eval_steps_per_second": 2.537,
1647
+ "step": 91
1648
+ },
1649
+ {
1650
+ "epoch": 1.472,
1651
+ "grad_norm": 9.848048210144043,
1652
+ "learning_rate": 1.3114754098360657e-05,
1653
+ "loss": 0.689,
1654
+ "step": 92
1655
+ },
1656
+ {
1657
+ "epoch": 1.472,
1658
+ "eval_exact_match": 0.04081632653061224,
1659
+ "eval_f1_a": 0.45263157894736844,
1660
+ "eval_f1_m": 0.4581459923296658,
1661
+ "eval_loss": 0.7082812786102295,
1662
+ "eval_runtime": 12.4011,
1663
+ "eval_samples_per_second": 20.159,
1664
+ "eval_steps_per_second": 2.58,
1665
+ "step": 92
1666
+ },
1667
+ {
1668
+ "epoch": 1.488,
1669
+ "grad_norm": 17.459680557250977,
1670
+ "learning_rate": 1.2704918032786885e-05,
1671
+ "loss": 0.8424,
1672
+ "step": 93
1673
+ },
1674
+ {
1675
+ "epoch": 1.488,
1676
+ "eval_exact_match": 0.04081632653061224,
1677
+ "eval_f1_a": 0.450261780104712,
1678
+ "eval_f1_m": 0.45620235773296997,
1679
+ "eval_loss": 0.7073437571525574,
1680
+ "eval_runtime": 12.4025,
1681
+ "eval_samples_per_second": 20.157,
1682
+ "eval_steps_per_second": 2.58,
1683
+ "step": 93
1684
+ },
1685
+ {
1686
+ "epoch": 1.504,
1687
+ "grad_norm": 7.91733980178833,
1688
+ "learning_rate": 1.2295081967213116e-05,
1689
+ "loss": 0.6224,
1690
+ "step": 94
1691
+ },
1692
+ {
1693
+ "epoch": 1.504,
1694
+ "eval_exact_match": 0.04081632653061224,
1695
+ "eval_f1_a": 0.4468085106382979,
1696
+ "eval_f1_m": 0.4566639709496852,
1697
+ "eval_loss": 0.7057031393051147,
1698
+ "eval_runtime": 12.4036,
1699
+ "eval_samples_per_second": 20.155,
1700
+ "eval_steps_per_second": 2.58,
1701
+ "step": 94
1702
+ },
1703
+ {
1704
+ "epoch": 1.52,
1705
+ "grad_norm": 17.14729881286621,
1706
+ "learning_rate": 1.1885245901639344e-05,
1707
+ "loss": 0.5743,
1708
+ "step": 95
1709
+ },
1710
+ {
1711
+ "epoch": 1.52,
1712
+ "eval_exact_match": 0.04081632653061224,
1713
+ "eval_f1_a": 0.4491978609625668,
1714
+ "eval_f1_m": 0.4579678258249688,
1715
+ "eval_loss": 0.7053046822547913,
1716
+ "eval_runtime": 12.4028,
1717
+ "eval_samples_per_second": 20.157,
1718
+ "eval_steps_per_second": 2.58,
1719
+ "step": 95
1720
+ },
1721
+ {
1722
+ "epoch": 1.536,
1723
+ "grad_norm": 10.893065452575684,
1724
+ "learning_rate": 1.1475409836065575e-05,
1725
+ "loss": 0.7041,
1726
+ "step": 96
1727
+ },
1728
+ {
1729
+ "epoch": 1.536,
1730
+ "eval_exact_match": 0.04081632653061224,
1731
+ "eval_f1_a": 0.44324324324324327,
1732
+ "eval_f1_m": 0.45810145070349156,
1733
+ "eval_loss": 0.7051210999488831,
1734
+ "eval_runtime": 12.3907,
1735
+ "eval_samples_per_second": 20.176,
1736
+ "eval_steps_per_second": 2.583,
1737
+ "step": 96
1738
+ },
1739
+ {
1740
+ "epoch": 1.552,
1741
+ "grad_norm": 9.716265678405762,
1742
+ "learning_rate": 1.1065573770491803e-05,
1743
+ "loss": 0.6443,
1744
+ "step": 97
1745
+ },
1746
+ {
1747
+ "epoch": 1.552,
1748
+ "eval_exact_match": 0.02040816326530612,
1749
+ "eval_f1_a": 0.4262295081967213,
1750
+ "eval_f1_m": 0.4496709356403233,
1751
+ "eval_loss": 0.7061210870742798,
1752
+ "eval_runtime": 12.2983,
1753
+ "eval_samples_per_second": 20.328,
1754
+ "eval_steps_per_second": 2.602,
1755
+ "step": 97
1756
+ },
1757
+ {
1758
+ "epoch": 1.568,
1759
+ "grad_norm": 17.185867309570312,
1760
+ "learning_rate": 1.0655737704918032e-05,
1761
+ "loss": 0.6588,
1762
+ "step": 98
1763
+ },
1764
+ {
1765
+ "epoch": 1.568,
1766
+ "eval_exact_match": 0.02040816326530612,
1767
+ "eval_f1_a": 0.430939226519337,
1768
+ "eval_f1_m": 0.4559391572146673,
1769
+ "eval_loss": 0.704464852809906,
1770
+ "eval_runtime": 12.4036,
1771
+ "eval_samples_per_second": 20.155,
1772
+ "eval_steps_per_second": 2.58,
1773
+ "step": 98
1774
+ },
1775
+ {
1776
+ "epoch": 1.584,
1777
+ "grad_norm": 9.352198600769043,
1778
+ "learning_rate": 1.0245901639344262e-05,
1779
+ "loss": 0.6669,
1780
+ "step": 99
1781
+ },
1782
+ {
1783
+ "epoch": 1.584,
1784
+ "eval_exact_match": 0.02040816326530612,
1785
+ "eval_f1_a": 0.41860465116279066,
1786
+ "eval_f1_m": 0.45876048721186774,
1787
+ "eval_loss": 0.7052500247955322,
1788
+ "eval_runtime": 12.2869,
1789
+ "eval_samples_per_second": 20.347,
1790
+ "eval_steps_per_second": 2.604,
1791
+ "step": 99
1792
+ },
1793
+ {
1794
+ "epoch": 1.6,
1795
+ "grad_norm": 10.983183860778809,
1796
+ "learning_rate": 9.836065573770493e-06,
1797
+ "loss": 0.667,
1798
+ "step": 100
1799
+ },
1800
+ {
1801
+ "epoch": 1.6,
1802
+ "eval_exact_match": 0.02040816326530612,
1803
+ "eval_f1_a": 0.3976608187134503,
1804
+ "eval_f1_m": 0.44378116214850916,
1805
+ "eval_loss": 0.7053242325782776,
1806
+ "eval_runtime": 12.292,
1807
+ "eval_samples_per_second": 20.338,
1808
+ "eval_steps_per_second": 2.603,
1809
+ "step": 100
1810
+ },
1811
+ {
1812
+ "epoch": 1.616,
1813
+ "grad_norm": 8.352677345275879,
1814
+ "learning_rate": 9.426229508196721e-06,
1815
+ "loss": 0.6028,
1816
+ "step": 101
1817
+ },
1818
+ {
1819
+ "epoch": 1.616,
1820
+ "eval_exact_match": 0.02040816326530612,
1821
+ "eval_f1_a": 0.3780487804878049,
1822
+ "eval_f1_m": 0.440360093421318,
1823
+ "eval_loss": 0.7049765586853027,
1824
+ "eval_runtime": 12.3969,
1825
+ "eval_samples_per_second": 20.166,
1826
+ "eval_steps_per_second": 2.581,
1827
+ "step": 101
1828
+ },
1829
+ {
1830
+ "epoch": 1.6320000000000001,
1831
+ "grad_norm": 18.418481826782227,
1832
+ "learning_rate": 9.016393442622952e-06,
1833
+ "loss": 0.7084,
1834
+ "step": 102
1835
+ },
1836
+ {
1837
+ "epoch": 1.6320000000000001,
1838
+ "eval_exact_match": 0.02040816326530612,
1839
+ "eval_f1_a": 0.3827160493827161,
1840
+ "eval_f1_m": 0.4446715982430268,
1841
+ "eval_loss": 0.7034218907356262,
1842
+ "eval_runtime": 12.3991,
1843
+ "eval_samples_per_second": 20.163,
1844
+ "eval_steps_per_second": 2.581,
1845
+ "step": 102
1846
+ },
1847
+ {
1848
+ "epoch": 1.6480000000000001,
1849
+ "grad_norm": 10.633659362792969,
1850
+ "learning_rate": 8.60655737704918e-06,
1851
+ "loss": 0.6221,
1852
+ "step": 103
1853
+ },
1854
+ {
1855
+ "epoch": 1.6480000000000001,
1856
+ "eval_exact_match": 0.02040816326530612,
1857
+ "eval_f1_a": 0.33766233766233766,
1858
+ "eval_f1_m": 0.42304736080246286,
1859
+ "eval_loss": 0.7042617201805115,
1860
+ "eval_runtime": 12.2954,
1861
+ "eval_samples_per_second": 20.333,
1862
+ "eval_steps_per_second": 2.603,
1863
+ "step": 103
1864
+ },
1865
+ {
1866
+ "epoch": 1.6640000000000001,
1867
+ "grad_norm": 12.377849578857422,
1868
+ "learning_rate": 8.196721311475409e-06,
1869
+ "loss": 0.5413,
1870
+ "step": 104
1871
+ },
1872
+ {
1873
+ "epoch": 1.6640000000000001,
1874
+ "eval_exact_match": 0.02040816326530612,
1875
+ "eval_f1_a": 0.3289473684210526,
1876
+ "eval_f1_m": 0.4253797223184978,
1877
+ "eval_loss": 0.7033711075782776,
1878
+ "eval_runtime": 12.3003,
1879
+ "eval_samples_per_second": 20.325,
1880
+ "eval_steps_per_second": 2.602,
1881
+ "step": 104
1882
+ },
1883
+ {
1884
+ "epoch": 1.6800000000000002,
1885
+ "grad_norm": 12.551422119140625,
1886
+ "learning_rate": 7.78688524590164e-06,
1887
+ "loss": 0.682,
1888
+ "step": 105
1889
+ },
1890
+ {
1891
+ "epoch": 1.6800000000000002,
1892
+ "eval_exact_match": 0.0,
1893
+ "eval_f1_a": 0.2953020134228188,
1894
+ "eval_f1_m": 0.4078515022392573,
1895
+ "eval_loss": 0.7049804925918579,
1896
+ "eval_runtime": 12.495,
1897
+ "eval_samples_per_second": 20.008,
1898
+ "eval_steps_per_second": 2.561,
1899
+ "step": 105
1900
+ },
1901
+ {
1902
+ "epoch": 1.696,
1903
+ "grad_norm": 8.237449645996094,
1904
+ "learning_rate": 7.3770491803278695e-06,
1905
+ "loss": 0.7448,
1906
+ "step": 106
1907
+ },
1908
+ {
1909
+ "epoch": 1.696,
1910
+ "eval_exact_match": 0.0,
1911
+ "eval_f1_a": 0.3013698630136986,
1912
+ "eval_f1_m": 0.41549131140967865,
1913
+ "eval_loss": 0.7063398361206055,
1914
+ "eval_runtime": 12.4967,
1915
+ "eval_samples_per_second": 20.005,
1916
+ "eval_steps_per_second": 2.561,
1917
+ "step": 106
1918
+ },
1919
+ {
1920
+ "epoch": 1.712,
1921
+ "grad_norm": 7.336461067199707,
1922
+ "learning_rate": 6.967213114754098e-06,
1923
+ "loss": 0.4493,
1924
+ "step": 107
1925
+ },
1926
+ {
1927
+ "epoch": 1.712,
1928
+ "eval_exact_match": 0.0,
1929
+ "eval_f1_a": 0.2553191489361702,
1930
+ "eval_f1_m": 0.4051889720257066,
1931
+ "eval_loss": 0.7075039148330688,
1932
+ "eval_runtime": 12.403,
1933
+ "eval_samples_per_second": 20.156,
1934
+ "eval_steps_per_second": 2.58,
1935
+ "step": 107
1936
+ },
1937
+ {
1938
+ "epoch": 1.728,
1939
+ "grad_norm": 8.348127365112305,
1940
+ "learning_rate": 6.557377049180328e-06,
1941
+ "loss": 0.8977,
1942
+ "step": 108
1943
+ },
1944
+ {
1945
+ "epoch": 1.728,
1946
+ "eval_exact_match": 0.0,
1947
+ "eval_f1_a": 0.2608695652173913,
1948
+ "eval_f1_m": 0.4141215926930211,
1949
+ "eval_loss": 0.7066875100135803,
1950
+ "eval_runtime": 12.3958,
1951
+ "eval_samples_per_second": 20.168,
1952
+ "eval_steps_per_second": 2.582,
1953
+ "step": 108
1954
+ },
1955
+ {
1956
+ "epoch": 1.744,
1957
+ "grad_norm": 6.762801170349121,
1958
+ "learning_rate": 6.147540983606558e-06,
1959
+ "loss": 0.6888,
1960
+ "step": 109
1961
+ },
1962
+ {
1963
+ "epoch": 1.744,
1964
+ "eval_exact_match": 0.0,
1965
+ "eval_f1_a": 0.23529411764705882,
1966
+ "eval_f1_m": 0.4052051689806791,
1967
+ "eval_loss": 0.7079609632492065,
1968
+ "eval_runtime": 12.2957,
1969
+ "eval_samples_per_second": 20.332,
1970
+ "eval_steps_per_second": 2.603,
1971
+ "step": 109
1972
+ },
1973
+ {
1974
+ "epoch": 1.76,
1975
+ "grad_norm": 7.54478645324707,
1976
+ "learning_rate": 5.737704918032787e-06,
1977
+ "loss": 0.7079,
1978
+ "step": 110
1979
+ },
1980
+ {
1981
+ "epoch": 1.76,
1982
+ "eval_exact_match": 0.0,
1983
+ "eval_f1_a": 0.196969696969697,
1984
+ "eval_f1_m": 0.3930353659945496,
1985
+ "eval_loss": 0.7083203196525574,
1986
+ "eval_runtime": 12.4001,
1987
+ "eval_samples_per_second": 20.161,
1988
+ "eval_steps_per_second": 2.581,
1989
+ "step": 110
1990
+ },
1991
+ {
1992
+ "epoch": 1.776,
1993
+ "grad_norm": 6.84658145904541,
1994
+ "learning_rate": 5.327868852459016e-06,
1995
+ "loss": 0.6766,
1996
+ "step": 111
1997
+ },
1998
+ {
1999
+ "epoch": 1.776,
2000
+ "eval_exact_match": 0.0,
2001
+ "eval_f1_a": 0.2105263157894737,
2002
+ "eval_f1_m": 0.39676157176157173,
2003
+ "eval_loss": 0.7091640830039978,
2004
+ "eval_runtime": 12.4008,
2005
+ "eval_samples_per_second": 20.16,
2006
+ "eval_steps_per_second": 2.58,
2007
+ "step": 111
2008
+ },
2009
+ {
2010
+ "epoch": 1.792,
2011
+ "grad_norm": 16.06954002380371,
2012
+ "learning_rate": 4.918032786885246e-06,
2013
+ "loss": 0.6298,
2014
+ "step": 112
2015
+ },
2016
+ {
2017
+ "epoch": 1.792,
2018
+ "eval_exact_match": 0.0,
2019
+ "eval_f1_a": 0.15748031496062992,
2020
+ "eval_f1_m": 0.37861616841208684,
2021
+ "eval_loss": 0.7095195055007935,
2022
+ "eval_runtime": 12.4951,
2023
+ "eval_samples_per_second": 20.008,
2024
+ "eval_steps_per_second": 2.561,
2025
+ "step": 112
2026
+ },
2027
+ {
2028
+ "epoch": 1.808,
2029
+ "grad_norm": 5.902178764343262,
2030
+ "learning_rate": 4.508196721311476e-06,
2031
+ "loss": 0.5864,
2032
+ "step": 113
2033
+ },
2034
+ {
2035
+ "epoch": 1.808,
2036
+ "eval_exact_match": 0.0,
2037
+ "eval_f1_a": 0.15873015873015875,
2038
+ "eval_f1_m": 0.38075416646845217,
2039
+ "eval_loss": 0.7111679911613464,
2040
+ "eval_runtime": 12.4019,
2041
+ "eval_samples_per_second": 20.158,
2042
+ "eval_steps_per_second": 2.58,
2043
+ "step": 113
2044
+ },
2045
+ {
2046
+ "epoch": 1.8239999999999998,
2047
+ "grad_norm": 17.344051361083984,
2048
+ "learning_rate": 4.098360655737704e-06,
2049
+ "loss": 0.7265,
2050
+ "step": 114
2051
+ },
2052
+ {
2053
+ "epoch": 1.8239999999999998,
2054
+ "eval_exact_match": 0.0,
2055
+ "eval_f1_a": 0.16,
2056
+ "eval_f1_m": 0.38211471068613934,
2057
+ "eval_loss": 0.7119726538658142,
2058
+ "eval_runtime": 12.3958,
2059
+ "eval_samples_per_second": 20.168,
2060
+ "eval_steps_per_second": 2.582,
2061
+ "step": 114
2062
+ },
2063
+ {
2064
+ "epoch": 1.8399999999999999,
2065
+ "grad_norm": 13.185153007507324,
2066
+ "learning_rate": 3.6885245901639347e-06,
2067
+ "loss": 0.5856,
2068
+ "step": 115
2069
+ },
2070
+ {
2071
+ "epoch": 1.8399999999999999,
2072
+ "eval_exact_match": 0.0,
2073
+ "eval_f1_a": 0.15873015873015875,
2074
+ "eval_f1_m": 0.37997671262977384,
2075
+ "eval_loss": 0.7116367220878601,
2076
+ "eval_runtime": 12.3902,
2077
+ "eval_samples_per_second": 20.177,
2078
+ "eval_steps_per_second": 2.583,
2079
+ "step": 115
2080
+ },
2081
+ {
2082
+ "epoch": 1.8559999999999999,
2083
+ "grad_norm": 6.701231956481934,
2084
+ "learning_rate": 3.278688524590164e-06,
2085
+ "loss": 0.718,
2086
+ "step": 116
2087
+ },
2088
+ {
2089
+ "epoch": 1.8559999999999999,
2090
+ "eval_exact_match": 0.0,
2091
+ "eval_f1_a": 0.09836065573770492,
2092
+ "eval_f1_m": 0.361842380975034,
2093
+ "eval_loss": 0.7136015892028809,
2094
+ "eval_runtime": 12.294,
2095
+ "eval_samples_per_second": 20.335,
2096
+ "eval_steps_per_second": 2.603,
2097
+ "step": 116
2098
+ },
2099
+ {
2100
+ "epoch": 1.8719999999999999,
2101
+ "grad_norm": 11.730875015258789,
2102
+ "learning_rate": 2.8688524590163937e-06,
2103
+ "loss": 0.7096,
2104
+ "step": 117
2105
+ },
2106
+ {
2107
+ "epoch": 1.8719999999999999,
2108
+ "eval_exact_match": 0.0,
2109
+ "eval_f1_a": 0.08333333333333334,
2110
+ "eval_f1_m": 0.35354549079038877,
2111
+ "eval_loss": 0.7125195264816284,
2112
+ "eval_runtime": 12.3992,
2113
+ "eval_samples_per_second": 20.163,
2114
+ "eval_steps_per_second": 2.581,
2115
+ "step": 117
2116
+ },
2117
+ {
2118
+ "epoch": 1.888,
2119
+ "grad_norm": 11.764548301696777,
2120
+ "learning_rate": 2.459016393442623e-06,
2121
+ "loss": 0.7648,
2122
+ "step": 118
2123
+ },
2124
+ {
2125
+ "epoch": 1.888,
2126
+ "eval_exact_match": 0.0,
2127
+ "eval_f1_a": 0.10169491525423728,
2128
+ "eval_f1_m": 0.36885033900340025,
2129
+ "eval_loss": 0.7145312428474426,
2130
+ "eval_runtime": 12.2937,
2131
+ "eval_samples_per_second": 20.336,
2132
+ "eval_steps_per_second": 2.603,
2133
+ "step": 118
2134
+ },
2135
+ {
2136
+ "epoch": 1.904,
2137
+ "grad_norm": 7.357093334197998,
2138
+ "learning_rate": 2.049180327868852e-06,
2139
+ "loss": 0.66,
2140
+ "step": 119
2141
+ },
2142
+ {
2143
+ "epoch": 1.904,
2144
+ "eval_exact_match": 0.0,
2145
+ "eval_f1_a": 0.08403361344537816,
2146
+ "eval_f1_m": 0.3581008843763946,
2147
+ "eval_loss": 0.7142617106437683,
2148
+ "eval_runtime": 12.2932,
2149
+ "eval_samples_per_second": 20.336,
2150
+ "eval_steps_per_second": 2.603,
2151
+ "step": 119
2152
+ },
2153
+ {
2154
+ "epoch": 1.92,
2155
+ "grad_norm": 12.222306251525879,
2156
+ "learning_rate": 1.639344262295082e-06,
2157
+ "loss": 0.7134,
2158
+ "step": 120
2159
+ },
2160
+ {
2161
+ "epoch": 1.92,
2162
+ "eval_exact_match": 0.0,
2163
+ "eval_f1_a": 0.08620689655172413,
2164
+ "eval_f1_m": 0.3664693866224478,
2165
+ "eval_loss": 0.7143398523330688,
2166
+ "eval_runtime": 12.4037,
2167
+ "eval_samples_per_second": 20.155,
2168
+ "eval_steps_per_second": 2.58,
2169
+ "step": 120
2170
+ },
2171
+ {
2172
+ "epoch": 1.936,
2173
+ "grad_norm": 11.94378662109375,
2174
+ "learning_rate": 1.2295081967213116e-06,
2175
+ "loss": 0.6269,
2176
+ "step": 121
2177
+ },
2178
+ {
2179
+ "epoch": 1.936,
2180
+ "eval_exact_match": 0.0,
2181
+ "eval_f1_a": 0.06837606837606838,
2182
+ "eval_f1_m": 0.35452724712928796,
2183
+ "eval_loss": 0.7161992192268372,
2184
+ "eval_runtime": 12.2894,
2185
+ "eval_samples_per_second": 20.343,
2186
+ "eval_steps_per_second": 2.604,
2187
+ "step": 121
2188
+ },
2189
+ {
2190
+ "epoch": 1.952,
2191
+ "grad_norm": 10.898783683776855,
2192
+ "learning_rate": 8.19672131147541e-07,
2193
+ "loss": 0.6692,
2194
+ "step": 122
2195
+ },
2196
+ {
2197
+ "epoch": 1.952,
2198
+ "eval_exact_match": 0.0,
2199
+ "eval_f1_a": 0.06779661016949154,
2200
+ "eval_f1_m": 0.35377437355228475,
2201
+ "eval_loss": 0.7155507802963257,
2202
+ "eval_runtime": 12.2897,
2203
+ "eval_samples_per_second": 20.342,
2204
+ "eval_steps_per_second": 2.604,
2205
+ "step": 122
2206
+ },
2207
+ {
2208
+ "epoch": 1.968,
2209
+ "grad_norm": 11.34163761138916,
2210
+ "learning_rate": 4.098360655737705e-07,
2211
+ "loss": 0.7837,
2212
+ "step": 123
2213
+ },
2214
+ {
2215
+ "epoch": 1.968,
2216
+ "eval_exact_match": 0.0,
2217
+ "eval_f1_a": 0.06837606837606838,
2218
+ "eval_f1_m": 0.3542995836363184,
2219
+ "eval_loss": 0.7164883017539978,
2220
+ "eval_runtime": 12.3974,
2221
+ "eval_samples_per_second": 20.166,
2222
+ "eval_steps_per_second": 2.581,
2223
+ "step": 123
2224
+ },
2225
+ {
2226
+ "epoch": 1.984,
2227
+ "grad_norm": 9.480073928833008,
2228
+ "learning_rate": 0.0,
2229
+ "loss": 0.5815,
2230
+ "step": 124
2231
+ },
2232
+ {
2233
+ "epoch": 1.984,
2234
+ "eval_exact_match": 0.0,
2235
+ "eval_f1_a": 0.053097345132743355,
2236
+ "eval_f1_m": 0.35416089179594584,
2237
+ "eval_loss": 0.7167656421661377,
2238
+ "eval_runtime": 12.2934,
2239
+ "eval_samples_per_second": 20.336,
2240
+ "eval_steps_per_second": 2.603,
2241
+ "step": 124
2242
+ },
2243
+ {
2244
+ "epoch": 1.984,
2245
+ "step": 124,
2246
+ "total_flos": 107566092058624.0,
2247
+ "train_loss": 0.7248697588520665,
2248
+ "train_runtime": 2001.1585,
2249
+ "train_samples_per_second": 0.999,
2250
+ "train_steps_per_second": 0.062
2251
+ }
2252
+ ],
2253
+ "logging_steps": 1,
2254
+ "max_steps": 124,
2255
+ "num_input_tokens_seen": 0,
2256
+ "num_train_epochs": 2,
2257
+ "save_steps": 500,
2258
+ "total_flos": 107566092058624.0,
2259
+ "train_batch_size": 2,
2260
+ "trial_name": null,
2261
+ "trial_params": null
2262
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5daa9d904db1c0d816f3e51f5e00ef720b45916ab080defd25795dfb5ee4107
3
+ size 5048
vocab.json ADDED
The diff for this file is too large to render. See raw diff