Model save
Browse files- README.md +67 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Aug11_20-49-38_COE-CS-sv003/events.out.tfevents.1723409405.COE-CS-sv003.3806331.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1478 -0
README.md
ADDED
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
+
datasets:
|
9 |
+
- generator
|
10 |
+
model-index:
|
11 |
+
- name: prometheus-7b-directp0.1_seed42_syntax
|
12 |
+
results: []
|
13 |
+
---
|
14 |
+
|
15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
16 |
+
should probably proofread and complete it, then remove this comment. -->
|
17 |
+
|
18 |
+
# prometheus-7b-directp0.1_seed42_syntax
|
19 |
+
|
20 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset.
|
21 |
+
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.3341
|
23 |
+
|
24 |
+
## Model description
|
25 |
+
|
26 |
+
More information needed
|
27 |
+
|
28 |
+
## Intended uses & limitations
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Training and evaluation data
|
33 |
+
|
34 |
+
More information needed
|
35 |
+
|
36 |
+
## Training procedure
|
37 |
+
|
38 |
+
### Training hyperparameters
|
39 |
+
|
40 |
+
The following hyperparameters were used during training:
|
41 |
+
- learning_rate: 1e-05
|
42 |
+
- train_batch_size: 2
|
43 |
+
- eval_batch_size: 1
|
44 |
+
- seed: 42
|
45 |
+
- distributed_type: multi-GPU
|
46 |
+
- num_devices: 4
|
47 |
+
- gradient_accumulation_steps: 4
|
48 |
+
- total_train_batch_size: 32
|
49 |
+
- total_eval_batch_size: 4
|
50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
51 |
+
- lr_scheduler_type: cosine
|
52 |
+
- lr_scheduler_warmup_ratio: 0.1
|
53 |
+
- num_epochs: 1
|
54 |
+
|
55 |
+
### Training results
|
56 |
+
|
57 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
58 |
+
|:-------------:|:------:|:----:|:---------------:|
|
59 |
+
| 0.3258 | 0.9995 | 1019 | 0.3341 |
|
60 |
+
|
61 |
+
|
62 |
+
### Framework versions
|
63 |
+
|
64 |
+
- Transformers 4.43.4
|
65 |
+
- Pytorch 2.3.1+cu121
|
66 |
+
- Datasets 2.19.1
|
67 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.9995095635115253,
|
3 |
+
"total_flos": 213305524224000.0,
|
4 |
+
"train_loss": 0.4876700218573169,
|
5 |
+
"train_runtime": 22910.3839,
|
6 |
+
"train_samples": 108847,
|
7 |
+
"train_samples_per_second": 1.424,
|
8 |
+
"train_steps_per_second": 0.044
|
9 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"eos_token_id": 2,
|
5 |
+
"transformers_version": "4.43.4"
|
6 |
+
}
|
runs/Aug11_20-49-38_COE-CS-sv003/events.out.tfevents.1723409405.COE-CS-sv003.3806331.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3a6eefad1e9612dcaca71666dce428ada61c29081d61b9ba53bb0a825abebaa0
|
3 |
+
size 48809
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.9995095635115253,
|
3 |
+
"total_flos": 213305524224000.0,
|
4 |
+
"train_loss": 0.4876700218573169,
|
5 |
+
"train_runtime": 22910.3839,
|
6 |
+
"train_samples": 108847,
|
7 |
+
"train_samples_per_second": 1.424,
|
8 |
+
"train_steps_per_second": 0.044
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1478 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9995095635115253,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1019,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.000980872976949485,
|
13 |
+
"grad_norm": 23.71615728078218,
|
14 |
+
"learning_rate": 9.803921568627452e-08,
|
15 |
+
"loss": 1.3172,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.004904364884747425,
|
20 |
+
"grad_norm": 21.50955102466688,
|
21 |
+
"learning_rate": 4.901960784313725e-07,
|
22 |
+
"loss": 1.3156,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.00980872976949485,
|
27 |
+
"grad_norm": 8.347835856988212,
|
28 |
+
"learning_rate": 9.80392156862745e-07,
|
29 |
+
"loss": 1.2118,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.014713094654242276,
|
34 |
+
"grad_norm": 9.70215349989816,
|
35 |
+
"learning_rate": 1.4705882352941177e-06,
|
36 |
+
"loss": 1.0495,
|
37 |
+
"step": 15
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.0196174595389897,
|
41 |
+
"grad_norm": 2.935709350105428,
|
42 |
+
"learning_rate": 1.96078431372549e-06,
|
43 |
+
"loss": 0.9169,
|
44 |
+
"step": 20
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.024521824423737126,
|
48 |
+
"grad_norm": 2.3710459957915373,
|
49 |
+
"learning_rate": 2.450980392156863e-06,
|
50 |
+
"loss": 0.8718,
|
51 |
+
"step": 25
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.029426189308484552,
|
55 |
+
"grad_norm": 2.217748118460408,
|
56 |
+
"learning_rate": 2.9411764705882355e-06,
|
57 |
+
"loss": 0.8404,
|
58 |
+
"step": 30
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.03433055419323198,
|
62 |
+
"grad_norm": 2.2456090605656223,
|
63 |
+
"learning_rate": 3.431372549019608e-06,
|
64 |
+
"loss": 0.8207,
|
65 |
+
"step": 35
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.0392349190779794,
|
69 |
+
"grad_norm": 2.1968777765698135,
|
70 |
+
"learning_rate": 3.92156862745098e-06,
|
71 |
+
"loss": 0.8027,
|
72 |
+
"step": 40
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.04413928396272683,
|
76 |
+
"grad_norm": 2.3343295421758956,
|
77 |
+
"learning_rate": 4.411764705882353e-06,
|
78 |
+
"loss": 0.7888,
|
79 |
+
"step": 45
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.04904364884747425,
|
83 |
+
"grad_norm": 2.3845961870372956,
|
84 |
+
"learning_rate": 4.901960784313726e-06,
|
85 |
+
"loss": 0.7703,
|
86 |
+
"step": 50
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.053948013732221675,
|
90 |
+
"grad_norm": 2.3216256302247933,
|
91 |
+
"learning_rate": 5.392156862745098e-06,
|
92 |
+
"loss": 0.7541,
|
93 |
+
"step": 55
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.058852378616969105,
|
97 |
+
"grad_norm": 2.378678579603692,
|
98 |
+
"learning_rate": 5.882352941176471e-06,
|
99 |
+
"loss": 0.749,
|
100 |
+
"step": 60
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.06375674350171653,
|
104 |
+
"grad_norm": 2.355836921671654,
|
105 |
+
"learning_rate": 6.372549019607843e-06,
|
106 |
+
"loss": 0.7258,
|
107 |
+
"step": 65
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.06866110838646396,
|
111 |
+
"grad_norm": 2.480621419103395,
|
112 |
+
"learning_rate": 6.862745098039216e-06,
|
113 |
+
"loss": 0.7132,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.07356547327121138,
|
118 |
+
"grad_norm": 2.196577695253739,
|
119 |
+
"learning_rate": 7.352941176470589e-06,
|
120 |
+
"loss": 0.7168,
|
121 |
+
"step": 75
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.0784698381559588,
|
125 |
+
"grad_norm": 2.418840940827789,
|
126 |
+
"learning_rate": 7.84313725490196e-06,
|
127 |
+
"loss": 0.7051,
|
128 |
+
"step": 80
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.08337420304070622,
|
132 |
+
"grad_norm": 2.4161115457147577,
|
133 |
+
"learning_rate": 8.333333333333334e-06,
|
134 |
+
"loss": 0.6993,
|
135 |
+
"step": 85
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.08827856792545366,
|
139 |
+
"grad_norm": 2.3049037332530804,
|
140 |
+
"learning_rate": 8.823529411764707e-06,
|
141 |
+
"loss": 0.6948,
|
142 |
+
"step": 90
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.09318293281020108,
|
146 |
+
"grad_norm": 2.345395326875072,
|
147 |
+
"learning_rate": 9.31372549019608e-06,
|
148 |
+
"loss": 0.6859,
|
149 |
+
"step": 95
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.0980872976949485,
|
153 |
+
"grad_norm": 2.299729975358926,
|
154 |
+
"learning_rate": 9.803921568627451e-06,
|
155 |
+
"loss": 0.6836,
|
156 |
+
"step": 100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.10299166257969593,
|
160 |
+
"grad_norm": 2.2610515025654117,
|
161 |
+
"learning_rate": 9.999735917410952e-06,
|
162 |
+
"loss": 0.6794,
|
163 |
+
"step": 105
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.10789602746444335,
|
167 |
+
"grad_norm": 2.450010148482251,
|
168 |
+
"learning_rate": 9.998122180387662e-06,
|
169 |
+
"loss": 0.6765,
|
170 |
+
"step": 110
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.11280039234919079,
|
174 |
+
"grad_norm": 2.221120129726642,
|
175 |
+
"learning_rate": 9.995041891820093e-06,
|
176 |
+
"loss": 0.6746,
|
177 |
+
"step": 115
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.11770475723393821,
|
181 |
+
"grad_norm": 2.1057039809456377,
|
182 |
+
"learning_rate": 9.990495955528073e-06,
|
183 |
+
"loss": 0.6644,
|
184 |
+
"step": 120
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.12260912211868563,
|
188 |
+
"grad_norm": 2.090893956936151,
|
189 |
+
"learning_rate": 9.984485705382538e-06,
|
190 |
+
"loss": 0.6695,
|
191 |
+
"step": 125
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.12751348700343307,
|
195 |
+
"grad_norm": 2.0602352081070543,
|
196 |
+
"learning_rate": 9.977012904914133e-06,
|
197 |
+
"loss": 0.6519,
|
198 |
+
"step": 130
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.13241785188818048,
|
202 |
+
"grad_norm": 2.0932998186794927,
|
203 |
+
"learning_rate": 9.968079746795759e-06,
|
204 |
+
"loss": 0.6657,
|
205 |
+
"step": 135
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.1373222167729279,
|
209 |
+
"grad_norm": 2.136782363642342,
|
210 |
+
"learning_rate": 9.957688852199201e-06,
|
211 |
+
"loss": 0.6557,
|
212 |
+
"step": 140
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.14222658165767532,
|
216 |
+
"grad_norm": 2.1261422382559623,
|
217 |
+
"learning_rate": 9.945843270026021e-06,
|
218 |
+
"loss": 0.6495,
|
219 |
+
"step": 145
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.14713094654242276,
|
223 |
+
"grad_norm": 2.067127907721412,
|
224 |
+
"learning_rate": 9.932546476012942e-06,
|
225 |
+
"loss": 0.6411,
|
226 |
+
"step": 150
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.1520353114271702,
|
230 |
+
"grad_norm": 2.040847654490592,
|
231 |
+
"learning_rate": 9.91780237171201e-06,
|
232 |
+
"loss": 0.6416,
|
233 |
+
"step": 155
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.1569396763119176,
|
237 |
+
"grad_norm": 2.0912635989059787,
|
238 |
+
"learning_rate": 9.901615283345782e-06,
|
239 |
+
"loss": 0.6503,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.16184404119666504,
|
244 |
+
"grad_norm": 2.8372493265750873,
|
245 |
+
"learning_rate": 9.883989960537934e-06,
|
246 |
+
"loss": 0.6424,
|
247 |
+
"step": 165
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.16674840608141245,
|
251 |
+
"grad_norm": 2.0493928993359307,
|
252 |
+
"learning_rate": 9.86493157491962e-06,
|
253 |
+
"loss": 0.6387,
|
254 |
+
"step": 170
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.17165277096615988,
|
258 |
+
"grad_norm": 2.083065623023665,
|
259 |
+
"learning_rate": 9.84444571861201e-06,
|
260 |
+
"loss": 0.6362,
|
261 |
+
"step": 175
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.17655713585090732,
|
265 |
+
"grad_norm": 2.066269679181421,
|
266 |
+
"learning_rate": 9.822538402585451e-06,
|
267 |
+
"loss": 0.6277,
|
268 |
+
"step": 180
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.18146150073565473,
|
272 |
+
"grad_norm": 2.1282623287152167,
|
273 |
+
"learning_rate": 9.799216054895715e-06,
|
274 |
+
"loss": 0.6274,
|
275 |
+
"step": 185
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.18636586562040217,
|
279 |
+
"grad_norm": 2.0335813176241095,
|
280 |
+
"learning_rate": 9.774485518797892e-06,
|
281 |
+
"loss": 0.6155,
|
282 |
+
"step": 190
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.19127023050514957,
|
286 |
+
"grad_norm": 1.9384285409252677,
|
287 |
+
"learning_rate": 9.748354050738416e-06,
|
288 |
+
"loss": 0.638,
|
289 |
+
"step": 195
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.196174595389897,
|
293 |
+
"grad_norm": 2.006376224222685,
|
294 |
+
"learning_rate": 9.720829318225897e-06,
|
295 |
+
"loss": 0.613,
|
296 |
+
"step": 200
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.20107896027464445,
|
300 |
+
"grad_norm": 2.088420694162267,
|
301 |
+
"learning_rate": 9.691919397581304e-06,
|
302 |
+
"loss": 0.6139,
|
303 |
+
"step": 205
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.20598332515939186,
|
307 |
+
"grad_norm": 2.0409644262977515,
|
308 |
+
"learning_rate": 9.66163277156821e-06,
|
309 |
+
"loss": 0.6068,
|
310 |
+
"step": 210
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.2108876900441393,
|
314 |
+
"grad_norm": 2.0023467594132973,
|
315 |
+
"learning_rate": 9.629978326903778e-06,
|
316 |
+
"loss": 0.6084,
|
317 |
+
"step": 215
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.2157920549288867,
|
321 |
+
"grad_norm": 1.904954310523434,
|
322 |
+
"learning_rate": 9.596965351651204e-06,
|
323 |
+
"loss": 0.6045,
|
324 |
+
"step": 220
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.22069641981363414,
|
328 |
+
"grad_norm": 1.964486390414263,
|
329 |
+
"learning_rate": 9.562603532494432e-06,
|
330 |
+
"loss": 0.6197,
|
331 |
+
"step": 225
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.22560078469838157,
|
335 |
+
"grad_norm": 2.0628213777647577,
|
336 |
+
"learning_rate": 9.526902951895857e-06,
|
337 |
+
"loss": 0.5853,
|
338 |
+
"step": 230
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.23050514958312898,
|
342 |
+
"grad_norm": 2.067943464789154,
|
343 |
+
"learning_rate": 9.48987408513794e-06,
|
344 |
+
"loss": 0.5892,
|
345 |
+
"step": 235
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.23540951446787642,
|
349 |
+
"grad_norm": 2.1163968024177757,
|
350 |
+
"learning_rate": 9.451527797249538e-06,
|
351 |
+
"loss": 0.5866,
|
352 |
+
"step": 240
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.24031387935262383,
|
356 |
+
"grad_norm": 2.218142662563601,
|
357 |
+
"learning_rate": 9.411875339817886e-06,
|
358 |
+
"loss": 0.5923,
|
359 |
+
"step": 245
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.24521824423737126,
|
363 |
+
"grad_norm": 2.0789211498714057,
|
364 |
+
"learning_rate": 9.370928347687149e-06,
|
365 |
+
"loss": 0.6067,
|
366 |
+
"step": 250
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.2501226091221187,
|
370 |
+
"grad_norm": 2.00243401967041,
|
371 |
+
"learning_rate": 9.328698835544516e-06,
|
372 |
+
"loss": 0.5733,
|
373 |
+
"step": 255
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.25502697400686614,
|
377 |
+
"grad_norm": 1.9232684596963454,
|
378 |
+
"learning_rate": 9.285199194394854e-06,
|
379 |
+
"loss": 0.6039,
|
380 |
+
"step": 260
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.2599313388916135,
|
384 |
+
"grad_norm": 2.4559722910125523,
|
385 |
+
"learning_rate": 9.240442187924922e-06,
|
386 |
+
"loss": 0.5837,
|
387 |
+
"step": 265
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.26483570377636095,
|
391 |
+
"grad_norm": 2.0642822474573235,
|
392 |
+
"learning_rate": 9.19444094875825e-06,
|
393 |
+
"loss": 0.5816,
|
394 |
+
"step": 270
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.2697400686611084,
|
398 |
+
"grad_norm": 1.8431714309717544,
|
399 |
+
"learning_rate": 9.147208974601762e-06,
|
400 |
+
"loss": 0.5891,
|
401 |
+
"step": 275
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.2746444335458558,
|
405 |
+
"grad_norm": 2.455382996504095,
|
406 |
+
"learning_rate": 9.098760124285255e-06,
|
407 |
+
"loss": 0.5739,
|
408 |
+
"step": 280
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.27954879843060326,
|
412 |
+
"grad_norm": 2.105877839231797,
|
413 |
+
"learning_rate": 9.049108613694958e-06,
|
414 |
+
"loss": 0.5664,
|
415 |
+
"step": 285
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.28445316331535064,
|
419 |
+
"grad_norm": 2.1572999228459637,
|
420 |
+
"learning_rate": 8.998269011602283e-06,
|
421 |
+
"loss": 0.5654,
|
422 |
+
"step": 290
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.2893575282000981,
|
426 |
+
"grad_norm": 2.2354758276064257,
|
427 |
+
"learning_rate": 8.94625623538905e-06,
|
428 |
+
"loss": 0.5718,
|
429 |
+
"step": 295
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.2942618930848455,
|
433 |
+
"grad_norm": 1.9951110551471705,
|
434 |
+
"learning_rate": 8.893085546670426e-06,
|
435 |
+
"loss": 0.5647,
|
436 |
+
"step": 300
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.29916625796959295,
|
440 |
+
"grad_norm": 1.940858410217326,
|
441 |
+
"learning_rate": 8.838772546816857e-06,
|
442 |
+
"loss": 0.5503,
|
443 |
+
"step": 305
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.3040706228543404,
|
447 |
+
"grad_norm": 2.1124206964514567,
|
448 |
+
"learning_rate": 8.783333172376292e-06,
|
449 |
+
"loss": 0.5625,
|
450 |
+
"step": 310
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.30897498773908777,
|
454 |
+
"grad_norm": 1.9926807872955052,
|
455 |
+
"learning_rate": 8.726783690398091e-06,
|
456 |
+
"loss": 0.5406,
|
457 |
+
"step": 315
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.3138793526238352,
|
461 |
+
"grad_norm": 2.3776434685854664,
|
462 |
+
"learning_rate": 8.669140693659928e-06,
|
463 |
+
"loss": 0.5412,
|
464 |
+
"step": 320
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.31878371750858264,
|
468 |
+
"grad_norm": 2.0505624375679194,
|
469 |
+
"learning_rate": 8.610421095799129e-06,
|
470 |
+
"loss": 0.5465,
|
471 |
+
"step": 325
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.3236880823933301,
|
475 |
+
"grad_norm": 1.9938642512875002,
|
476 |
+
"learning_rate": 8.550642126349873e-06,
|
477 |
+
"loss": 0.5448,
|
478 |
+
"step": 330
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.3285924472780775,
|
482 |
+
"grad_norm": 1.8794344216432206,
|
483 |
+
"learning_rate": 8.489821325687682e-06,
|
484 |
+
"loss": 0.5309,
|
485 |
+
"step": 335
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.3334968121628249,
|
489 |
+
"grad_norm": 1.9586224504819914,
|
490 |
+
"learning_rate": 8.427976539882725e-06,
|
491 |
+
"loss": 0.5256,
|
492 |
+
"step": 340
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.33840117704757233,
|
496 |
+
"grad_norm": 1.9633684416354464,
|
497 |
+
"learning_rate": 8.365125915463406e-06,
|
498 |
+
"loss": 0.528,
|
499 |
+
"step": 345
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.34330554193231977,
|
503 |
+
"grad_norm": 1.9574848872158568,
|
504 |
+
"learning_rate": 8.301287894091812e-06,
|
505 |
+
"loss": 0.5345,
|
506 |
+
"step": 350
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.3482099068170672,
|
510 |
+
"grad_norm": 2.046001716842364,
|
511 |
+
"learning_rate": 8.236481207152539e-06,
|
512 |
+
"loss": 0.5392,
|
513 |
+
"step": 355
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.35311427170181464,
|
517 |
+
"grad_norm": 2.0110023874224257,
|
518 |
+
"learning_rate": 8.170724870256526e-06,
|
519 |
+
"loss": 0.5171,
|
520 |
+
"step": 360
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.358018636586562,
|
524 |
+
"grad_norm": 1.8982030843350457,
|
525 |
+
"learning_rate": 8.104038177661484e-06,
|
526 |
+
"loss": 0.5245,
|
527 |
+
"step": 365
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.36292300147130946,
|
531 |
+
"grad_norm": 1.9231079403397293,
|
532 |
+
"learning_rate": 8.036440696610566e-06,
|
533 |
+
"loss": 0.52,
|
534 |
+
"step": 370
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.3678273663560569,
|
538 |
+
"grad_norm": 1.9562337288746108,
|
539 |
+
"learning_rate": 7.967952261590936e-06,
|
540 |
+
"loss": 0.5087,
|
541 |
+
"step": 375
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.37273173124080433,
|
545 |
+
"grad_norm": 1.9474638682438907,
|
546 |
+
"learning_rate": 7.898592968513919e-06,
|
547 |
+
"loss": 0.5085,
|
548 |
+
"step": 380
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.37763609612555177,
|
552 |
+
"grad_norm": 1.9123482797519735,
|
553 |
+
"learning_rate": 7.828383168818457e-06,
|
554 |
+
"loss": 0.5131,
|
555 |
+
"step": 385
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.38254046101029915,
|
559 |
+
"grad_norm": 2.057943038506519,
|
560 |
+
"learning_rate": 7.757343463499577e-06,
|
561 |
+
"loss": 0.4981,
|
562 |
+
"step": 390
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.3874448258950466,
|
566 |
+
"grad_norm": 1.9838558136826598,
|
567 |
+
"learning_rate": 7.685494697063627e-06,
|
568 |
+
"loss": 0.5158,
|
569 |
+
"step": 395
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.392349190779794,
|
573 |
+
"grad_norm": 1.9089161018582137,
|
574 |
+
"learning_rate": 7.612857951412085e-06,
|
575 |
+
"loss": 0.5115,
|
576 |
+
"step": 400
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.39725355566454146,
|
580 |
+
"grad_norm": 2.053508845540271,
|
581 |
+
"learning_rate": 7.5394545396556864e-06,
|
582 |
+
"loss": 0.4983,
|
583 |
+
"step": 405
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.4021579205492889,
|
587 |
+
"grad_norm": 2.0381945487707225,
|
588 |
+
"learning_rate": 7.465305999860728e-06,
|
589 |
+
"loss": 0.4864,
|
590 |
+
"step": 410
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.4070622854340363,
|
594 |
+
"grad_norm": 2.060283387944364,
|
595 |
+
"learning_rate": 7.390434088729348e-06,
|
596 |
+
"loss": 0.4858,
|
597 |
+
"step": 415
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.4119666503187837,
|
601 |
+
"grad_norm": 2.108098690183231,
|
602 |
+
"learning_rate": 7.314860775215674e-06,
|
603 |
+
"loss": 0.4894,
|
604 |
+
"step": 420
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.41687101520353115,
|
608 |
+
"grad_norm": 1.9523420077195515,
|
609 |
+
"learning_rate": 7.2386082340796715e-06,
|
610 |
+
"loss": 0.5032,
|
611 |
+
"step": 425
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.4217753800882786,
|
615 |
+
"grad_norm": 1.9725770065321593,
|
616 |
+
"learning_rate": 7.1616988393806245e-06,
|
617 |
+
"loss": 0.4917,
|
618 |
+
"step": 430
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.426679744973026,
|
622 |
+
"grad_norm": 1.989450857443718,
|
623 |
+
"learning_rate": 7.0841551579121144e-06,
|
624 |
+
"loss": 0.488,
|
625 |
+
"step": 435
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.4315841098577734,
|
629 |
+
"grad_norm": 1.874422665252578,
|
630 |
+
"learning_rate": 7.005999942580478e-06,
|
631 |
+
"loss": 0.4871,
|
632 |
+
"step": 440
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.43648847474252084,
|
636 |
+
"grad_norm": 1.9628105831400913,
|
637 |
+
"learning_rate": 6.927256125728624e-06,
|
638 |
+
"loss": 0.4774,
|
639 |
+
"step": 445
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.4413928396272683,
|
643 |
+
"grad_norm": 2.052494969259568,
|
644 |
+
"learning_rate": 6.8479468124072146e-06,
|
645 |
+
"loss": 0.4846,
|
646 |
+
"step": 450
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.4462972045120157,
|
650 |
+
"grad_norm": 2.196992014411905,
|
651 |
+
"learning_rate": 6.768095273595176e-06,
|
652 |
+
"loss": 0.4761,
|
653 |
+
"step": 455
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.45120156939676315,
|
657 |
+
"grad_norm": 2.140187037948035,
|
658 |
+
"learning_rate": 6.6877249393715115e-06,
|
659 |
+
"loss": 0.4716,
|
660 |
+
"step": 460
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.4561059342815105,
|
664 |
+
"grad_norm": 2.165318448275331,
|
665 |
+
"learning_rate": 6.60685939204044e-06,
|
666 |
+
"loss": 0.462,
|
667 |
+
"step": 465
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.46101029916625796,
|
671 |
+
"grad_norm": 2.0062687355632485,
|
672 |
+
"learning_rate": 6.525522359211858e-06,
|
673 |
+
"loss": 0.4592,
|
674 |
+
"step": 470
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.4659146640510054,
|
678 |
+
"grad_norm": 1.9449243429974221,
|
679 |
+
"learning_rate": 6.443737706839175e-06,
|
680 |
+
"loss": 0.4662,
|
681 |
+
"step": 475
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.47081902893575284,
|
685 |
+
"grad_norm": 2.0062916535890816,
|
686 |
+
"learning_rate": 6.36152943221656e-06,
|
687 |
+
"loss": 0.4618,
|
688 |
+
"step": 480
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.4757233938205002,
|
692 |
+
"grad_norm": 1.9536933195743733,
|
693 |
+
"learning_rate": 6.278921656937631e-06,
|
694 |
+
"loss": 0.4586,
|
695 |
+
"step": 485
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.48062775870524765,
|
699 |
+
"grad_norm": 2.1253405020175706,
|
700 |
+
"learning_rate": 6.195938619817694e-06,
|
701 |
+
"loss": 0.4643,
|
702 |
+
"step": 490
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.4855321235899951,
|
706 |
+
"grad_norm": 2.0546536638691695,
|
707 |
+
"learning_rate": 6.112604669781572e-06,
|
708 |
+
"loss": 0.4553,
|
709 |
+
"step": 495
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.4904364884747425,
|
713 |
+
"grad_norm": 1.9563960276975647,
|
714 |
+
"learning_rate": 6.0289442587191405e-06,
|
715 |
+
"loss": 0.4537,
|
716 |
+
"step": 500
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.49534085335948996,
|
720 |
+
"grad_norm": 1.9189930888442277,
|
721 |
+
"learning_rate": 5.944981934310627e-06,
|
722 |
+
"loss": 0.4555,
|
723 |
+
"step": 505
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.5002452182442374,
|
727 |
+
"grad_norm": 1.8664708836063784,
|
728 |
+
"learning_rate": 5.860742332823831e-06,
|
729 |
+
"loss": 0.4515,
|
730 |
+
"step": 510
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.5051495831289848,
|
734 |
+
"grad_norm": 1.9544083422711673,
|
735 |
+
"learning_rate": 5.776250171885329e-06,
|
736 |
+
"loss": 0.447,
|
737 |
+
"step": 515
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.5100539480137323,
|
741 |
+
"grad_norm": 2.0179051671028385,
|
742 |
+
"learning_rate": 5.691530243227824e-06,
|
743 |
+
"loss": 0.4386,
|
744 |
+
"step": 520
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.5149583128984796,
|
748 |
+
"grad_norm": 2.0846338532033752,
|
749 |
+
"learning_rate": 5.6066074054157385e-06,
|
750 |
+
"loss": 0.4355,
|
751 |
+
"step": 525
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.519862677783227,
|
755 |
+
"grad_norm": 2.0263314582631153,
|
756 |
+
"learning_rate": 5.521506576551196e-06,
|
757 |
+
"loss": 0.4401,
|
758 |
+
"step": 530
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.5247670426679745,
|
762 |
+
"grad_norm": 1.8778135562708458,
|
763 |
+
"learning_rate": 5.436252726962553e-06,
|
764 |
+
"loss": 0.4341,
|
765 |
+
"step": 535
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.5296714075527219,
|
769 |
+
"grad_norm": 1.9757916591258975,
|
770 |
+
"learning_rate": 5.350870871877577e-06,
|
771 |
+
"loss": 0.4364,
|
772 |
+
"step": 540
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.5345757724374693,
|
776 |
+
"grad_norm": 1.9531961318347626,
|
777 |
+
"learning_rate": 5.265386064083481e-06,
|
778 |
+
"loss": 0.4323,
|
779 |
+
"step": 545
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.5394801373222168,
|
783 |
+
"grad_norm": 2.0725678291543628,
|
784 |
+
"learning_rate": 5.179823386575908e-06,
|
785 |
+
"loss": 0.4364,
|
786 |
+
"step": 550
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.5443845022069642,
|
790 |
+
"grad_norm": 1.860156556204824,
|
791 |
+
"learning_rate": 5.09420794519907e-06,
|
792 |
+
"loss": 0.4329,
|
793 |
+
"step": 555
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.5492888670917117,
|
797 |
+
"grad_norm": 1.8698424821946518,
|
798 |
+
"learning_rate": 5.008564861279188e-06,
|
799 |
+
"loss": 0.4143,
|
800 |
+
"step": 560
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.5541932319764591,
|
804 |
+
"grad_norm": 1.9655955550507962,
|
805 |
+
"learning_rate": 4.922919264253368e-06,
|
806 |
+
"loss": 0.4248,
|
807 |
+
"step": 565
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.5590975968612065,
|
811 |
+
"grad_norm": 1.9066406300773608,
|
812 |
+
"learning_rate": 4.837296284296113e-06,
|
813 |
+
"loss": 0.4186,
|
814 |
+
"step": 570
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.5640019617459539,
|
818 |
+
"grad_norm": 1.80977031179867,
|
819 |
+
"learning_rate": 4.75172104494561e-06,
|
820 |
+
"loss": 0.4156,
|
821 |
+
"step": 575
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.5689063266307013,
|
825 |
+
"grad_norm": 1.991595721949942,
|
826 |
+
"learning_rate": 4.666218655731981e-06,
|
827 |
+
"loss": 0.4156,
|
828 |
+
"step": 580
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.5738106915154487,
|
832 |
+
"grad_norm": 1.910239431707766,
|
833 |
+
"learning_rate": 4.580814204809618e-06,
|
834 |
+
"loss": 0.3942,
|
835 |
+
"step": 585
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.5787150564001962,
|
839 |
+
"grad_norm": 1.9231788230679294,
|
840 |
+
"learning_rate": 4.495532751595813e-06,
|
841 |
+
"loss": 0.4131,
|
842 |
+
"step": 590
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.5836194212849436,
|
846 |
+
"grad_norm": 1.903193102248486,
|
847 |
+
"learning_rate": 4.410399319417806e-06,
|
848 |
+
"loss": 0.4128,
|
849 |
+
"step": 595
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.588523786169691,
|
853 |
+
"grad_norm": 1.8468443833129051,
|
854 |
+
"learning_rate": 4.325438888170429e-06,
|
855 |
+
"loss": 0.4007,
|
856 |
+
"step": 600
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.5934281510544385,
|
860 |
+
"grad_norm": 1.8391801455451426,
|
861 |
+
"learning_rate": 4.2406763869864965e-06,
|
862 |
+
"loss": 0.4127,
|
863 |
+
"step": 605
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.5983325159391859,
|
867 |
+
"grad_norm": 1.9158169069206314,
|
868 |
+
"learning_rate": 4.156136686922083e-06,
|
869 |
+
"loss": 0.4102,
|
870 |
+
"step": 610
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.6032368808239333,
|
874 |
+
"grad_norm": 1.8336931901990852,
|
875 |
+
"learning_rate": 4.071844593658841e-06,
|
876 |
+
"loss": 0.3978,
|
877 |
+
"step": 615
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.6081412457086808,
|
881 |
+
"grad_norm": 1.9026371408899738,
|
882 |
+
"learning_rate": 3.987824840225512e-06,
|
883 |
+
"loss": 0.4009,
|
884 |
+
"step": 620
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.6130456105934281,
|
888 |
+
"grad_norm": 1.9012934657121805,
|
889 |
+
"learning_rate": 3.904102079740753e-06,
|
890 |
+
"loss": 0.3923,
|
891 |
+
"step": 625
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.6179499754781755,
|
895 |
+
"grad_norm": 1.8972920333300498,
|
896 |
+
"learning_rate": 3.820700878179389e-06,
|
897 |
+
"loss": 0.3894,
|
898 |
+
"step": 630
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.622854340362923,
|
902 |
+
"grad_norm": 1.8312865024256686,
|
903 |
+
"learning_rate": 3.73764570716427e-06,
|
904 |
+
"loss": 0.3822,
|
905 |
+
"step": 635
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.6277587052476704,
|
909 |
+
"grad_norm": 1.9183680244376244,
|
910 |
+
"learning_rate": 3.654960936785783e-06,
|
911 |
+
"loss": 0.3926,
|
912 |
+
"step": 640
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.6326630701324178,
|
916 |
+
"grad_norm": 1.828330870696398,
|
917 |
+
"learning_rate": 3.572670828451177e-06,
|
918 |
+
"loss": 0.3924,
|
919 |
+
"step": 645
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.6375674350171653,
|
923 |
+
"grad_norm": 1.8312336733506578,
|
924 |
+
"learning_rate": 3.4907995277657624e-06,
|
925 |
+
"loss": 0.3984,
|
926 |
+
"step": 650
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.6424717999019127,
|
930 |
+
"grad_norm": 2.0339474627749436,
|
931 |
+
"learning_rate": 3.4093710574480926e-06,
|
932 |
+
"loss": 0.3737,
|
933 |
+
"step": 655
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.6473761647866602,
|
937 |
+
"grad_norm": 1.8512202300281744,
|
938 |
+
"learning_rate": 3.3284093102812144e-06,
|
939 |
+
"loss": 0.3896,
|
940 |
+
"step": 660
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.6522805296714076,
|
944 |
+
"grad_norm": 1.935099436122936,
|
945 |
+
"learning_rate": 3.2479380421020336e-06,
|
946 |
+
"loss": 0.3744,
|
947 |
+
"step": 665
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.657184894556155,
|
951 |
+
"grad_norm": 1.877516062960988,
|
952 |
+
"learning_rate": 3.167980864830855e-06,
|
953 |
+
"loss": 0.3872,
|
954 |
+
"step": 670
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.6620892594409024,
|
958 |
+
"grad_norm": 1.8575937601231947,
|
959 |
+
"learning_rate": 3.0885612395431765e-06,
|
960 |
+
"loss": 0.3811,
|
961 |
+
"step": 675
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.6669936243256498,
|
965 |
+
"grad_norm": 1.799905567165699,
|
966 |
+
"learning_rate": 3.009702469585713e-06,
|
967 |
+
"loss": 0.3793,
|
968 |
+
"step": 680
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.6718979892103972,
|
972 |
+
"grad_norm": 1.9383699331479365,
|
973 |
+
"learning_rate": 2.93142769373873e-06,
|
974 |
+
"loss": 0.3712,
|
975 |
+
"step": 685
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.6768023540951447,
|
979 |
+
"grad_norm": 1.9008755304822846,
|
980 |
+
"learning_rate": 2.853759879426644e-06,
|
981 |
+
"loss": 0.3738,
|
982 |
+
"step": 690
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.6817067189798921,
|
986 |
+
"grad_norm": 1.8531666768609951,
|
987 |
+
"learning_rate": 2.7767218159789067e-06,
|
988 |
+
"loss": 0.3619,
|
989 |
+
"step": 695
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.6866110838646395,
|
993 |
+
"grad_norm": 1.9787710274083234,
|
994 |
+
"learning_rate": 2.7003361079431547e-06,
|
995 |
+
"loss": 0.3733,
|
996 |
+
"step": 700
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.691515448749387,
|
1000 |
+
"grad_norm": 1.797813113073451,
|
1001 |
+
"learning_rate": 2.624625168452568e-06,
|
1002 |
+
"loss": 0.3762,
|
1003 |
+
"step": 705
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.6964198136341344,
|
1007 |
+
"grad_norm": 1.9305643695542356,
|
1008 |
+
"learning_rate": 2.5496112126493995e-06,
|
1009 |
+
"loss": 0.3712,
|
1010 |
+
"step": 710
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.7013241785188818,
|
1014 |
+
"grad_norm": 1.7398004529962572,
|
1015 |
+
"learning_rate": 2.4753162511665936e-06,
|
1016 |
+
"loss": 0.366,
|
1017 |
+
"step": 715
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.7062285434036293,
|
1021 |
+
"grad_norm": 2.0252060555010902,
|
1022 |
+
"learning_rate": 2.401762083669419e-06,
|
1023 |
+
"loss": 0.3626,
|
1024 |
+
"step": 720
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.7111329082883766,
|
1028 |
+
"grad_norm": 1.7500094335967311,
|
1029 |
+
"learning_rate": 2.3289702924589914e-06,
|
1030 |
+
"loss": 0.3624,
|
1031 |
+
"step": 725
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.716037273173124,
|
1035 |
+
"grad_norm": 1.752557189300548,
|
1036 |
+
"learning_rate": 2.256962236139598e-06,
|
1037 |
+
"loss": 0.3677,
|
1038 |
+
"step": 730
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.7209416380578715,
|
1042 |
+
"grad_norm": 1.784599164157859,
|
1043 |
+
"learning_rate": 2.18575904335163e-06,
|
1044 |
+
"loss": 0.3647,
|
1045 |
+
"step": 735
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.7258460029426189,
|
1049 |
+
"grad_norm": 1.9166229348625452,
|
1050 |
+
"learning_rate": 2.115381606572018e-06,
|
1051 |
+
"loss": 0.3614,
|
1052 |
+
"step": 740
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.7307503678273664,
|
1056 |
+
"grad_norm": 1.7664227370032382,
|
1057 |
+
"learning_rate": 2.0458505759839433e-06,
|
1058 |
+
"loss": 0.3539,
|
1059 |
+
"step": 745
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.7356547327121138,
|
1063 |
+
"grad_norm": 1.7823664294805341,
|
1064 |
+
"learning_rate": 1.9771863534176544e-06,
|
1065 |
+
"loss": 0.3649,
|
1066 |
+
"step": 750
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.7405590975968612,
|
1070 |
+
"grad_norm": 1.798651877731334,
|
1071 |
+
"learning_rate": 1.90940908636415e-06,
|
1072 |
+
"loss": 0.3584,
|
1073 |
+
"step": 755
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.7454634624816087,
|
1077 |
+
"grad_norm": 1.7897055029417046,
|
1078 |
+
"learning_rate": 1.8425386620634961e-06,
|
1079 |
+
"loss": 0.3575,
|
1080 |
+
"step": 760
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.7503678273663561,
|
1084 |
+
"grad_norm": 1.8567965533511697,
|
1085 |
+
"learning_rate": 1.7765947016694902e-06,
|
1086 |
+
"loss": 0.3597,
|
1087 |
+
"step": 765
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.7552721922511035,
|
1091 |
+
"grad_norm": 1.7069946121225148,
|
1092 |
+
"learning_rate": 1.711596554492428e-06,
|
1093 |
+
"loss": 0.3569,
|
1094 |
+
"step": 770
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.7601765571358509,
|
1098 |
+
"grad_norm": 1.7171226071674441,
|
1099 |
+
"learning_rate": 1.64756329232161e-06,
|
1100 |
+
"loss": 0.3508,
|
1101 |
+
"step": 775
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.7650809220205983,
|
1105 |
+
"grad_norm": 1.851171218498648,
|
1106 |
+
"learning_rate": 1.5845137038292851e-06,
|
1107 |
+
"loss": 0.3505,
|
1108 |
+
"step": 780
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.7699852869053457,
|
1112 |
+
"grad_norm": 1.7875922878269628,
|
1113 |
+
"learning_rate": 1.5224662890576781e-06,
|
1114 |
+
"loss": 0.3404,
|
1115 |
+
"step": 785
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.7748896517900932,
|
1119 |
+
"grad_norm": 1.6638468790620988,
|
1120 |
+
"learning_rate": 1.4614392539906892e-06,
|
1121 |
+
"loss": 0.3522,
|
1122 |
+
"step": 790
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.7797940166748406,
|
1126 |
+
"grad_norm": 1.7409632860061264,
|
1127 |
+
"learning_rate": 1.4014505052118893e-06,
|
1128 |
+
"loss": 0.353,
|
1129 |
+
"step": 795
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.784698381559588,
|
1133 |
+
"grad_norm": 1.815807330350682,
|
1134 |
+
"learning_rate": 1.3425176446503618e-06,
|
1135 |
+
"loss": 0.3414,
|
1136 |
+
"step": 800
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.7896027464443355,
|
1140 |
+
"grad_norm": 1.8126274287911948,
|
1141 |
+
"learning_rate": 1.2846579644159291e-06,
|
1142 |
+
"loss": 0.3425,
|
1143 |
+
"step": 805
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.7945071113290829,
|
1147 |
+
"grad_norm": 1.7603201856240744,
|
1148 |
+
"learning_rate": 1.2278884417253033e-06,
|
1149 |
+
"loss": 0.3453,
|
1150 |
+
"step": 810
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.7994114762138304,
|
1154 |
+
"grad_norm": 1.7250692746456593,
|
1155 |
+
"learning_rate": 1.172225733920616e-06,
|
1156 |
+
"loss": 0.3456,
|
1157 |
+
"step": 815
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.8043158410985778,
|
1161 |
+
"grad_norm": 1.6574970310295125,
|
1162 |
+
"learning_rate": 1.1176861735818107e-06,
|
1163 |
+
"loss": 0.3357,
|
1164 |
+
"step": 820
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.8092202059833251,
|
1168 |
+
"grad_norm": 1.7019799673778844,
|
1169 |
+
"learning_rate": 1.0642857637343346e-06,
|
1170 |
+
"loss": 0.3406,
|
1171 |
+
"step": 825
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.8141245708680726,
|
1175 |
+
"grad_norm": 1.6768204828704758,
|
1176 |
+
"learning_rate": 1.0120401731535213e-06,
|
1177 |
+
"loss": 0.353,
|
1178 |
+
"step": 830
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.81902893575282,
|
1182 |
+
"grad_norm": 1.6058567769661107,
|
1183 |
+
"learning_rate": 9.609647317670468e-07,
|
1184 |
+
"loss": 0.3413,
|
1185 |
+
"step": 835
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.8239333006375674,
|
1189 |
+
"grad_norm": 1.722906282315617,
|
1190 |
+
"learning_rate": 9.110744261568206e-07,
|
1191 |
+
"loss": 0.3329,
|
1192 |
+
"step": 840
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.8288376655223149,
|
1196 |
+
"grad_norm": 1.722830977466264,
|
1197 |
+
"learning_rate": 8.623838951616076e-07,
|
1198 |
+
"loss": 0.3339,
|
1199 |
+
"step": 845
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.8337420304070623,
|
1203 |
+
"grad_norm": 1.7043628706248817,
|
1204 |
+
"learning_rate": 8.149074255816996e-07,
|
1205 |
+
"loss": 0.3327,
|
1206 |
+
"step": 850
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.8386463952918097,
|
1210 |
+
"grad_norm": 1.6620351875279176,
|
1211 |
+
"learning_rate": 7.68658947986874e-07,
|
1212 |
+
"loss": 0.3409,
|
1213 |
+
"step": 855
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.8435507601765572,
|
1217 |
+
"grad_norm": 1.6249652725084447,
|
1218 |
+
"learning_rate": 7.236520326288721e-07,
|
1219 |
+
"loss": 0.3345,
|
1220 |
+
"step": 860
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.8484551250613046,
|
1224 |
+
"grad_norm": 1.6797312028896152,
|
1225 |
+
"learning_rate": 6.79899885459619e-07,
|
1226 |
+
"loss": 0.3371,
|
1227 |
+
"step": 865
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.853359489946052,
|
1231 |
+
"grad_norm": 1.6133846444360203,
|
1232 |
+
"learning_rate": 6.374153442563192e-07,
|
1233 |
+
"loss": 0.3291,
|
1234 |
+
"step": 870
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.8582638548307994,
|
1238 |
+
"grad_norm": 1.639961696466558,
|
1239 |
+
"learning_rate": 5.962108748545942e-07,
|
1240 |
+
"loss": 0.3405,
|
1241 |
+
"step": 875
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.8631682197155468,
|
1245 |
+
"grad_norm": 1.7763694537319363,
|
1246 |
+
"learning_rate": 5.562985674907467e-07,
|
1247 |
+
"loss": 0.3377,
|
1248 |
+
"step": 880
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.8680725846002942,
|
1252 |
+
"grad_norm": 1.7034248324336427,
|
1253 |
+
"learning_rate": 5.176901332542378e-07,
|
1254 |
+
"loss": 0.3406,
|
1255 |
+
"step": 885
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.8729769494850417,
|
1259 |
+
"grad_norm": 1.6095220015390743,
|
1260 |
+
"learning_rate": 4.803969006514175e-07,
|
1261 |
+
"loss": 0.33,
|
1262 |
+
"step": 890
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.8778813143697891,
|
1266 |
+
"grad_norm": 1.7111560907547738,
|
1267 |
+
"learning_rate": 4.444298122815055e-07,
|
1268 |
+
"loss": 0.335,
|
1269 |
+
"step": 895
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.8827856792545365,
|
1273 |
+
"grad_norm": 1.6246588106178113,
|
1274 |
+
"learning_rate": 4.0979942162580387e-07,
|
1275 |
+
"loss": 0.3289,
|
1276 |
+
"step": 900
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.887690044139284,
|
1280 |
+
"grad_norm": 1.6674759393718703,
|
1281 |
+
"learning_rate": 3.76515889951099e-07,
|
1282 |
+
"loss": 0.3287,
|
1283 |
+
"step": 905
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.8925944090240314,
|
1287 |
+
"grad_norm": 1.6046692843902606,
|
1288 |
+
"learning_rate": 3.445889833281296e-07,
|
1289 |
+
"loss": 0.3324,
|
1290 |
+
"step": 910
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.8974987739087789,
|
1294 |
+
"grad_norm": 1.7079729479541892,
|
1295 |
+
"learning_rate": 3.140280697660247e-07,
|
1296 |
+
"loss": 0.3258,
|
1297 |
+
"step": 915
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.9024031387935263,
|
1301 |
+
"grad_norm": 1.6511127606613865,
|
1302 |
+
"learning_rate": 2.8484211646353677e-07,
|
1303 |
+
"loss": 0.3266,
|
1304 |
+
"step": 920
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.9073075036782736,
|
1308 |
+
"grad_norm": 1.6108380967394342,
|
1309 |
+
"learning_rate": 2.570396871778796e-07,
|
1310 |
+
"loss": 0.3285,
|
1311 |
+
"step": 925
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.912211868563021,
|
1315 |
+
"grad_norm": 1.7501453734514094,
|
1316 |
+
"learning_rate": 2.3062893971195211e-07,
|
1317 |
+
"loss": 0.3299,
|
1318 |
+
"step": 930
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.9171162334477685,
|
1322 |
+
"grad_norm": 1.7120576016328979,
|
1323 |
+
"learning_rate": 2.0561762352066638e-07,
|
1324 |
+
"loss": 0.3261,
|
1325 |
+
"step": 935
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.9220205983325159,
|
1329 |
+
"grad_norm": 1.6683684954483,
|
1330 |
+
"learning_rate": 1.8201307743709927e-07,
|
1331 |
+
"loss": 0.328,
|
1332 |
+
"step": 940
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.9269249632172634,
|
1336 |
+
"grad_norm": 1.6981337204644897,
|
1337 |
+
"learning_rate": 1.5982222751913079e-07,
|
1338 |
+
"loss": 0.331,
|
1339 |
+
"step": 945
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.9318293281020108,
|
1343 |
+
"grad_norm": 1.6609453351002048,
|
1344 |
+
"learning_rate": 1.390515850171953e-07,
|
1345 |
+
"loss": 0.3234,
|
1346 |
+
"step": 950
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.9367336929867582,
|
1350 |
+
"grad_norm": 1.6636906278300567,
|
1351 |
+
"learning_rate": 1.1970724446374592e-07,
|
1352 |
+
"loss": 0.3336,
|
1353 |
+
"step": 955
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.9416380578715057,
|
1357 |
+
"grad_norm": 1.5711270056253417,
|
1358 |
+
"learning_rate": 1.0179488188499675e-07,
|
1359 |
+
"loss": 0.3299,
|
1360 |
+
"step": 960
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.9465424227562531,
|
1364 |
+
"grad_norm": 1.6157567675118045,
|
1365 |
+
"learning_rate": 8.531975313545715e-08,
|
1366 |
+
"loss": 0.3398,
|
1367 |
+
"step": 965
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.9514467876410004,
|
1371 |
+
"grad_norm": 1.6523867015186364,
|
1372 |
+
"learning_rate": 7.028669235575714e-08,
|
1373 |
+
"loss": 0.3294,
|
1374 |
+
"step": 970
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.9563511525257479,
|
1378 |
+
"grad_norm": 1.740288361114131,
|
1379 |
+
"learning_rate": 5.670011055421365e-08,
|
1380 |
+
"loss": 0.3335,
|
1381 |
+
"step": 975
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.9612555174104953,
|
1385 |
+
"grad_norm": 1.621637272095815,
|
1386 |
+
"learning_rate": 4.4563994312546435e-08,
|
1387 |
+
"loss": 0.3295,
|
1388 |
+
"step": 980
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.9661598822952427,
|
1392 |
+
"grad_norm": 1.657618038769045,
|
1393 |
+
"learning_rate": 3.3881904616137054e-08,
|
1394 |
+
"loss": 0.3266,
|
1395 |
+
"step": 985
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.9710642471799902,
|
1399 |
+
"grad_norm": 1.6751008606908966,
|
1400 |
+
"learning_rate": 2.4656975809160267e-08,
|
1401 |
+
"loss": 0.3315,
|
1402 |
+
"step": 990
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.9759686120647376,
|
1406 |
+
"grad_norm": 1.6937145693196844,
|
1407 |
+
"learning_rate": 1.689191467490303e-08,
|
1408 |
+
"loss": 0.3313,
|
1409 |
+
"step": 995
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.980872976949485,
|
1413 |
+
"grad_norm": 1.6407710130512811,
|
1414 |
+
"learning_rate": 1.058899964154092e-08,
|
1415 |
+
"loss": 0.3278,
|
1416 |
+
"step": 1000
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.9857773418342325,
|
1420 |
+
"grad_norm": 1.6749685115327173,
|
1421 |
+
"learning_rate": 5.750080113598455e-09,
|
1422 |
+
"loss": 0.3246,
|
1423 |
+
"step": 1005
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.9906817067189799,
|
1427 |
+
"grad_norm": 1.5703580183454553,
|
1428 |
+
"learning_rate": 2.376575929297076e-09,
|
1429 |
+
"loss": 0.3257,
|
1430 |
+
"step": 1010
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.9955860716037274,
|
1434 |
+
"grad_norm": 1.560579334168069,
|
1435 |
+
"learning_rate": 4.694769439445024e-10,
|
1436 |
+
"loss": 0.3258,
|
1437 |
+
"step": 1015
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.9995095635115253,
|
1441 |
+
"eval_loss": 0.33406102657318115,
|
1442 |
+
"eval_runtime": 96.9695,
|
1443 |
+
"eval_samples_per_second": 3.114,
|
1444 |
+
"eval_steps_per_second": 0.784,
|
1445 |
+
"step": 1019
|
1446 |
+
},
|
1447 |
+
{
|
1448 |
+
"epoch": 0.9995095635115253,
|
1449 |
+
"step": 1019,
|
1450 |
+
"total_flos": 213305524224000.0,
|
1451 |
+
"train_loss": 0.4876700218573169,
|
1452 |
+
"train_runtime": 22910.3839,
|
1453 |
+
"train_samples_per_second": 1.424,
|
1454 |
+
"train_steps_per_second": 0.044
|
1455 |
+
}
|
1456 |
+
],
|
1457 |
+
"logging_steps": 5,
|
1458 |
+
"max_steps": 1019,
|
1459 |
+
"num_input_tokens_seen": 0,
|
1460 |
+
"num_train_epochs": 1,
|
1461 |
+
"save_steps": 100,
|
1462 |
+
"stateful_callbacks": {
|
1463 |
+
"TrainerControl": {
|
1464 |
+
"args": {
|
1465 |
+
"should_epoch_stop": false,
|
1466 |
+
"should_evaluate": false,
|
1467 |
+
"should_log": false,
|
1468 |
+
"should_save": true,
|
1469 |
+
"should_training_stop": true
|
1470 |
+
},
|
1471 |
+
"attributes": {}
|
1472 |
+
}
|
1473 |
+
},
|
1474 |
+
"total_flos": 213305524224000.0,
|
1475 |
+
"train_batch_size": 2,
|
1476 |
+
"trial_name": null,
|
1477 |
+
"trial_params": null
|
1478 |
+
}
|