Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- train_results.json +9 -0
- trainer_state.json +2660 -0
README.md
ADDED
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: transformers
|
3 |
+
license: apache-2.0
|
4 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
5 |
+
tags:
|
6 |
+
- trl
|
7 |
+
- sft
|
8 |
+
- generated_from_trainer
|
9 |
+
model-index:
|
10 |
+
- name: feedback_p0.1_seed42_level2_syntaxcleanbatch16
|
11 |
+
results: []
|
12 |
+
---
|
13 |
+
|
14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
+
should probably proofread and complete it, then remove this comment. -->
|
16 |
+
|
17 |
+
# feedback_p0.1_seed42_level2_syntaxcleanbatch16
|
18 |
+
|
19 |
+
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 None dataset.
|
20 |
+
|
21 |
+
## Model description
|
22 |
+
|
23 |
+
More information needed
|
24 |
+
|
25 |
+
## Intended uses & limitations
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Training and evaluation data
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training procedure
|
34 |
+
|
35 |
+
### Training hyperparameters
|
36 |
+
|
37 |
+
The following hyperparameters were used during training:
|
38 |
+
- learning_rate: 1e-05
|
39 |
+
- train_batch_size: 4
|
40 |
+
- eval_batch_size: 1
|
41 |
+
- seed: 42
|
42 |
+
- distributed_type: multi-GPU
|
43 |
+
- num_devices: 4
|
44 |
+
- total_train_batch_size: 16
|
45 |
+
- total_eval_batch_size: 4
|
46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
+
- lr_scheduler_type: cosine
|
48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
49 |
+
- num_epochs: 1
|
50 |
+
|
51 |
+
### Training results
|
52 |
+
|
53 |
+
|
54 |
+
|
55 |
+
### Framework versions
|
56 |
+
|
57 |
+
- Transformers 4.44.2
|
58 |
+
- Pytorch 2.3.1+cu121
|
59 |
+
- Datasets 2.19.1
|
60 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 194932459438080.0,
|
4 |
+
"train_loss": 0.5118633475710831,
|
5 |
+
"train_runtime": 20414.4633,
|
6 |
+
"train_samples": 98932,
|
7 |
+
"train_samples_per_second": 1.459,
|
8 |
+
"train_steps_per_second": 0.091
|
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": "4.44.2"
|
6 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 194932459438080.0,
|
4 |
+
"train_loss": 0.5118633475710831,
|
5 |
+
"train_runtime": 20414.4633,
|
6 |
+
"train_samples": 98932,
|
7 |
+
"train_samples_per_second": 1.459,
|
8 |
+
"train_steps_per_second": 0.091
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,2660 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1862,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0005370569280343716,
|
13 |
+
"grad_norm": 24.25113947879833,
|
14 |
+
"learning_rate": 5.3475935828877005e-08,
|
15 |
+
"loss": 1.3237,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.0026852846401718583,
|
20 |
+
"grad_norm": 23.161748165504072,
|
21 |
+
"learning_rate": 2.6737967914438503e-07,
|
22 |
+
"loss": 1.3357,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.0053705692803437165,
|
27 |
+
"grad_norm": 15.788644713919888,
|
28 |
+
"learning_rate": 5.347593582887701e-07,
|
29 |
+
"loss": 1.2928,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.008055853920515575,
|
34 |
+
"grad_norm": 12.065360956658548,
|
35 |
+
"learning_rate": 8.021390374331551e-07,
|
36 |
+
"loss": 1.1734,
|
37 |
+
"step": 15
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.010741138560687433,
|
41 |
+
"grad_norm": 8.451481103732192,
|
42 |
+
"learning_rate": 1.0695187165775401e-06,
|
43 |
+
"loss": 1.053,
|
44 |
+
"step": 20
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.01342642320085929,
|
48 |
+
"grad_norm": 3.607181792793628,
|
49 |
+
"learning_rate": 1.3368983957219254e-06,
|
50 |
+
"loss": 0.9479,
|
51 |
+
"step": 25
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.01611170784103115,
|
55 |
+
"grad_norm": 3.4122039231622128,
|
56 |
+
"learning_rate": 1.6042780748663103e-06,
|
57 |
+
"loss": 0.9164,
|
58 |
+
"step": 30
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.018796992481203006,
|
62 |
+
"grad_norm": 2.9894854055400732,
|
63 |
+
"learning_rate": 1.8716577540106954e-06,
|
64 |
+
"loss": 0.8869,
|
65 |
+
"step": 35
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.021482277121374866,
|
69 |
+
"grad_norm": 3.052529317760186,
|
70 |
+
"learning_rate": 2.1390374331550802e-06,
|
71 |
+
"loss": 0.8571,
|
72 |
+
"step": 40
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.024167561761546726,
|
76 |
+
"grad_norm": 2.9434989562495484,
|
77 |
+
"learning_rate": 2.4064171122994653e-06,
|
78 |
+
"loss": 0.857,
|
79 |
+
"step": 45
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.02685284640171858,
|
83 |
+
"grad_norm": 3.0139599082135424,
|
84 |
+
"learning_rate": 2.673796791443851e-06,
|
85 |
+
"loss": 0.8329,
|
86 |
+
"step": 50
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.02953813104189044,
|
90 |
+
"grad_norm": 2.8752390849195306,
|
91 |
+
"learning_rate": 2.9411764705882355e-06,
|
92 |
+
"loss": 0.8183,
|
93 |
+
"step": 55
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.0322234156820623,
|
97 |
+
"grad_norm": 2.9815477478474164,
|
98 |
+
"learning_rate": 3.2085561497326205e-06,
|
99 |
+
"loss": 0.8154,
|
100 |
+
"step": 60
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.03490870032223416,
|
104 |
+
"grad_norm": 3.1961157011348433,
|
105 |
+
"learning_rate": 3.4759358288770056e-06,
|
106 |
+
"loss": 0.8127,
|
107 |
+
"step": 65
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.03759398496240601,
|
111 |
+
"grad_norm": 3.011408828479708,
|
112 |
+
"learning_rate": 3.7433155080213907e-06,
|
113 |
+
"loss": 0.7876,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.040279269602577876,
|
118 |
+
"grad_norm": 3.074730944639507,
|
119 |
+
"learning_rate": 4.010695187165775e-06,
|
120 |
+
"loss": 0.7812,
|
121 |
+
"step": 75
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.04296455424274973,
|
125 |
+
"grad_norm": 3.0954880610087416,
|
126 |
+
"learning_rate": 4.2780748663101604e-06,
|
127 |
+
"loss": 0.7938,
|
128 |
+
"step": 80
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.04564983888292159,
|
132 |
+
"grad_norm": 3.103351806444735,
|
133 |
+
"learning_rate": 4.5454545454545455e-06,
|
134 |
+
"loss": 0.7783,
|
135 |
+
"step": 85
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.04833512352309345,
|
139 |
+
"grad_norm": 3.1872294877792693,
|
140 |
+
"learning_rate": 4.812834224598931e-06,
|
141 |
+
"loss": 0.7573,
|
142 |
+
"step": 90
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.05102040816326531,
|
146 |
+
"grad_norm": 3.1638880990567455,
|
147 |
+
"learning_rate": 5.0802139037433165e-06,
|
148 |
+
"loss": 0.7558,
|
149 |
+
"step": 95
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.05370569280343716,
|
153 |
+
"grad_norm": 3.0115638804221754,
|
154 |
+
"learning_rate": 5.347593582887702e-06,
|
155 |
+
"loss": 0.7537,
|
156 |
+
"step": 100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.05639097744360902,
|
160 |
+
"grad_norm": 3.2941594696139145,
|
161 |
+
"learning_rate": 5.614973262032086e-06,
|
162 |
+
"loss": 0.7515,
|
163 |
+
"step": 105
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.05907626208378088,
|
167 |
+
"grad_norm": 3.3155309676050013,
|
168 |
+
"learning_rate": 5.882352941176471e-06,
|
169 |
+
"loss": 0.7397,
|
170 |
+
"step": 110
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.06176154672395274,
|
174 |
+
"grad_norm": 3.162740571135442,
|
175 |
+
"learning_rate": 6.149732620320856e-06,
|
176 |
+
"loss": 0.7296,
|
177 |
+
"step": 115
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.0644468313641246,
|
181 |
+
"grad_norm": 3.0840588519609016,
|
182 |
+
"learning_rate": 6.417112299465241e-06,
|
183 |
+
"loss": 0.7336,
|
184 |
+
"step": 120
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.06713211600429646,
|
188 |
+
"grad_norm": 3.150117025536966,
|
189 |
+
"learning_rate": 6.684491978609626e-06,
|
190 |
+
"loss": 0.7382,
|
191 |
+
"step": 125
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.06981740064446831,
|
195 |
+
"grad_norm": 2.9252296896598278,
|
196 |
+
"learning_rate": 6.951871657754011e-06,
|
197 |
+
"loss": 0.7211,
|
198 |
+
"step": 130
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.07250268528464017,
|
202 |
+
"grad_norm": 3.0320480967430092,
|
203 |
+
"learning_rate": 7.219251336898396e-06,
|
204 |
+
"loss": 0.73,
|
205 |
+
"step": 135
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.07518796992481203,
|
209 |
+
"grad_norm": 2.874198329021165,
|
210 |
+
"learning_rate": 7.486631016042781e-06,
|
211 |
+
"loss": 0.7434,
|
212 |
+
"step": 140
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.07787325456498388,
|
216 |
+
"grad_norm": 2.9173140627663656,
|
217 |
+
"learning_rate": 7.754010695187166e-06,
|
218 |
+
"loss": 0.7353,
|
219 |
+
"step": 145
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.08055853920515575,
|
223 |
+
"grad_norm": 2.9848439720434294,
|
224 |
+
"learning_rate": 8.02139037433155e-06,
|
225 |
+
"loss": 0.73,
|
226 |
+
"step": 150
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.08324382384532761,
|
230 |
+
"grad_norm": 2.87264895278571,
|
231 |
+
"learning_rate": 8.288770053475937e-06,
|
232 |
+
"loss": 0.7264,
|
233 |
+
"step": 155
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.08592910848549946,
|
237 |
+
"grad_norm": 2.864090437505294,
|
238 |
+
"learning_rate": 8.556149732620321e-06,
|
239 |
+
"loss": 0.716,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.08861439312567132,
|
244 |
+
"grad_norm": 2.9211346467401826,
|
245 |
+
"learning_rate": 8.823529411764707e-06,
|
246 |
+
"loss": 0.7213,
|
247 |
+
"step": 165
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.09129967776584318,
|
251 |
+
"grad_norm": 2.812817629694785,
|
252 |
+
"learning_rate": 9.090909090909091e-06,
|
253 |
+
"loss": 0.7239,
|
254 |
+
"step": 170
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.09398496240601503,
|
258 |
+
"grad_norm": 3.0966947636468323,
|
259 |
+
"learning_rate": 9.358288770053477e-06,
|
260 |
+
"loss": 0.7252,
|
261 |
+
"step": 175
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.0966702470461869,
|
265 |
+
"grad_norm": 2.8482566526889066,
|
266 |
+
"learning_rate": 9.625668449197861e-06,
|
267 |
+
"loss": 0.7111,
|
268 |
+
"step": 180
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.09935553168635876,
|
272 |
+
"grad_norm": 3.103402165074239,
|
273 |
+
"learning_rate": 9.893048128342247e-06,
|
274 |
+
"loss": 0.709,
|
275 |
+
"step": 185
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.10204081632653061,
|
279 |
+
"grad_norm": 2.833371746409837,
|
280 |
+
"learning_rate": 9.999920849895074e-06,
|
281 |
+
"loss": 0.7141,
|
282 |
+
"step": 190
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.10472610096670247,
|
286 |
+
"grad_norm": 2.7601482578055685,
|
287 |
+
"learning_rate": 9.99943716388422e-06,
|
288 |
+
"loss": 0.7091,
|
289 |
+
"step": 195
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.10741138560687433,
|
293 |
+
"grad_norm": 2.7694618688472334,
|
294 |
+
"learning_rate": 9.998513806628825e-06,
|
295 |
+
"loss": 0.6892,
|
296 |
+
"step": 200
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.11009667024704618,
|
300 |
+
"grad_norm": 2.7627170756712545,
|
301 |
+
"learning_rate": 9.997150859332753e-06,
|
302 |
+
"loss": 0.7057,
|
303 |
+
"step": 205
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.11278195488721804,
|
307 |
+
"grad_norm": 2.641313671359486,
|
308 |
+
"learning_rate": 9.995348441859243e-06,
|
309 |
+
"loss": 0.6968,
|
310 |
+
"step": 210
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.11546723952738991,
|
314 |
+
"grad_norm": 2.7210282435142163,
|
315 |
+
"learning_rate": 9.993106712720367e-06,
|
316 |
+
"loss": 0.7126,
|
317 |
+
"step": 215
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.11815252416756176,
|
321 |
+
"grad_norm": 2.6468526042665563,
|
322 |
+
"learning_rate": 9.990425869063085e-06,
|
323 |
+
"loss": 0.7133,
|
324 |
+
"step": 220
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.12083780880773362,
|
328 |
+
"grad_norm": 2.669248971535401,
|
329 |
+
"learning_rate": 9.987306146651908e-06,
|
330 |
+
"loss": 0.6862,
|
331 |
+
"step": 225
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.12352309344790548,
|
335 |
+
"grad_norm": 2.614929201255673,
|
336 |
+
"learning_rate": 9.983747819848168e-06,
|
337 |
+
"loss": 0.6822,
|
338 |
+
"step": 230
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.12620837808807733,
|
342 |
+
"grad_norm": 2.752612192498847,
|
343 |
+
"learning_rate": 9.97975120158589e-06,
|
344 |
+
"loss": 0.7086,
|
345 |
+
"step": 235
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.1288936627282492,
|
349 |
+
"grad_norm": 2.779012267697376,
|
350 |
+
"learning_rate": 9.975316643344257e-06,
|
351 |
+
"loss": 0.7081,
|
352 |
+
"step": 240
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.13157894736842105,
|
356 |
+
"grad_norm": 2.5890743255074278,
|
357 |
+
"learning_rate": 9.970444535116721e-06,
|
358 |
+
"loss": 0.6914,
|
359 |
+
"step": 245
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.13426423200859292,
|
363 |
+
"grad_norm": 2.6285297060534334,
|
364 |
+
"learning_rate": 9.965135305376694e-06,
|
365 |
+
"loss": 0.681,
|
366 |
+
"step": 250
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.13694951664876476,
|
370 |
+
"grad_norm": 2.5527710360428038,
|
371 |
+
"learning_rate": 9.959389421039863e-06,
|
372 |
+
"loss": 0.6912,
|
373 |
+
"step": 255
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.13963480128893663,
|
377 |
+
"grad_norm": 2.665201173590149,
|
378 |
+
"learning_rate": 9.95320738742313e-06,
|
379 |
+
"loss": 0.6911,
|
380 |
+
"step": 260
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.1423200859291085,
|
384 |
+
"grad_norm": 2.57466357634417,
|
385 |
+
"learning_rate": 9.946589748200185e-06,
|
386 |
+
"loss": 0.7051,
|
387 |
+
"step": 265
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.14500537056928034,
|
391 |
+
"grad_norm": 2.512739592787712,
|
392 |
+
"learning_rate": 9.939537085353668e-06,
|
393 |
+
"loss": 0.6789,
|
394 |
+
"step": 270
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.1476906552094522,
|
398 |
+
"grad_norm": 2.5488133338759544,
|
399 |
+
"learning_rate": 9.932050019124011e-06,
|
400 |
+
"loss": 0.6722,
|
401 |
+
"step": 275
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.15037593984962405,
|
405 |
+
"grad_norm": 2.563447287239947,
|
406 |
+
"learning_rate": 9.924129207954877e-06,
|
407 |
+
"loss": 0.68,
|
408 |
+
"step": 280
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.15306122448979592,
|
412 |
+
"grad_norm": 2.6455202812435066,
|
413 |
+
"learning_rate": 9.915775348435266e-06,
|
414 |
+
"loss": 0.6829,
|
415 |
+
"step": 285
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.15574650912996776,
|
419 |
+
"grad_norm": 2.607428689084183,
|
420 |
+
"learning_rate": 9.906989175238232e-06,
|
421 |
+
"loss": 0.6734,
|
422 |
+
"step": 290
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.15843179377013963,
|
426 |
+
"grad_norm": 2.632494678548499,
|
427 |
+
"learning_rate": 9.8977714610563e-06,
|
428 |
+
"loss": 0.6654,
|
429 |
+
"step": 295
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.1611170784103115,
|
433 |
+
"grad_norm": 2.5679386221927163,
|
434 |
+
"learning_rate": 9.888123016533496e-06,
|
435 |
+
"loss": 0.6464,
|
436 |
+
"step": 300
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.16380236305048335,
|
440 |
+
"grad_norm": 2.4455235169022123,
|
441 |
+
"learning_rate": 9.878044690194055e-06,
|
442 |
+
"loss": 0.6808,
|
443 |
+
"step": 305
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.16648764769065522,
|
447 |
+
"grad_norm": 2.6134103392173254,
|
448 |
+
"learning_rate": 9.86753736836781e-06,
|
449 |
+
"loss": 0.6689,
|
450 |
+
"step": 310
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.16917293233082706,
|
454 |
+
"grad_norm": 2.553706613546006,
|
455 |
+
"learning_rate": 9.856601975112227e-06,
|
456 |
+
"loss": 0.6588,
|
457 |
+
"step": 315
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.17185821697099893,
|
461 |
+
"grad_norm": 2.5066521118011407,
|
462 |
+
"learning_rate": 9.84523947213116e-06,
|
463 |
+
"loss": 0.6663,
|
464 |
+
"step": 320
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.17454350161117077,
|
468 |
+
"grad_norm": 2.337716810935506,
|
469 |
+
"learning_rate": 9.833450858690257e-06,
|
470 |
+
"loss": 0.6571,
|
471 |
+
"step": 325
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.17722878625134264,
|
475 |
+
"grad_norm": 2.568589396910675,
|
476 |
+
"learning_rate": 9.821237171529088e-06,
|
477 |
+
"loss": 0.6569,
|
478 |
+
"step": 330
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.1799140708915145,
|
482 |
+
"grad_norm": 2.385921872954393,
|
483 |
+
"learning_rate": 9.808599484769971e-06,
|
484 |
+
"loss": 0.656,
|
485 |
+
"step": 335
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.18259935553168635,
|
489 |
+
"grad_norm": 2.4819007537288362,
|
490 |
+
"learning_rate": 9.79553890982351e-06,
|
491 |
+
"loss": 0.6609,
|
492 |
+
"step": 340
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.18528464017185822,
|
496 |
+
"grad_norm": 2.437061379076581,
|
497 |
+
"learning_rate": 9.782056595290848e-06,
|
498 |
+
"loss": 0.6455,
|
499 |
+
"step": 345
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.18796992481203006,
|
503 |
+
"grad_norm": 2.564231295853269,
|
504 |
+
"learning_rate": 9.768153726862652e-06,
|
505 |
+
"loss": 0.6612,
|
506 |
+
"step": 350
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.19065520945220193,
|
510 |
+
"grad_norm": 2.4893325625782583,
|
511 |
+
"learning_rate": 9.753831527214854e-06,
|
512 |
+
"loss": 0.662,
|
513 |
+
"step": 355
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.1933404940923738,
|
517 |
+
"grad_norm": 2.4552866529124757,
|
518 |
+
"learning_rate": 9.739091255901105e-06,
|
519 |
+
"loss": 0.6486,
|
520 |
+
"step": 360
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.19602577873254565,
|
524 |
+
"grad_norm": 2.5401346090028154,
|
525 |
+
"learning_rate": 9.723934209242015e-06,
|
526 |
+
"loss": 0.6471,
|
527 |
+
"step": 365
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.19871106337271752,
|
531 |
+
"grad_norm": 2.5620972387781253,
|
532 |
+
"learning_rate": 9.708361720211146e-06,
|
533 |
+
"loss": 0.6419,
|
534 |
+
"step": 370
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.20139634801288936,
|
538 |
+
"grad_norm": 2.4951612644213554,
|
539 |
+
"learning_rate": 9.692375158317787e-06,
|
540 |
+
"loss": 0.6492,
|
541 |
+
"step": 375
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.20408163265306123,
|
545 |
+
"grad_norm": 2.4669992078317917,
|
546 |
+
"learning_rate": 9.675975929486512e-06,
|
547 |
+
"loss": 0.6546,
|
548 |
+
"step": 380
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.20676691729323307,
|
552 |
+
"grad_norm": 2.603925161419725,
|
553 |
+
"learning_rate": 9.659165475933537e-06,
|
554 |
+
"loss": 0.6433,
|
555 |
+
"step": 385
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.20945220193340494,
|
559 |
+
"grad_norm": 2.358390538888436,
|
560 |
+
"learning_rate": 9.641945276039885e-06,
|
561 |
+
"loss": 0.6436,
|
562 |
+
"step": 390
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.2121374865735768,
|
566 |
+
"grad_norm": 2.4363415566836184,
|
567 |
+
"learning_rate": 9.624316844221376e-06,
|
568 |
+
"loss": 0.626,
|
569 |
+
"step": 395
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.21482277121374865,
|
573 |
+
"grad_norm": 2.282557445492374,
|
574 |
+
"learning_rate": 9.606281730795435e-06,
|
575 |
+
"loss": 0.6267,
|
576 |
+
"step": 400
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.21750805585392052,
|
580 |
+
"grad_norm": 2.3912481386253654,
|
581 |
+
"learning_rate": 9.587841521844755e-06,
|
582 |
+
"loss": 0.6451,
|
583 |
+
"step": 405
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.22019334049409237,
|
587 |
+
"grad_norm": 2.468086968145363,
|
588 |
+
"learning_rate": 9.568997839077812e-06,
|
589 |
+
"loss": 0.6275,
|
590 |
+
"step": 410
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.22287862513426424,
|
594 |
+
"grad_norm": 2.777359054522718,
|
595 |
+
"learning_rate": 9.549752339686243e-06,
|
596 |
+
"loss": 0.6487,
|
597 |
+
"step": 415
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.22556390977443608,
|
601 |
+
"grad_norm": 2.435355179570479,
|
602 |
+
"learning_rate": 9.530106716199103e-06,
|
603 |
+
"loss": 0.6467,
|
604 |
+
"step": 420
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.22824919441460795,
|
608 |
+
"grad_norm": 2.383529479537672,
|
609 |
+
"learning_rate": 9.510062696334024e-06,
|
610 |
+
"loss": 0.6345,
|
611 |
+
"step": 425
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.23093447905477982,
|
615 |
+
"grad_norm": 2.4092098189126734,
|
616 |
+
"learning_rate": 9.489622042845266e-06,
|
617 |
+
"loss": 0.6239,
|
618 |
+
"step": 430
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.23361976369495166,
|
622 |
+
"grad_norm": 2.426365302575787,
|
623 |
+
"learning_rate": 9.468786553368694e-06,
|
624 |
+
"loss": 0.6238,
|
625 |
+
"step": 435
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.23630504833512353,
|
629 |
+
"grad_norm": 2.5087002358611574,
|
630 |
+
"learning_rate": 9.447558060263693e-06,
|
631 |
+
"loss": 0.6237,
|
632 |
+
"step": 440
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.23899033297529537,
|
636 |
+
"grad_norm": 2.387385697008228,
|
637 |
+
"learning_rate": 9.42593843045201e-06,
|
638 |
+
"loss": 0.622,
|
639 |
+
"step": 445
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.24167561761546724,
|
643 |
+
"grad_norm": 2.326708486574728,
|
644 |
+
"learning_rate": 9.403929565253582e-06,
|
645 |
+
"loss": 0.6306,
|
646 |
+
"step": 450
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.24436090225563908,
|
650 |
+
"grad_norm": 2.369288401258165,
|
651 |
+
"learning_rate": 9.381533400219319e-06,
|
652 |
+
"loss": 0.6073,
|
653 |
+
"step": 455
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.24704618689581095,
|
657 |
+
"grad_norm": 2.383046130914992,
|
658 |
+
"learning_rate": 9.358751904960885e-06,
|
659 |
+
"loss": 0.6205,
|
660 |
+
"step": 460
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.24973147153598282,
|
664 |
+
"grad_norm": 2.414995815331301,
|
665 |
+
"learning_rate": 9.335587082977484e-06,
|
666 |
+
"loss": 0.6106,
|
667 |
+
"step": 465
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.25241675617615467,
|
671 |
+
"grad_norm": 2.768675173604855,
|
672 |
+
"learning_rate": 9.312040971479663e-06,
|
673 |
+
"loss": 0.6272,
|
674 |
+
"step": 470
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.25510204081632654,
|
678 |
+
"grad_norm": 2.2921439710822185,
|
679 |
+
"learning_rate": 9.288115641210144e-06,
|
680 |
+
"loss": 0.6113,
|
681 |
+
"step": 475
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.2577873254564984,
|
685 |
+
"grad_norm": 2.3081671484334807,
|
686 |
+
"learning_rate": 9.26381319626173e-06,
|
687 |
+
"loss": 0.6102,
|
688 |
+
"step": 480
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.2604726100966702,
|
692 |
+
"grad_norm": 2.5833114870823355,
|
693 |
+
"learning_rate": 9.239135773892249e-06,
|
694 |
+
"loss": 0.6042,
|
695 |
+
"step": 485
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.2631578947368421,
|
699 |
+
"grad_norm": 2.3993711180589425,
|
700 |
+
"learning_rate": 9.214085544336597e-06,
|
701 |
+
"loss": 0.6043,
|
702 |
+
"step": 490
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.26584317937701396,
|
706 |
+
"grad_norm": 2.3928350581872073,
|
707 |
+
"learning_rate": 9.188664710615886e-06,
|
708 |
+
"loss": 0.6072,
|
709 |
+
"step": 495
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.26852846401718583,
|
713 |
+
"grad_norm": 2.291178931562558,
|
714 |
+
"learning_rate": 9.162875508343691e-06,
|
715 |
+
"loss": 0.607,
|
716 |
+
"step": 500
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.2712137486573577,
|
720 |
+
"grad_norm": 2.302429905502462,
|
721 |
+
"learning_rate": 9.136720205529446e-06,
|
722 |
+
"loss": 0.6212,
|
723 |
+
"step": 505
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.2738990332975295,
|
727 |
+
"grad_norm": 2.4002220088345227,
|
728 |
+
"learning_rate": 9.110201102378991e-06,
|
729 |
+
"loss": 0.6022,
|
730 |
+
"step": 510
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.2765843179377014,
|
734 |
+
"grad_norm": 2.345306335150251,
|
735 |
+
"learning_rate": 9.083320531092268e-06,
|
736 |
+
"loss": 0.5849,
|
737 |
+
"step": 515
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.27926960257787325,
|
741 |
+
"grad_norm": 2.361854410468674,
|
742 |
+
"learning_rate": 9.056080855658233e-06,
|
743 |
+
"loss": 0.5941,
|
744 |
+
"step": 520
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.2819548872180451,
|
748 |
+
"grad_norm": 2.329088369479831,
|
749 |
+
"learning_rate": 9.02848447164695e-06,
|
750 |
+
"loss": 0.592,
|
751 |
+
"step": 525
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.284640171858217,
|
755 |
+
"grad_norm": 2.9145104665950834,
|
756 |
+
"learning_rate": 9.00053380599891e-06,
|
757 |
+
"loss": 0.6079,
|
758 |
+
"step": 530
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.2873254564983888,
|
762 |
+
"grad_norm": 2.30854865495231,
|
763 |
+
"learning_rate": 8.972231316811608e-06,
|
764 |
+
"loss": 0.5902,
|
765 |
+
"step": 535
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.2900107411385607,
|
769 |
+
"grad_norm": 2.710514687574968,
|
770 |
+
"learning_rate": 8.943579493123354e-06,
|
771 |
+
"loss": 0.6139,
|
772 |
+
"step": 540
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.29269602577873255,
|
776 |
+
"grad_norm": 2.3749882685361907,
|
777 |
+
"learning_rate": 8.914580854694389e-06,
|
778 |
+
"loss": 0.5982,
|
779 |
+
"step": 545
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.2953813104189044,
|
783 |
+
"grad_norm": 2.341531483158182,
|
784 |
+
"learning_rate": 8.885237951785275e-06,
|
785 |
+
"loss": 0.6009,
|
786 |
+
"step": 550
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.2980665950590763,
|
790 |
+
"grad_norm": 2.31541797689058,
|
791 |
+
"learning_rate": 8.855553364932628e-06,
|
792 |
+
"loss": 0.5539,
|
793 |
+
"step": 555
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.3007518796992481,
|
797 |
+
"grad_norm": 2.3459951500114773,
|
798 |
+
"learning_rate": 8.82552970472216e-06,
|
799 |
+
"loss": 0.5793,
|
800 |
+
"step": 560
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.30343716433942,
|
804 |
+
"grad_norm": 2.305688250517311,
|
805 |
+
"learning_rate": 8.795169611559108e-06,
|
806 |
+
"loss": 0.6016,
|
807 |
+
"step": 565
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.30612244897959184,
|
811 |
+
"grad_norm": 2.5187456404531936,
|
812 |
+
"learning_rate": 8.764475755436011e-06,
|
813 |
+
"loss": 0.5785,
|
814 |
+
"step": 570
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.3088077336197637,
|
818 |
+
"grad_norm": 2.3186145964360336,
|
819 |
+
"learning_rate": 8.733450835697914e-06,
|
820 |
+
"loss": 0.5952,
|
821 |
+
"step": 575
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.31149301825993553,
|
825 |
+
"grad_norm": 2.4852031283314675,
|
826 |
+
"learning_rate": 8.702097580804962e-06,
|
827 |
+
"loss": 0.5921,
|
828 |
+
"step": 580
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.3141783029001074,
|
832 |
+
"grad_norm": 2.48247156500519,
|
833 |
+
"learning_rate": 8.670418748092465e-06,
|
834 |
+
"loss": 0.5824,
|
835 |
+
"step": 585
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.31686358754027927,
|
839 |
+
"grad_norm": 2.4487511752763003,
|
840 |
+
"learning_rate": 8.638417123528393e-06,
|
841 |
+
"loss": 0.5766,
|
842 |
+
"step": 590
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.31954887218045114,
|
846 |
+
"grad_norm": 2.5523301393566973,
|
847 |
+
"learning_rate": 8.60609552146837e-06,
|
848 |
+
"loss": 0.5778,
|
849 |
+
"step": 595
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.322234156820623,
|
853 |
+
"grad_norm": 2.463880439656359,
|
854 |
+
"learning_rate": 8.573456784408162e-06,
|
855 |
+
"loss": 0.5876,
|
856 |
+
"step": 600
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.3249194414607948,
|
860 |
+
"grad_norm": 2.279384394364597,
|
861 |
+
"learning_rate": 8.540503782733718e-06,
|
862 |
+
"loss": 0.5564,
|
863 |
+
"step": 605
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.3276047261009667,
|
867 |
+
"grad_norm": 2.3581766823519703,
|
868 |
+
"learning_rate": 8.507239414468707e-06,
|
869 |
+
"loss": 0.5711,
|
870 |
+
"step": 610
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.33029001074113856,
|
874 |
+
"grad_norm": 2.386244897526467,
|
875 |
+
"learning_rate": 8.473666605019673e-06,
|
876 |
+
"loss": 0.5606,
|
877 |
+
"step": 615
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.33297529538131043,
|
881 |
+
"grad_norm": 2.6538204423039176,
|
882 |
+
"learning_rate": 8.439788306918759e-06,
|
883 |
+
"loss": 0.5679,
|
884 |
+
"step": 620
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.3356605800214823,
|
888 |
+
"grad_norm": 2.3536598330429994,
|
889 |
+
"learning_rate": 8.405607499564044e-06,
|
890 |
+
"loss": 0.5686,
|
891 |
+
"step": 625
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.3383458646616541,
|
895 |
+
"grad_norm": 2.5411210857238737,
|
896 |
+
"learning_rate": 8.371127188957527e-06,
|
897 |
+
"loss": 0.5568,
|
898 |
+
"step": 630
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.341031149301826,
|
902 |
+
"grad_norm": 2.3248484069987447,
|
903 |
+
"learning_rate": 8.336350407440766e-06,
|
904 |
+
"loss": 0.5603,
|
905 |
+
"step": 635
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.34371643394199786,
|
909 |
+
"grad_norm": 2.531937446280703,
|
910 |
+
"learning_rate": 8.3012802134282e-06,
|
911 |
+
"loss": 0.559,
|
912 |
+
"step": 640
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.3464017185821697,
|
916 |
+
"grad_norm": 2.453303019061587,
|
917 |
+
"learning_rate": 8.26591969113818e-06,
|
918 |
+
"loss": 0.5515,
|
919 |
+
"step": 645
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.34908700322234154,
|
923 |
+
"grad_norm": 2.69008655565826,
|
924 |
+
"learning_rate": 8.230271950321733e-06,
|
925 |
+
"loss": 0.5581,
|
926 |
+
"step": 650
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.3517722878625134,
|
930 |
+
"grad_norm": 2.2858934099243546,
|
931 |
+
"learning_rate": 8.194340125989068e-06,
|
932 |
+
"loss": 0.5584,
|
933 |
+
"step": 655
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.3544575725026853,
|
937 |
+
"grad_norm": 2.4227589614633307,
|
938 |
+
"learning_rate": 8.158127378133886e-06,
|
939 |
+
"loss": 0.5338,
|
940 |
+
"step": 660
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.35714285714285715,
|
944 |
+
"grad_norm": 2.473273208390763,
|
945 |
+
"learning_rate": 8.121636891455462e-06,
|
946 |
+
"loss": 0.5488,
|
947 |
+
"step": 665
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.359828141783029,
|
951 |
+
"grad_norm": 4.029648668242547,
|
952 |
+
"learning_rate": 8.084871875078574e-06,
|
953 |
+
"loss": 0.5558,
|
954 |
+
"step": 670
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.36251342642320084,
|
958 |
+
"grad_norm": 2.3823297359151487,
|
959 |
+
"learning_rate": 8.047835562271289e-06,
|
960 |
+
"loss": 0.5347,
|
961 |
+
"step": 675
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.3651987110633727,
|
965 |
+
"grad_norm": 2.3206951441879062,
|
966 |
+
"learning_rate": 8.010531210160594e-06,
|
967 |
+
"loss": 0.5474,
|
968 |
+
"step": 680
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.3678839957035446,
|
972 |
+
"grad_norm": 2.401861048928935,
|
973 |
+
"learning_rate": 7.972962099445979e-06,
|
974 |
+
"loss": 0.5336,
|
975 |
+
"step": 685
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.37056928034371645,
|
979 |
+
"grad_norm": 2.283419604263929,
|
980 |
+
"learning_rate": 7.935131534110893e-06,
|
981 |
+
"loss": 0.5423,
|
982 |
+
"step": 690
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.3732545649838883,
|
986 |
+
"grad_norm": 2.4091957808848377,
|
987 |
+
"learning_rate": 7.897042841132195e-06,
|
988 |
+
"loss": 0.5512,
|
989 |
+
"step": 695
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.37593984962406013,
|
993 |
+
"grad_norm": 2.360101346375749,
|
994 |
+
"learning_rate": 7.858699370187558e-06,
|
995 |
+
"loss": 0.5521,
|
996 |
+
"step": 700
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.378625134264232,
|
1000 |
+
"grad_norm": 2.578398145192331,
|
1001 |
+
"learning_rate": 7.820104493360883e-06,
|
1002 |
+
"loss": 0.5446,
|
1003 |
+
"step": 705
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.38131041890440387,
|
1007 |
+
"grad_norm": 2.392308313238444,
|
1008 |
+
"learning_rate": 7.781261604845754e-06,
|
1009 |
+
"loss": 0.539,
|
1010 |
+
"step": 710
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.38399570354457574,
|
1014 |
+
"grad_norm": 2.4396652511764434,
|
1015 |
+
"learning_rate": 7.742174120646929e-06,
|
1016 |
+
"loss": 0.5459,
|
1017 |
+
"step": 715
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.3866809881847476,
|
1021 |
+
"grad_norm": 2.4045120660307138,
|
1022 |
+
"learning_rate": 7.70284547827992e-06,
|
1023 |
+
"loss": 0.5245,
|
1024 |
+
"step": 720
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.3893662728249194,
|
1028 |
+
"grad_norm": 2.2765914991878358,
|
1029 |
+
"learning_rate": 7.663279136468696e-06,
|
1030 |
+
"loss": 0.5493,
|
1031 |
+
"step": 725
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.3920515574650913,
|
1035 |
+
"grad_norm": 2.350607787315242,
|
1036 |
+
"learning_rate": 7.623478574841499e-06,
|
1037 |
+
"loss": 0.5274,
|
1038 |
+
"step": 730
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.39473684210526316,
|
1042 |
+
"grad_norm": 2.4564288550698166,
|
1043 |
+
"learning_rate": 7.58344729362483e-06,
|
1044 |
+
"loss": 0.5272,
|
1045 |
+
"step": 735
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.39742212674543503,
|
1049 |
+
"grad_norm": 2.308846635556029,
|
1050 |
+
"learning_rate": 7.5431888133356334e-06,
|
1051 |
+
"loss": 0.5256,
|
1052 |
+
"step": 740
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.40010741138560685,
|
1056 |
+
"grad_norm": 2.3648049921250105,
|
1057 |
+
"learning_rate": 7.502706674471678e-06,
|
1058 |
+
"loss": 0.5374,
|
1059 |
+
"step": 745
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.4027926960257787,
|
1063 |
+
"grad_norm": 2.372539412693635,
|
1064 |
+
"learning_rate": 7.462004437200195e-06,
|
1065 |
+
"loss": 0.5322,
|
1066 |
+
"step": 750
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.4054779806659506,
|
1070 |
+
"grad_norm": 2.35495618473913,
|
1071 |
+
"learning_rate": 7.42108568104479e-06,
|
1072 |
+
"loss": 0.5278,
|
1073 |
+
"step": 755
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.40816326530612246,
|
1077 |
+
"grad_norm": 2.432730442205152,
|
1078 |
+
"learning_rate": 7.379954004570628e-06,
|
1079 |
+
"loss": 0.543,
|
1080 |
+
"step": 760
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.41084854994629433,
|
1084 |
+
"grad_norm": 2.383640686140904,
|
1085 |
+
"learning_rate": 7.338613025067977e-06,
|
1086 |
+
"loss": 0.5493,
|
1087 |
+
"step": 765
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.41353383458646614,
|
1091 |
+
"grad_norm": 2.2937243918245307,
|
1092 |
+
"learning_rate": 7.2970663782340765e-06,
|
1093 |
+
"loss": 0.5159,
|
1094 |
+
"step": 770
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.416219119226638,
|
1098 |
+
"grad_norm": 2.584638869408312,
|
1099 |
+
"learning_rate": 7.25531771785341e-06,
|
1100 |
+
"loss": 0.528,
|
1101 |
+
"step": 775
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.4189044038668099,
|
1105 |
+
"grad_norm": 2.4008027863806363,
|
1106 |
+
"learning_rate": 7.2133707154763646e-06,
|
1107 |
+
"loss": 0.5264,
|
1108 |
+
"step": 780
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.42158968850698175,
|
1112 |
+
"grad_norm": 2.388791471928522,
|
1113 |
+
"learning_rate": 7.17122906009635e-06,
|
1114 |
+
"loss": 0.5123,
|
1115 |
+
"step": 785
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.4242749731471536,
|
1119 |
+
"grad_norm": 2.3442732809582774,
|
1120 |
+
"learning_rate": 7.128896457825364e-06,
|
1121 |
+
"loss": 0.5149,
|
1122 |
+
"step": 790
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.42696025778732544,
|
1126 |
+
"grad_norm": 2.279111093706145,
|
1127 |
+
"learning_rate": 7.08637663156807e-06,
|
1128 |
+
"loss": 0.5168,
|
1129 |
+
"step": 795
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.4296455424274973,
|
1133 |
+
"grad_norm": 2.4062295480857805,
|
1134 |
+
"learning_rate": 7.043673320694386e-06,
|
1135 |
+
"loss": 0.5015,
|
1136 |
+
"step": 800
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.4323308270676692,
|
1140 |
+
"grad_norm": 2.5033445807132932,
|
1141 |
+
"learning_rate": 7.000790280710625e-06,
|
1142 |
+
"loss": 0.5221,
|
1143 |
+
"step": 805
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.43501611170784105,
|
1147 |
+
"grad_norm": 2.3456722409693698,
|
1148 |
+
"learning_rate": 6.957731282929224e-06,
|
1149 |
+
"loss": 0.498,
|
1150 |
+
"step": 810
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.4377013963480129,
|
1154 |
+
"grad_norm": 2.3434221234555936,
|
1155 |
+
"learning_rate": 6.914500114137082e-06,
|
1156 |
+
"loss": 0.4955,
|
1157 |
+
"step": 815
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.44038668098818473,
|
1161 |
+
"grad_norm": 2.351670401180038,
|
1162 |
+
"learning_rate": 6.871100576262526e-06,
|
1163 |
+
"loss": 0.5041,
|
1164 |
+
"step": 820
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.4430719656283566,
|
1168 |
+
"grad_norm": 2.357200356793434,
|
1169 |
+
"learning_rate": 6.827536486040964e-06,
|
1170 |
+
"loss": 0.5131,
|
1171 |
+
"step": 825
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.44575725026852847,
|
1175 |
+
"grad_norm": 2.3716872327140948,
|
1176 |
+
"learning_rate": 6.783811674679216e-06,
|
1177 |
+
"loss": 0.5072,
|
1178 |
+
"step": 830
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.44844253490870034,
|
1182 |
+
"grad_norm": 2.3837956868904753,
|
1183 |
+
"learning_rate": 6.7399299875185875e-06,
|
1184 |
+
"loss": 0.5072,
|
1185 |
+
"step": 835
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.45112781954887216,
|
1189 |
+
"grad_norm": 2.468749709382942,
|
1190 |
+
"learning_rate": 6.695895283696691e-06,
|
1191 |
+
"loss": 0.4916,
|
1192 |
+
"step": 840
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.453813104189044,
|
1196 |
+
"grad_norm": 2.2977664623966576,
|
1197 |
+
"learning_rate": 6.651711435808061e-06,
|
1198 |
+
"loss": 0.5073,
|
1199 |
+
"step": 845
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.4564983888292159,
|
1203 |
+
"grad_norm": 2.3750347681190367,
|
1204 |
+
"learning_rate": 6.607382329563581e-06,
|
1205 |
+
"loss": 0.5042,
|
1206 |
+
"step": 850
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.45918367346938777,
|
1210 |
+
"grad_norm": 2.3201817159617604,
|
1211 |
+
"learning_rate": 6.562911863448752e-06,
|
1212 |
+
"loss": 0.505,
|
1213 |
+
"step": 855
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.46186895810955964,
|
1217 |
+
"grad_norm": 2.317046101118156,
|
1218 |
+
"learning_rate": 6.518303948380854e-06,
|
1219 |
+
"loss": 0.5027,
|
1220 |
+
"step": 860
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.46455424274973145,
|
1224 |
+
"grad_norm": 2.4450948832382844,
|
1225 |
+
"learning_rate": 6.473562507364995e-06,
|
1226 |
+
"loss": 0.5057,
|
1227 |
+
"step": 865
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.4672395273899033,
|
1231 |
+
"grad_norm": 2.342313602440348,
|
1232 |
+
"learning_rate": 6.428691475149107e-06,
|
1233 |
+
"loss": 0.4859,
|
1234 |
+
"step": 870
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.4699248120300752,
|
1238 |
+
"grad_norm": 2.3462970754354084,
|
1239 |
+
"learning_rate": 6.383694797877915e-06,
|
1240 |
+
"loss": 0.4714,
|
1241 |
+
"step": 875
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.47261009667024706,
|
1245 |
+
"grad_norm": 2.5885071330202627,
|
1246 |
+
"learning_rate": 6.338576432745891e-06,
|
1247 |
+
"loss": 0.4875,
|
1248 |
+
"step": 880
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.47529538131041893,
|
1252 |
+
"grad_norm": 2.2979744167450358,
|
1253 |
+
"learning_rate": 6.293340347649234e-06,
|
1254 |
+
"loss": 0.4951,
|
1255 |
+
"step": 885
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.47798066595059074,
|
1259 |
+
"grad_norm": 2.291237469268037,
|
1260 |
+
"learning_rate": 6.247990520836935e-06,
|
1261 |
+
"loss": 0.4883,
|
1262 |
+
"step": 890
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.4806659505907626,
|
1266 |
+
"grad_norm": 2.3728016920526565,
|
1267 |
+
"learning_rate": 6.202530940560897e-06,
|
1268 |
+
"loss": 0.485,
|
1269 |
+
"step": 895
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.4833512352309345,
|
1273 |
+
"grad_norm": 2.610281937328259,
|
1274 |
+
"learning_rate": 6.156965604725202e-06,
|
1275 |
+
"loss": 0.4882,
|
1276 |
+
"step": 900
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.48603651987110635,
|
1280 |
+
"grad_norm": 2.3718642604920834,
|
1281 |
+
"learning_rate": 6.111298520534514e-06,
|
1282 |
+
"loss": 0.4866,
|
1283 |
+
"step": 905
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.48872180451127817,
|
1287 |
+
"grad_norm": 2.3757882082854573,
|
1288 |
+
"learning_rate": 6.065533704141666e-06,
|
1289 |
+
"loss": 0.4798,
|
1290 |
+
"step": 910
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.49140708915145004,
|
1294 |
+
"grad_norm": 2.3521649976871712,
|
1295 |
+
"learning_rate": 6.019675180294469e-06,
|
1296 |
+
"loss": 0.4876,
|
1297 |
+
"step": 915
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.4940923737916219,
|
1301 |
+
"grad_norm": 2.263694366342218,
|
1302 |
+
"learning_rate": 5.973726981981756e-06,
|
1303 |
+
"loss": 0.4838,
|
1304 |
+
"step": 920
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.4967776584317938,
|
1308 |
+
"grad_norm": 2.574827738131497,
|
1309 |
+
"learning_rate": 5.9276931500787045e-06,
|
1310 |
+
"loss": 0.4893,
|
1311 |
+
"step": 925
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.49946294307196565,
|
1315 |
+
"grad_norm": 2.3280039732972355,
|
1316 |
+
"learning_rate": 5.8815777329914655e-06,
|
1317 |
+
"loss": 0.4748,
|
1318 |
+
"step": 930
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.5021482277121375,
|
1322 |
+
"grad_norm": 2.472054452574987,
|
1323 |
+
"learning_rate": 5.83538478630113e-06,
|
1324 |
+
"loss": 0.4906,
|
1325 |
+
"step": 935
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.5048335123523093,
|
1329 |
+
"grad_norm": 2.3804196285899186,
|
1330 |
+
"learning_rate": 5.789118372407061e-06,
|
1331 |
+
"loss": 0.4631,
|
1332 |
+
"step": 940
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.5075187969924813,
|
1336 |
+
"grad_norm": 2.3558638927371742,
|
1337 |
+
"learning_rate": 5.7427825601696376e-06,
|
1338 |
+
"loss": 0.4691,
|
1339 |
+
"step": 945
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.5102040816326531,
|
1343 |
+
"grad_norm": 2.359879850321753,
|
1344 |
+
"learning_rate": 5.696381424552411e-06,
|
1345 |
+
"loss": 0.4639,
|
1346 |
+
"step": 950
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.5128893662728249,
|
1350 |
+
"grad_norm": 2.322787328071383,
|
1351 |
+
"learning_rate": 5.649919046263742e-06,
|
1352 |
+
"loss": 0.465,
|
1353 |
+
"step": 955
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.5155746509129968,
|
1357 |
+
"grad_norm": 2.4521402907626646,
|
1358 |
+
"learning_rate": 5.6033995113979336e-06,
|
1359 |
+
"loss": 0.4623,
|
1360 |
+
"step": 960
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.5182599355531686,
|
1364 |
+
"grad_norm": 2.2573376873830284,
|
1365 |
+
"learning_rate": 5.556826911075867e-06,
|
1366 |
+
"loss": 0.4642,
|
1367 |
+
"step": 965
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.5209452201933404,
|
1371 |
+
"grad_norm": 2.398522707712972,
|
1372 |
+
"learning_rate": 5.510205341085224e-06,
|
1373 |
+
"loss": 0.4754,
|
1374 |
+
"step": 970
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.5236305048335124,
|
1378 |
+
"grad_norm": 2.4170684561458757,
|
1379 |
+
"learning_rate": 5.463538901520278e-06,
|
1380 |
+
"loss": 0.4602,
|
1381 |
+
"step": 975
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.5263157894736842,
|
1385 |
+
"grad_norm": 2.3207712466054096,
|
1386 |
+
"learning_rate": 5.416831696421325e-06,
|
1387 |
+
"loss": 0.464,
|
1388 |
+
"step": 980
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.5290010741138561,
|
1392 |
+
"grad_norm": 2.4520936790108108,
|
1393 |
+
"learning_rate": 5.370087833413747e-06,
|
1394 |
+
"loss": 0.47,
|
1395 |
+
"step": 985
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.5316863587540279,
|
1399 |
+
"grad_norm": 2.382760350655894,
|
1400 |
+
"learning_rate": 5.323311423346775e-06,
|
1401 |
+
"loss": 0.4744,
|
1402 |
+
"step": 990
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.5343716433941997,
|
1406 |
+
"grad_norm": 2.392728462627693,
|
1407 |
+
"learning_rate": 5.2765065799319646e-06,
|
1408 |
+
"loss": 0.4539,
|
1409 |
+
"step": 995
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.5370569280343717,
|
1413 |
+
"grad_norm": 2.317069450690583,
|
1414 |
+
"learning_rate": 5.229677419381417e-06,
|
1415 |
+
"loss": 0.4724,
|
1416 |
+
"step": 1000
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.5397422126745435,
|
1420 |
+
"grad_norm": 2.181814505403613,
|
1421 |
+
"learning_rate": 5.182828060045783e-06,
|
1422 |
+
"loss": 0.4709,
|
1423 |
+
"step": 1005
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.5424274973147154,
|
1427 |
+
"grad_norm": 2.2356286647456862,
|
1428 |
+
"learning_rate": 5.1359626220520804e-06,
|
1429 |
+
"loss": 0.4543,
|
1430 |
+
"step": 1010
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.5451127819548872,
|
1434 |
+
"grad_norm": 2.309541338501052,
|
1435 |
+
"learning_rate": 5.08908522694135e-06,
|
1436 |
+
"loss": 0.4445,
|
1437 |
+
"step": 1015
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.547798066595059,
|
1441 |
+
"grad_norm": 2.334557739373175,
|
1442 |
+
"learning_rate": 5.042199997306185e-06,
|
1443 |
+
"loss": 0.4579,
|
1444 |
+
"step": 1020
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 0.550483351235231,
|
1448 |
+
"grad_norm": 2.3861434626116633,
|
1449 |
+
"learning_rate": 4.995311056428192e-06,
|
1450 |
+
"loss": 0.4487,
|
1451 |
+
"step": 1025
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 0.5531686358754028,
|
1455 |
+
"grad_norm": 2.277598529970781,
|
1456 |
+
"learning_rate": 4.948422527915348e-06,
|
1457 |
+
"loss": 0.459,
|
1458 |
+
"step": 1030
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 0.5558539205155747,
|
1462 |
+
"grad_norm": 2.5154062392897885,
|
1463 |
+
"learning_rate": 4.901538535339369e-06,
|
1464 |
+
"loss": 0.4471,
|
1465 |
+
"step": 1035
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 0.5585392051557465,
|
1469 |
+
"grad_norm": 2.290644492599463,
|
1470 |
+
"learning_rate": 4.854663201873066e-06,
|
1471 |
+
"loss": 0.4534,
|
1472 |
+
"step": 1040
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 0.5612244897959183,
|
1476 |
+
"grad_norm": 2.3507551432764715,
|
1477 |
+
"learning_rate": 4.807800649927723e-06,
|
1478 |
+
"loss": 0.4463,
|
1479 |
+
"step": 1045
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 0.5639097744360902,
|
1483 |
+
"grad_norm": 2.1774339041234225,
|
1484 |
+
"learning_rate": 4.760955000790572e-06,
|
1485 |
+
"loss": 0.4365,
|
1486 |
+
"step": 1050
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 0.5665950590762621,
|
1490 |
+
"grad_norm": 2.1529579750128454,
|
1491 |
+
"learning_rate": 4.714130374262339e-06,
|
1492 |
+
"loss": 0.4227,
|
1493 |
+
"step": 1055
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.569280343716434,
|
1497 |
+
"grad_norm": 2.320093713572853,
|
1498 |
+
"learning_rate": 4.667330888294934e-06,
|
1499 |
+
"loss": 0.4212,
|
1500 |
+
"step": 1060
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 0.5719656283566058,
|
1504 |
+
"grad_norm": 2.2930843074583427,
|
1505 |
+
"learning_rate": 4.620560658629303e-06,
|
1506 |
+
"loss": 0.4284,
|
1507 |
+
"step": 1065
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 0.5746509129967776,
|
1511 |
+
"grad_norm": 2.2548004697834165,
|
1512 |
+
"learning_rate": 4.573823798433469e-06,
|
1513 |
+
"loss": 0.4446,
|
1514 |
+
"step": 1070
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 0.5773361976369495,
|
1518 |
+
"grad_norm": 2.6223716299704347,
|
1519 |
+
"learning_rate": 4.527124417940812e-06,
|
1520 |
+
"loss": 0.4569,
|
1521 |
+
"step": 1075
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 0.5800214822771214,
|
1525 |
+
"grad_norm": 2.2996066174355776,
|
1526 |
+
"learning_rate": 4.48046662408858e-06,
|
1527 |
+
"loss": 0.443,
|
1528 |
+
"step": 1080
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 0.5827067669172933,
|
1532 |
+
"grad_norm": 2.331188250468318,
|
1533 |
+
"learning_rate": 4.43385452015673e-06,
|
1534 |
+
"loss": 0.4373,
|
1535 |
+
"step": 1085
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 0.5853920515574651,
|
1539 |
+
"grad_norm": 2.2636646634515527,
|
1540 |
+
"learning_rate": 4.387292205407055e-06,
|
1541 |
+
"loss": 0.4359,
|
1542 |
+
"step": 1090
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 0.5880773361976369,
|
1546 |
+
"grad_norm": 2.2831377194175033,
|
1547 |
+
"learning_rate": 4.3407837747226765e-06,
|
1548 |
+
"loss": 0.4364,
|
1549 |
+
"step": 1095
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 0.5907626208378088,
|
1553 |
+
"grad_norm": 2.2946882078675146,
|
1554 |
+
"learning_rate": 4.29433331824793e-06,
|
1555 |
+
"loss": 0.4261,
|
1556 |
+
"step": 1100
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.5934479054779807,
|
1560 |
+
"grad_norm": 2.2943723069744664,
|
1561 |
+
"learning_rate": 4.247944921028663e-06,
|
1562 |
+
"loss": 0.4358,
|
1563 |
+
"step": 1105
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 0.5961331901181526,
|
1567 |
+
"grad_norm": 2.579316764381701,
|
1568 |
+
"learning_rate": 4.201622662652972e-06,
|
1569 |
+
"loss": 0.4204,
|
1570 |
+
"step": 1110
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 0.5988184747583244,
|
1574 |
+
"grad_norm": 2.1742173999609737,
|
1575 |
+
"learning_rate": 4.155370616892422e-06,
|
1576 |
+
"loss": 0.4269,
|
1577 |
+
"step": 1115
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 0.6015037593984962,
|
1581 |
+
"grad_norm": 2.6709708559911935,
|
1582 |
+
"learning_rate": 4.1091928513438005e-06,
|
1583 |
+
"loss": 0.4259,
|
1584 |
+
"step": 1120
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 0.6041890440386681,
|
1588 |
+
"grad_norm": 2.1624211921715277,
|
1589 |
+
"learning_rate": 4.063093427071376e-06,
|
1590 |
+
"loss": 0.4311,
|
1591 |
+
"step": 1125
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 0.60687432867884,
|
1595 |
+
"grad_norm": 2.3121026535251294,
|
1596 |
+
"learning_rate": 4.0170763982497716e-06,
|
1597 |
+
"loss": 0.4123,
|
1598 |
+
"step": 1130
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 0.6095596133190118,
|
1602 |
+
"grad_norm": 2.291719266377087,
|
1603 |
+
"learning_rate": 3.971145811807397e-06,
|
1604 |
+
"loss": 0.4332,
|
1605 |
+
"step": 1135
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 0.6122448979591837,
|
1609 |
+
"grad_norm": 2.151850103320739,
|
1610 |
+
"learning_rate": 3.925305707070572e-06,
|
1611 |
+
"loss": 0.4242,
|
1612 |
+
"step": 1140
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 0.6149301825993555,
|
1616 |
+
"grad_norm": 2.199401051328622,
|
1617 |
+
"learning_rate": 3.879560115408279e-06,
|
1618 |
+
"loss": 0.4158,
|
1619 |
+
"step": 1145
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 0.6176154672395274,
|
1623 |
+
"grad_norm": 2.229032836245145,
|
1624 |
+
"learning_rate": 3.833913059877622e-06,
|
1625 |
+
"loss": 0.4221,
|
1626 |
+
"step": 1150
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 0.6203007518796992,
|
1630 |
+
"grad_norm": 2.29573891109709,
|
1631 |
+
"learning_rate": 3.788368554870034e-06,
|
1632 |
+
"loss": 0.4213,
|
1633 |
+
"step": 1155
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 0.6229860365198711,
|
1637 |
+
"grad_norm": 2.1322887575075784,
|
1638 |
+
"learning_rate": 3.7429306057582306e-06,
|
1639 |
+
"loss": 0.4191,
|
1640 |
+
"step": 1160
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 0.625671321160043,
|
1644 |
+
"grad_norm": 2.2395834148078198,
|
1645 |
+
"learning_rate": 3.6976032085439563e-06,
|
1646 |
+
"loss": 0.4203,
|
1647 |
+
"step": 1165
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 0.6283566058002148,
|
1651 |
+
"grad_norm": 2.4717439033137625,
|
1652 |
+
"learning_rate": 3.6523903495065634e-06,
|
1653 |
+
"loss": 0.4186,
|
1654 |
+
"step": 1170
|
1655 |
+
},
|
1656 |
+
{
|
1657 |
+
"epoch": 0.6310418904403867,
|
1658 |
+
"grad_norm": 2.1334153639365976,
|
1659 |
+
"learning_rate": 3.6072960048524443e-06,
|
1660 |
+
"loss": 0.4142,
|
1661 |
+
"step": 1175
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 0.6337271750805585,
|
1665 |
+
"grad_norm": 2.427052147510359,
|
1666 |
+
"learning_rate": 3.562324140365343e-06,
|
1667 |
+
"loss": 0.3999,
|
1668 |
+
"step": 1180
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 0.6364124597207304,
|
1672 |
+
"grad_norm": 2.343168557743432,
|
1673 |
+
"learning_rate": 3.5174787110575904e-06,
|
1674 |
+
"loss": 0.4236,
|
1675 |
+
"step": 1185
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 0.6390977443609023,
|
1679 |
+
"grad_norm": 2.1562840384388213,
|
1680 |
+
"learning_rate": 3.4727636608222813e-06,
|
1681 |
+
"loss": 0.4245,
|
1682 |
+
"step": 1190
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.6417830290010741,
|
1686 |
+
"grad_norm": 2.36892688222517,
|
1687 |
+
"learning_rate": 3.428182922086437e-06,
|
1688 |
+
"loss": 0.4228,
|
1689 |
+
"step": 1195
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 0.644468313641246,
|
1693 |
+
"grad_norm": 2.156449921799051,
|
1694 |
+
"learning_rate": 3.3837404154651677e-06,
|
1695 |
+
"loss": 0.415,
|
1696 |
+
"step": 1200
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 0.6471535982814178,
|
1700 |
+
"grad_norm": 2.368201965573027,
|
1701 |
+
"learning_rate": 3.3394400494168745e-06,
|
1702 |
+
"loss": 0.4045,
|
1703 |
+
"step": 1205
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 0.6498388829215896,
|
1707 |
+
"grad_norm": 2.272852040156483,
|
1708 |
+
"learning_rate": 3.2952857198995304e-06,
|
1709 |
+
"loss": 0.4206,
|
1710 |
+
"step": 1210
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 0.6525241675617616,
|
1714 |
+
"grad_norm": 2.1923519803290206,
|
1715 |
+
"learning_rate": 3.2512813100280494e-06,
|
1716 |
+
"loss": 0.3881,
|
1717 |
+
"step": 1215
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 0.6552094522019334,
|
1721 |
+
"grad_norm": 2.1711514227665556,
|
1722 |
+
"learning_rate": 3.2074306897327913e-06,
|
1723 |
+
"loss": 0.4089,
|
1724 |
+
"step": 1220
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 0.6578947368421053,
|
1728 |
+
"grad_norm": 2.209964102937834,
|
1729 |
+
"learning_rate": 3.163737715419216e-06,
|
1730 |
+
"loss": 0.4131,
|
1731 |
+
"step": 1225
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 0.6605800214822771,
|
1735 |
+
"grad_norm": 2.143618617488237,
|
1736 |
+
"learning_rate": 3.1202062296287507e-06,
|
1737 |
+
"loss": 0.4015,
|
1738 |
+
"step": 1230
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 0.6632653061224489,
|
1742 |
+
"grad_norm": 2.1799040308872875,
|
1743 |
+
"learning_rate": 3.076840060700848e-06,
|
1744 |
+
"loss": 0.4186,
|
1745 |
+
"step": 1235
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 0.6659505907626209,
|
1749 |
+
"grad_norm": 2.152210047047405,
|
1750 |
+
"learning_rate": 3.0336430224363067e-06,
|
1751 |
+
"loss": 0.4149,
|
1752 |
+
"step": 1240
|
1753 |
+
},
|
1754 |
+
{
|
1755 |
+
"epoch": 0.6686358754027927,
|
1756 |
+
"grad_norm": 2.2584562292798362,
|
1757 |
+
"learning_rate": 2.990618913761876e-06,
|
1758 |
+
"loss": 0.3926,
|
1759 |
+
"step": 1245
|
1760 |
+
},
|
1761 |
+
{
|
1762 |
+
"epoch": 0.6713211600429646,
|
1763 |
+
"grad_norm": 2.291181695240791,
|
1764 |
+
"learning_rate": 2.9477715183961634e-06,
|
1765 |
+
"loss": 0.3955,
|
1766 |
+
"step": 1250
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 0.6740064446831364,
|
1770 |
+
"grad_norm": 2.1922026655553717,
|
1771 |
+
"learning_rate": 2.905104604516872e-06,
|
1772 |
+
"loss": 0.394,
|
1773 |
+
"step": 1255
|
1774 |
+
},
|
1775 |
+
{
|
1776 |
+
"epoch": 0.6766917293233082,
|
1777 |
+
"grad_norm": 2.3493386069538227,
|
1778 |
+
"learning_rate": 2.8626219244294074e-06,
|
1779 |
+
"loss": 0.3969,
|
1780 |
+
"step": 1260
|
1781 |
+
},
|
1782 |
+
{
|
1783 |
+
"epoch": 0.6793770139634802,
|
1784 |
+
"grad_norm": 2.2115983870340172,
|
1785 |
+
"learning_rate": 2.820327214236904e-06,
|
1786 |
+
"loss": 0.3937,
|
1787 |
+
"step": 1265
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 0.682062298603652,
|
1791 |
+
"grad_norm": 2.1951996772318934,
|
1792 |
+
"learning_rate": 2.7782241935116337e-06,
|
1793 |
+
"loss": 0.4049,
|
1794 |
+
"step": 1270
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 0.6847475832438239,
|
1798 |
+
"grad_norm": 2.425201994001083,
|
1799 |
+
"learning_rate": 2.7363165649679045e-06,
|
1800 |
+
"loss": 0.403,
|
1801 |
+
"step": 1275
|
1802 |
+
},
|
1803 |
+
{
|
1804 |
+
"epoch": 0.6874328678839957,
|
1805 |
+
"grad_norm": 2.2240784535487883,
|
1806 |
+
"learning_rate": 2.6946080141364295e-06,
|
1807 |
+
"loss": 0.3873,
|
1808 |
+
"step": 1280
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 0.6901181525241675,
|
1812 |
+
"grad_norm": 2.2329860565273054,
|
1813 |
+
"learning_rate": 2.6531022090401946e-06,
|
1814 |
+
"loss": 0.3902,
|
1815 |
+
"step": 1285
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 0.6928034371643395,
|
1819 |
+
"grad_norm": 2.327403916708101,
|
1820 |
+
"learning_rate": 2.611802799871893e-06,
|
1821 |
+
"loss": 0.3913,
|
1822 |
+
"step": 1290
|
1823 |
+
},
|
1824 |
+
{
|
1825 |
+
"epoch": 0.6954887218045113,
|
1826 |
+
"grad_norm": 2.1904991482392018,
|
1827 |
+
"learning_rate": 2.570713418672893e-06,
|
1828 |
+
"loss": 0.3919,
|
1829 |
+
"step": 1295
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 0.6981740064446831,
|
1833 |
+
"grad_norm": 2.193162643266735,
|
1834 |
+
"learning_rate": 2.529837679013841e-06,
|
1835 |
+
"loss": 0.3878,
|
1836 |
+
"step": 1300
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 0.700859291084855,
|
1840 |
+
"grad_norm": 2.37196563077284,
|
1841 |
+
"learning_rate": 2.4891791756768587e-06,
|
1842 |
+
"loss": 0.393,
|
1843 |
+
"step": 1305
|
1844 |
+
},
|
1845 |
+
{
|
1846 |
+
"epoch": 0.7035445757250268,
|
1847 |
+
"grad_norm": 2.165348884869985,
|
1848 |
+
"learning_rate": 2.4487414843394015e-06,
|
1849 |
+
"loss": 0.3931,
|
1850 |
+
"step": 1310
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 0.7062298603651987,
|
1854 |
+
"grad_norm": 2.1614220141607277,
|
1855 |
+
"learning_rate": 2.4085281612598027e-06,
|
1856 |
+
"loss": 0.3912,
|
1857 |
+
"step": 1315
|
1858 |
+
},
|
1859 |
+
{
|
1860 |
+
"epoch": 0.7089151450053706,
|
1861 |
+
"grad_norm": 2.169220432212257,
|
1862 |
+
"learning_rate": 2.368542742964519e-06,
|
1863 |
+
"loss": 0.3869,
|
1864 |
+
"step": 1320
|
1865 |
+
},
|
1866 |
+
{
|
1867 |
+
"epoch": 0.7116004296455424,
|
1868 |
+
"grad_norm": 2.2172750452575154,
|
1869 |
+
"learning_rate": 2.3287887459371193e-06,
|
1870 |
+
"loss": 0.39,
|
1871 |
+
"step": 1325
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 0.7142857142857143,
|
1875 |
+
"grad_norm": 2.386818932717909,
|
1876 |
+
"learning_rate": 2.2892696663090223e-06,
|
1877 |
+
"loss": 0.3852,
|
1878 |
+
"step": 1330
|
1879 |
+
},
|
1880 |
+
{
|
1881 |
+
"epoch": 0.7169709989258861,
|
1882 |
+
"grad_norm": 2.1061167664528395,
|
1883 |
+
"learning_rate": 2.2499889795520342e-06,
|
1884 |
+
"loss": 0.3759,
|
1885 |
+
"step": 1335
|
1886 |
+
},
|
1887 |
+
{
|
1888 |
+
"epoch": 0.719656283566058,
|
1889 |
+
"grad_norm": 2.3100938865418454,
|
1890 |
+
"learning_rate": 2.2109501401727097e-06,
|
1891 |
+
"loss": 0.385,
|
1892 |
+
"step": 1340
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 0.7223415682062299,
|
1896 |
+
"grad_norm": 2.108999556383698,
|
1897 |
+
"learning_rate": 2.1721565814085355e-06,
|
1898 |
+
"loss": 0.3909,
|
1899 |
+
"step": 1345
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 0.7250268528464017,
|
1903 |
+
"grad_norm": 2.2870436769769182,
|
1904 |
+
"learning_rate": 2.133611714926005e-06,
|
1905 |
+
"loss": 0.387,
|
1906 |
+
"step": 1350
|
1907 |
+
},
|
1908 |
+
{
|
1909 |
+
"epoch": 0.7277121374865736,
|
1910 |
+
"grad_norm": 2.0105731033958594,
|
1911 |
+
"learning_rate": 2.0953189305205847e-06,
|
1912 |
+
"loss": 0.3784,
|
1913 |
+
"step": 1355
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 0.7303974221267454,
|
1917 |
+
"grad_norm": 2.0901917154894716,
|
1918 |
+
"learning_rate": 2.057281595818592e-06,
|
1919 |
+
"loss": 0.3859,
|
1920 |
+
"step": 1360
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 0.7330827067669173,
|
1924 |
+
"grad_norm": 2.225721981804376,
|
1925 |
+
"learning_rate": 2.0195030559810387e-06,
|
1926 |
+
"loss": 0.3877,
|
1927 |
+
"step": 1365
|
1928 |
+
},
|
1929 |
+
{
|
1930 |
+
"epoch": 0.7357679914070892,
|
1931 |
+
"grad_norm": 2.1392668817849643,
|
1932 |
+
"learning_rate": 1.981986633409447e-06,
|
1933 |
+
"loss": 0.3846,
|
1934 |
+
"step": 1370
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 0.738453276047261,
|
1938 |
+
"grad_norm": 2.255949862237254,
|
1939 |
+
"learning_rate": 1.944735627453654e-06,
|
1940 |
+
"loss": 0.3702,
|
1941 |
+
"step": 1375
|
1942 |
+
},
|
1943 |
+
{
|
1944 |
+
"epoch": 0.7411385606874329,
|
1945 |
+
"grad_norm": 2.223172130508635,
|
1946 |
+
"learning_rate": 1.907753314121662e-06,
|
1947 |
+
"loss": 0.3706,
|
1948 |
+
"step": 1380
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 0.7438238453276047,
|
1952 |
+
"grad_norm": 2.120940618444079,
|
1953 |
+
"learning_rate": 1.8710429457915269e-06,
|
1954 |
+
"loss": 0.3723,
|
1955 |
+
"step": 1385
|
1956 |
+
},
|
1957 |
+
{
|
1958 |
+
"epoch": 0.7465091299677766,
|
1959 |
+
"grad_norm": 2.1971361335849506,
|
1960 |
+
"learning_rate": 1.834607750925333e-06,
|
1961 |
+
"loss": 0.3834,
|
1962 |
+
"step": 1390
|
1963 |
+
},
|
1964 |
+
{
|
1965 |
+
"epoch": 0.7491944146079484,
|
1966 |
+
"grad_norm": 2.0298699482093423,
|
1967 |
+
"learning_rate": 1.7984509337852724e-06,
|
1968 |
+
"loss": 0.3781,
|
1969 |
+
"step": 1395
|
1970 |
+
},
|
1971 |
+
{
|
1972 |
+
"epoch": 0.7518796992481203,
|
1973 |
+
"grad_norm": 1.9834275411391264,
|
1974 |
+
"learning_rate": 1.7625756741518407e-06,
|
1975 |
+
"loss": 0.3788,
|
1976 |
+
"step": 1400
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 0.7545649838882922,
|
1980 |
+
"grad_norm": 2.0910842672856993,
|
1981 |
+
"learning_rate": 1.7269851270441978e-06,
|
1982 |
+
"loss": 0.3754,
|
1983 |
+
"step": 1405
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 0.757250268528464,
|
1987 |
+
"grad_norm": 2.0335731739498093,
|
1988 |
+
"learning_rate": 1.6916824224427098e-06,
|
1989 |
+
"loss": 0.3725,
|
1990 |
+
"step": 1410
|
1991 |
+
},
|
1992 |
+
{
|
1993 |
+
"epoch": 0.7599355531686359,
|
1994 |
+
"grad_norm": 2.225103100697698,
|
1995 |
+
"learning_rate": 1.6566706650136704e-06,
|
1996 |
+
"loss": 0.3718,
|
1997 |
+
"step": 1415
|
1998 |
+
},
|
1999 |
+
{
|
2000 |
+
"epoch": 0.7626208378088077,
|
2001 |
+
"grad_norm": 2.240436179080048,
|
2002 |
+
"learning_rate": 1.6219529338362756e-06,
|
2003 |
+
"loss": 0.3749,
|
2004 |
+
"step": 1420
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 0.7653061224489796,
|
2008 |
+
"grad_norm": 2.2290538372638045,
|
2009 |
+
"learning_rate": 1.587532282131835e-06,
|
2010 |
+
"loss": 0.3825,
|
2011 |
+
"step": 1425
|
2012 |
+
},
|
2013 |
+
{
|
2014 |
+
"epoch": 0.7679914070891515,
|
2015 |
+
"grad_norm": 2.0263712228907957,
|
2016 |
+
"learning_rate": 1.5534117369952523e-06,
|
2017 |
+
"loss": 0.371,
|
2018 |
+
"step": 1430
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 0.7706766917293233,
|
2022 |
+
"grad_norm": 2.2797523248416005,
|
2023 |
+
"learning_rate": 1.5195942991288214e-06,
|
2024 |
+
"loss": 0.3638,
|
2025 |
+
"step": 1435
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 0.7733619763694952,
|
2029 |
+
"grad_norm": 2.0115399295361733,
|
2030 |
+
"learning_rate": 1.4860829425783151e-06,
|
2031 |
+
"loss": 0.3733,
|
2032 |
+
"step": 1440
|
2033 |
+
},
|
2034 |
+
{
|
2035 |
+
"epoch": 0.776047261009667,
|
2036 |
+
"grad_norm": 2.2080947234810493,
|
2037 |
+
"learning_rate": 1.4528806144714546e-06,
|
2038 |
+
"loss": 0.3778,
|
2039 |
+
"step": 1445
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 0.7787325456498388,
|
2043 |
+
"grad_norm": 2.290998397138441,
|
2044 |
+
"learning_rate": 1.4199902347587186e-06,
|
2045 |
+
"loss": 0.3656,
|
2046 |
+
"step": 1450
|
2047 |
+
},
|
2048 |
+
{
|
2049 |
+
"epoch": 0.7814178302900108,
|
2050 |
+
"grad_norm": 2.228838980742681,
|
2051 |
+
"learning_rate": 1.3874146959565387e-06,
|
2052 |
+
"loss": 0.3768,
|
2053 |
+
"step": 1455
|
2054 |
+
},
|
2055 |
+
{
|
2056 |
+
"epoch": 0.7841031149301826,
|
2057 |
+
"grad_norm": 2.257284588661445,
|
2058 |
+
"learning_rate": 1.3551568628929434e-06,
|
2059 |
+
"loss": 0.3702,
|
2060 |
+
"step": 1460
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 0.7867883995703545,
|
2064 |
+
"grad_norm": 2.0867760648839884,
|
2065 |
+
"learning_rate": 1.3232195724555919e-06,
|
2066 |
+
"loss": 0.3613,
|
2067 |
+
"step": 1465
|
2068 |
+
},
|
2069 |
+
{
|
2070 |
+
"epoch": 0.7894736842105263,
|
2071 |
+
"grad_norm": 2.0536005270696687,
|
2072 |
+
"learning_rate": 1.291605633342301e-06,
|
2073 |
+
"loss": 0.3672,
|
2074 |
+
"step": 1470
|
2075 |
+
},
|
2076 |
+
{
|
2077 |
+
"epoch": 0.7921589688506981,
|
2078 |
+
"grad_norm": 2.1344491365761695,
|
2079 |
+
"learning_rate": 1.2603178258140264e-06,
|
2080 |
+
"loss": 0.363,
|
2081 |
+
"step": 1475
|
2082 |
+
},
|
2083 |
+
{
|
2084 |
+
"epoch": 0.7948442534908701,
|
2085 |
+
"grad_norm": 2.1698541824902953,
|
2086 |
+
"learning_rate": 1.22935890145036e-06,
|
2087 |
+
"loss": 0.3663,
|
2088 |
+
"step": 1480
|
2089 |
+
},
|
2090 |
+
{
|
2091 |
+
"epoch": 0.7975295381310419,
|
2092 |
+
"grad_norm": 2.1279360172580213,
|
2093 |
+
"learning_rate": 1.198731582907547e-06,
|
2094 |
+
"loss": 0.3645,
|
2095 |
+
"step": 1485
|
2096 |
+
},
|
2097 |
+
{
|
2098 |
+
"epoch": 0.8002148227712137,
|
2099 |
+
"grad_norm": 2.169055503223085,
|
2100 |
+
"learning_rate": 1.1684385636790368e-06,
|
2101 |
+
"loss": 0.3599,
|
2102 |
+
"step": 1490
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 0.8029001074113856,
|
2106 |
+
"grad_norm": 1.9187045539119953,
|
2107 |
+
"learning_rate": 1.1384825078586114e-06,
|
2108 |
+
"loss": 0.3522,
|
2109 |
+
"step": 1495
|
2110 |
+
},
|
2111 |
+
{
|
2112 |
+
"epoch": 0.8055853920515574,
|
2113 |
+
"grad_norm": 2.2725376974932945,
|
2114 |
+
"learning_rate": 1.108866049906096e-06,
|
2115 |
+
"loss": 0.3627,
|
2116 |
+
"step": 1500
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 0.8082706766917294,
|
2120 |
+
"grad_norm": 2.0445590327981824,
|
2121 |
+
"learning_rate": 1.079591794415667e-06,
|
2122 |
+
"loss": 0.3652,
|
2123 |
+
"step": 1505
|
2124 |
+
},
|
2125 |
+
{
|
2126 |
+
"epoch": 0.8109559613319012,
|
2127 |
+
"grad_norm": 2.160848009226544,
|
2128 |
+
"learning_rate": 1.0506623158867952e-06,
|
2129 |
+
"loss": 0.3692,
|
2130 |
+
"step": 1510
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 0.813641245972073,
|
2134 |
+
"grad_norm": 2.073859022962504,
|
2135 |
+
"learning_rate": 1.0220801584978408e-06,
|
2136 |
+
"loss": 0.3639,
|
2137 |
+
"step": 1515
|
2138 |
+
},
|
2139 |
+
{
|
2140 |
+
"epoch": 0.8163265306122449,
|
2141 |
+
"grad_norm": 2.033806455267016,
|
2142 |
+
"learning_rate": 9.93847835882296e-07,
|
2143 |
+
"loss": 0.3517,
|
2144 |
+
"step": 1520
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 0.8190118152524167,
|
2148 |
+
"grad_norm": 2.26746303873919,
|
2149 |
+
"learning_rate": 9.659678309077374e-07,
|
2150 |
+
"loss": 0.3516,
|
2151 |
+
"step": 1525
|
2152 |
+
},
|
2153 |
+
{
|
2154 |
+
"epoch": 0.8216970998925887,
|
2155 |
+
"grad_norm": 2.141101672161054,
|
2156 |
+
"learning_rate": 9.384425954574617e-07,
|
2157 |
+
"loss": 0.3607,
|
2158 |
+
"step": 1530
|
2159 |
+
},
|
2160 |
+
{
|
2161 |
+
"epoch": 0.8243823845327605,
|
2162 |
+
"grad_norm": 2.131127304156173,
|
2163 |
+
"learning_rate": 9.112745502148629e-07,
|
2164 |
+
"loss": 0.3618,
|
2165 |
+
"step": 1535
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 0.8270676691729323,
|
2169 |
+
"grad_norm": 2.1498231144464026,
|
2170 |
+
"learning_rate": 8.844660844505498e-07,
|
2171 |
+
"loss": 0.3573,
|
2172 |
+
"step": 1540
|
2173 |
+
},
|
2174 |
+
{
|
2175 |
+
"epoch": 0.8297529538131042,
|
2176 |
+
"grad_norm": 1.9731227944157326,
|
2177 |
+
"learning_rate": 8.580195558122167e-07,
|
2178 |
+
"loss": 0.3691,
|
2179 |
+
"step": 1545
|
2180 |
+
},
|
2181 |
+
{
|
2182 |
+
"epoch": 0.832438238453276,
|
2183 |
+
"grad_norm": 2.068766813819825,
|
2184 |
+
"learning_rate": 8.319372901173045e-07,
|
2185 |
+
"loss": 0.3525,
|
2186 |
+
"step": 1550
|
2187 |
+
},
|
2188 |
+
{
|
2189 |
+
"epoch": 0.835123523093448,
|
2190 |
+
"grad_norm": 2.0577739725964075,
|
2191 |
+
"learning_rate": 8.062215811484652e-07,
|
2192 |
+
"loss": 0.3679,
|
2193 |
+
"step": 1555
|
2194 |
+
},
|
2195 |
+
{
|
2196 |
+
"epoch": 0.8378088077336198,
|
2197 |
+
"grad_norm": 2.0681341001622204,
|
2198 |
+
"learning_rate": 7.808746904518261e-07,
|
2199 |
+
"loss": 0.3589,
|
2200 |
+
"step": 1560
|
2201 |
+
},
|
2202 |
+
{
|
2203 |
+
"epoch": 0.8404940923737916,
|
2204 |
+
"grad_norm": 2.0118053393234216,
|
2205 |
+
"learning_rate": 7.558988471381063e-07,
|
2206 |
+
"loss": 0.3523,
|
2207 |
+
"step": 1565
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 0.8431793770139635,
|
2211 |
+
"grad_norm": 2.0792733438147453,
|
2212 |
+
"learning_rate": 7.312962476865831e-07,
|
2213 |
+
"loss": 0.3485,
|
2214 |
+
"step": 1570
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 0.8458646616541353,
|
2218 |
+
"grad_norm": 2.040693327935837,
|
2219 |
+
"learning_rate": 7.070690557519139e-07,
|
2220 |
+
"loss": 0.3605,
|
2221 |
+
"step": 1575
|
2222 |
+
},
|
2223 |
+
{
|
2224 |
+
"epoch": 0.8485499462943072,
|
2225 |
+
"grad_norm": 2.0885341784359195,
|
2226 |
+
"learning_rate": 6.832194019738686e-07,
|
2227 |
+
"loss": 0.3597,
|
2228 |
+
"step": 1580
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 0.8512352309344791,
|
2232 |
+
"grad_norm": 2.0494324488359315,
|
2233 |
+
"learning_rate": 6.597493837899377e-07,
|
2234 |
+
"loss": 0.334,
|
2235 |
+
"step": 1585
|
2236 |
+
},
|
2237 |
+
{
|
2238 |
+
"epoch": 0.8539205155746509,
|
2239 |
+
"grad_norm": 2.370416793000066,
|
2240 |
+
"learning_rate": 6.366610652508875e-07,
|
2241 |
+
"loss": 0.3699,
|
2242 |
+
"step": 1590
|
2243 |
+
},
|
2244 |
+
{
|
2245 |
+
"epoch": 0.8566058002148228,
|
2246 |
+
"grad_norm": 2.013431987936131,
|
2247 |
+
"learning_rate": 6.13956476839237e-07,
|
2248 |
+
"loss": 0.3433,
|
2249 |
+
"step": 1595
|
2250 |
+
},
|
2251 |
+
{
|
2252 |
+
"epoch": 0.8592910848549946,
|
2253 |
+
"grad_norm": 2.248933148588524,
|
2254 |
+
"learning_rate": 5.916376152906761e-07,
|
2255 |
+
"loss": 0.3619,
|
2256 |
+
"step": 1600
|
2257 |
+
},
|
2258 |
+
{
|
2259 |
+
"epoch": 0.8619763694951665,
|
2260 |
+
"grad_norm": 2.199576156596231,
|
2261 |
+
"learning_rate": 5.697064434184835e-07,
|
2262 |
+
"loss": 0.3585,
|
2263 |
+
"step": 1605
|
2264 |
+
},
|
2265 |
+
{
|
2266 |
+
"epoch": 0.8646616541353384,
|
2267 |
+
"grad_norm": 2.0492788724705564,
|
2268 |
+
"learning_rate": 5.481648899408943e-07,
|
2269 |
+
"loss": 0.351,
|
2270 |
+
"step": 1610
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 0.8673469387755102,
|
2274 |
+
"grad_norm": 2.1643262087841832,
|
2275 |
+
"learning_rate": 5.270148493114896e-07,
|
2276 |
+
"loss": 0.3613,
|
2277 |
+
"step": 1615
|
2278 |
+
},
|
2279 |
+
{
|
2280 |
+
"epoch": 0.8700322234156821,
|
2281 |
+
"grad_norm": 2.0363688955790775,
|
2282 |
+
"learning_rate": 5.062581815525819e-07,
|
2283 |
+
"loss": 0.3536,
|
2284 |
+
"step": 1620
|
2285 |
+
},
|
2286 |
+
{
|
2287 |
+
"epoch": 0.8727175080558539,
|
2288 |
+
"grad_norm": 2.2102662292719226,
|
2289 |
+
"learning_rate": 4.858967120916436e-07,
|
2290 |
+
"loss": 0.3691,
|
2291 |
+
"step": 1625
|
2292 |
+
},
|
2293 |
+
{
|
2294 |
+
"epoch": 0.8754027926960258,
|
2295 |
+
"grad_norm": 2.068954213267209,
|
2296 |
+
"learning_rate": 4.659322316007714e-07,
|
2297 |
+
"loss": 0.3579,
|
2298 |
+
"step": 1630
|
2299 |
+
},
|
2300 |
+
{
|
2301 |
+
"epoch": 0.8780880773361976,
|
2302 |
+
"grad_norm": 2.065009690643561,
|
2303 |
+
"learning_rate": 4.463664958392017e-07,
|
2304 |
+
"loss": 0.3476,
|
2305 |
+
"step": 1635
|
2306 |
+
},
|
2307 |
+
{
|
2308 |
+
"epoch": 0.8807733619763695,
|
2309 |
+
"grad_norm": 2.132923284888097,
|
2310 |
+
"learning_rate": 4.2720122549890607e-07,
|
2311 |
+
"loss": 0.3465,
|
2312 |
+
"step": 1640
|
2313 |
+
},
|
2314 |
+
{
|
2315 |
+
"epoch": 0.8834586466165414,
|
2316 |
+
"grad_norm": 2.0233890775516388,
|
2317 |
+
"learning_rate": 4.084381060532672e-07,
|
2318 |
+
"loss": 0.3436,
|
2319 |
+
"step": 1645
|
2320 |
+
},
|
2321 |
+
{
|
2322 |
+
"epoch": 0.8861439312567132,
|
2323 |
+
"grad_norm": 2.131767022151055,
|
2324 |
+
"learning_rate": 3.9007878760884856e-07,
|
2325 |
+
"loss": 0.3484,
|
2326 |
+
"step": 1650
|
2327 |
+
},
|
2328 |
+
{
|
2329 |
+
"epoch": 0.888829215896885,
|
2330 |
+
"grad_norm": 2.117063081591432,
|
2331 |
+
"learning_rate": 3.721248847602771e-07,
|
2332 |
+
"loss": 0.3557,
|
2333 |
+
"step": 1655
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 0.8915145005370569,
|
2337 |
+
"grad_norm": 2.0080840103832025,
|
2338 |
+
"learning_rate": 3.5457797644825376e-07,
|
2339 |
+
"loss": 0.3575,
|
2340 |
+
"step": 1660
|
2341 |
+
},
|
2342 |
+
{
|
2343 |
+
"epoch": 0.8941997851772288,
|
2344 |
+
"grad_norm": 1.9992195938670811,
|
2345 |
+
"learning_rate": 3.3743960582068745e-07,
|
2346 |
+
"loss": 0.3563,
|
2347 |
+
"step": 1665
|
2348 |
+
},
|
2349 |
+
{
|
2350 |
+
"epoch": 0.8968850698174007,
|
2351 |
+
"grad_norm": 2.0730852522100585,
|
2352 |
+
"learning_rate": 3.207112800969925e-07,
|
2353 |
+
"loss": 0.3673,
|
2354 |
+
"step": 1670
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 0.8995703544575725,
|
2358 |
+
"grad_norm": 2.124531809466262,
|
2359 |
+
"learning_rate": 3.0439447043553005e-07,
|
2360 |
+
"loss": 0.3495,
|
2361 |
+
"step": 1675
|
2362 |
+
},
|
2363 |
+
{
|
2364 |
+
"epoch": 0.9022556390977443,
|
2365 |
+
"grad_norm": 2.149507198420767,
|
2366 |
+
"learning_rate": 2.884906118042313e-07,
|
2367 |
+
"loss": 0.3464,
|
2368 |
+
"step": 1680
|
2369 |
+
},
|
2370 |
+
{
|
2371 |
+
"epoch": 0.9049409237379162,
|
2372 |
+
"grad_norm": 2.1429851911430773,
|
2373 |
+
"learning_rate": 2.730011028544044e-07,
|
2374 |
+
"loss": 0.3539,
|
2375 |
+
"step": 1685
|
2376 |
+
},
|
2377 |
+
{
|
2378 |
+
"epoch": 0.907626208378088,
|
2379 |
+
"grad_norm": 2.033562450055928,
|
2380 |
+
"learning_rate": 2.579273057977244e-07,
|
2381 |
+
"loss": 0.3683,
|
2382 |
+
"step": 1690
|
2383 |
+
},
|
2384 |
+
{
|
2385 |
+
"epoch": 0.91031149301826,
|
2386 |
+
"grad_norm": 2.099140956543066,
|
2387 |
+
"learning_rate": 2.4327054628643764e-07,
|
2388 |
+
"loss": 0.3524,
|
2389 |
+
"step": 1695
|
2390 |
+
},
|
2391 |
+
{
|
2392 |
+
"epoch": 0.9129967776584318,
|
2393 |
+
"grad_norm": 2.0654751576888324,
|
2394 |
+
"learning_rate": 2.29032113296781e-07,
|
2395 |
+
"loss": 0.3441,
|
2396 |
+
"step": 1700
|
2397 |
+
},
|
2398 |
+
{
|
2399 |
+
"epoch": 0.9156820622986036,
|
2400 |
+
"grad_norm": 2.065019806707046,
|
2401 |
+
"learning_rate": 2.1521325901561918e-07,
|
2402 |
+
"loss": 0.3489,
|
2403 |
+
"step": 1705
|
2404 |
+
},
|
2405 |
+
{
|
2406 |
+
"epoch": 0.9183673469387755,
|
2407 |
+
"grad_norm": 2.0595682015060177,
|
2408 |
+
"learning_rate": 2.018151987303263e-07,
|
2409 |
+
"loss": 0.3499,
|
2410 |
+
"step": 1710
|
2411 |
+
},
|
2412 |
+
{
|
2413 |
+
"epoch": 0.9210526315789473,
|
2414 |
+
"grad_norm": 2.1629799218903494,
|
2415 |
+
"learning_rate": 1.8883911072190619e-07,
|
2416 |
+
"loss": 0.354,
|
2417 |
+
"step": 1715
|
2418 |
+
},
|
2419 |
+
{
|
2420 |
+
"epoch": 0.9237379162191193,
|
2421 |
+
"grad_norm": 1.947786552992977,
|
2422 |
+
"learning_rate": 1.7628613616137048e-07,
|
2423 |
+
"loss": 0.3503,
|
2424 |
+
"step": 1720
|
2425 |
+
},
|
2426 |
+
{
|
2427 |
+
"epoch": 0.9264232008592911,
|
2428 |
+
"grad_norm": 2.087692154284924,
|
2429 |
+
"learning_rate": 1.6415737900937811e-07,
|
2430 |
+
"loss": 0.3501,
|
2431 |
+
"step": 1725
|
2432 |
+
},
|
2433 |
+
{
|
2434 |
+
"epoch": 0.9291084854994629,
|
2435 |
+
"grad_norm": 2.2226555122389406,
|
2436 |
+
"learning_rate": 1.5245390591914877e-07,
|
2437 |
+
"loss": 0.3426,
|
2438 |
+
"step": 1730
|
2439 |
+
},
|
2440 |
+
{
|
2441 |
+
"epoch": 0.9317937701396348,
|
2442 |
+
"grad_norm": 1.9281690244795662,
|
2443 |
+
"learning_rate": 1.4117674614265885e-07,
|
2444 |
+
"loss": 0.3598,
|
2445 |
+
"step": 1735
|
2446 |
+
},
|
2447 |
+
{
|
2448 |
+
"epoch": 0.9344790547798066,
|
2449 |
+
"grad_norm": 2.145136219649984,
|
2450 |
+
"learning_rate": 1.3032689144012457e-07,
|
2451 |
+
"loss": 0.3526,
|
2452 |
+
"step": 1740
|
2453 |
+
},
|
2454 |
+
{
|
2455 |
+
"epoch": 0.9371643394199786,
|
2456 |
+
"grad_norm": 2.0780704803664776,
|
2457 |
+
"learning_rate": 1.1990529599277656e-07,
|
2458 |
+
"loss": 0.3478,
|
2459 |
+
"step": 1745
|
2460 |
+
},
|
2461 |
+
{
|
2462 |
+
"epoch": 0.9398496240601504,
|
2463 |
+
"grad_norm": 2.1046408925852664,
|
2464 |
+
"learning_rate": 1.0991287631895498e-07,
|
2465 |
+
"loss": 0.3432,
|
2466 |
+
"step": 1750
|
2467 |
+
},
|
2468 |
+
{
|
2469 |
+
"epoch": 0.9425349087003222,
|
2470 |
+
"grad_norm": 1.8655403585442782,
|
2471 |
+
"learning_rate": 1.0035051119349881e-07,
|
2472 |
+
"loss": 0.3443,
|
2473 |
+
"step": 1755
|
2474 |
+
},
|
2475 |
+
{
|
2476 |
+
"epoch": 0.9452201933404941,
|
2477 |
+
"grad_norm": 1.9887511942254457,
|
2478 |
+
"learning_rate": 9.121904157046779e-08,
|
2479 |
+
"loss": 0.3483,
|
2480 |
+
"step": 1760
|
2481 |
+
},
|
2482 |
+
{
|
2483 |
+
"epoch": 0.9479054779806659,
|
2484 |
+
"grad_norm": 2.07044391687569,
|
2485 |
+
"learning_rate": 8.251927050918262e-08,
|
2486 |
+
"loss": 0.3487,
|
2487 |
+
"step": 1765
|
2488 |
+
},
|
2489 |
+
{
|
2490 |
+
"epoch": 0.9505907626208379,
|
2491 |
+
"grad_norm": 2.01578129872718,
|
2492 |
+
"learning_rate": 7.425196310360094e-08,
|
2493 |
+
"loss": 0.3576,
|
2494 |
+
"step": 1770
|
2495 |
+
},
|
2496 |
+
{
|
2497 |
+
"epoch": 0.9532760472610097,
|
2498 |
+
"grad_norm": 2.1282924163459325,
|
2499 |
+
"learning_rate": 6.64178464150339e-08,
|
2500 |
+
"loss": 0.3531,
|
2501 |
+
"step": 1775
|
2502 |
+
},
|
2503 |
+
{
|
2504 |
+
"epoch": 0.9559613319011815,
|
2505 |
+
"grad_norm": 2.0989453675717407,
|
2506 |
+
"learning_rate": 5.9017609408203445e-08,
|
2507 |
+
"loss": 0.3447,
|
2508 |
+
"step": 1780
|
2509 |
+
},
|
2510 |
+
{
|
2511 |
+
"epoch": 0.9586466165413534,
|
2512 |
+
"grad_norm": 2.0934033811862105,
|
2513 |
+
"learning_rate": 5.2051902890651316e-08,
|
2514 |
+
"loss": 0.3502,
|
2515 |
+
"step": 1785
|
2516 |
+
},
|
2517 |
+
{
|
2518 |
+
"epoch": 0.9613319011815252,
|
2519 |
+
"grad_norm": 2.1433944065391026,
|
2520 |
+
"learning_rate": 4.5521339455507075e-08,
|
2521 |
+
"loss": 0.3575,
|
2522 |
+
"step": 1790
|
2523 |
+
},
|
2524 |
+
{
|
2525 |
+
"epoch": 0.9640171858216972,
|
2526 |
+
"grad_norm": 2.0317297365975806,
|
2527 |
+
"learning_rate": 3.9426493427611177e-08,
|
2528 |
+
"loss": 0.3481,
|
2529 |
+
"step": 1795
|
2530 |
+
},
|
2531 |
+
{
|
2532 |
+
"epoch": 0.966702470461869,
|
2533 |
+
"grad_norm": 2.169192328216417,
|
2534 |
+
"learning_rate": 3.3767900813007623e-08,
|
2535 |
+
"loss": 0.3494,
|
2536 |
+
"step": 1800
|
2537 |
+
},
|
2538 |
+
{
|
2539 |
+
"epoch": 0.9693877551020408,
|
2540 |
+
"grad_norm": 2.0312846504114392,
|
2541 |
+
"learning_rate": 2.8546059251806092e-08,
|
2542 |
+
"loss": 0.3614,
|
2543 |
+
"step": 1805
|
2544 |
+
},
|
2545 |
+
{
|
2546 |
+
"epoch": 0.9720730397422127,
|
2547 |
+
"grad_norm": 2.280532113774367,
|
2548 |
+
"learning_rate": 2.3761427974416383e-08,
|
2549 |
+
"loss": 0.341,
|
2550 |
+
"step": 1810
|
2551 |
+
},
|
2552 |
+
{
|
2553 |
+
"epoch": 0.9747583243823845,
|
2554 |
+
"grad_norm": 2.0782148533974323,
|
2555 |
+
"learning_rate": 1.9414427761162423e-08,
|
2556 |
+
"loss": 0.3356,
|
2557 |
+
"step": 1815
|
2558 |
+
},
|
2559 |
+
{
|
2560 |
+
"epoch": 0.9774436090225563,
|
2561 |
+
"grad_norm": 2.0216470643000695,
|
2562 |
+
"learning_rate": 1.5505440905276835e-08,
|
2563 |
+
"loss": 0.3463,
|
2564 |
+
"step": 1820
|
2565 |
+
},
|
2566 |
+
{
|
2567 |
+
"epoch": 0.9801288936627283,
|
2568 |
+
"grad_norm": 2.005745009335829,
|
2569 |
+
"learning_rate": 1.203481117928007e-08,
|
2570 |
+
"loss": 0.3469,
|
2571 |
+
"step": 1825
|
2572 |
+
},
|
2573 |
+
{
|
2574 |
+
"epoch": 0.9828141783029001,
|
2575 |
+
"grad_norm": 2.091918639115214,
|
2576 |
+
"learning_rate": 9.002843804748473e-09,
|
2577 |
+
"loss": 0.3589,
|
2578 |
+
"step": 1830
|
2579 |
+
},
|
2580 |
+
{
|
2581 |
+
"epoch": 0.985499462943072,
|
2582 |
+
"grad_norm": 1.9926243886806494,
|
2583 |
+
"learning_rate": 6.409805425470761e-09,
|
2584 |
+
"loss": 0.3425,
|
2585 |
+
"step": 1835
|
2586 |
+
},
|
2587 |
+
{
|
2588 |
+
"epoch": 0.9881847475832438,
|
2589 |
+
"grad_norm": 1.9769758283227723,
|
2590 |
+
"learning_rate": 4.255924083999552e-09,
|
2591 |
+
"loss": 0.3621,
|
2592 |
+
"step": 1840
|
2593 |
+
},
|
2594 |
+
{
|
2595 |
+
"epoch": 0.9908700322234156,
|
2596 |
+
"grad_norm": 2.1185976376922038,
|
2597 |
+
"learning_rate": 2.5413892015951903e-09,
|
2598 |
+
"loss": 0.3617,
|
2599 |
+
"step": 1845
|
2600 |
+
},
|
2601 |
+
{
|
2602 |
+
"epoch": 0.9935553168635876,
|
2603 |
+
"grad_norm": 1.9588801206495003,
|
2604 |
+
"learning_rate": 1.266351561568513e-09,
|
2605 |
+
"loss": 0.3582,
|
2606 |
+
"step": 1850
|
2607 |
+
},
|
2608 |
+
{
|
2609 |
+
"epoch": 0.9962406015037594,
|
2610 |
+
"grad_norm": 2.0220772450114204,
|
2611 |
+
"learning_rate": 4.3092329601923664e-10,
|
2612 |
+
"loss": 0.3441,
|
2613 |
+
"step": 1855
|
2614 |
+
},
|
2615 |
+
{
|
2616 |
+
"epoch": 0.9989258861439313,
|
2617 |
+
"grad_norm": 2.0638722610945814,
|
2618 |
+
"learning_rate": 3.517787597440148e-11,
|
2619 |
+
"loss": 0.3507,
|
2620 |
+
"step": 1860
|
2621 |
+
},
|
2622 |
+
{
|
2623 |
+
"epoch": 1.0,
|
2624 |
+
"eval_runtime": 2.6591,
|
2625 |
+
"eval_samples_per_second": 3.761,
|
2626 |
+
"eval_steps_per_second": 1.128,
|
2627 |
+
"step": 1862
|
2628 |
+
},
|
2629 |
+
{
|
2630 |
+
"epoch": 1.0,
|
2631 |
+
"step": 1862,
|
2632 |
+
"total_flos": 194932459438080.0,
|
2633 |
+
"train_loss": 0.5118633475710831,
|
2634 |
+
"train_runtime": 20414.4633,
|
2635 |
+
"train_samples_per_second": 1.459,
|
2636 |
+
"train_steps_per_second": 0.091
|
2637 |
+
}
|
2638 |
+
],
|
2639 |
+
"logging_steps": 5,
|
2640 |
+
"max_steps": 1862,
|
2641 |
+
"num_input_tokens_seen": 0,
|
2642 |
+
"num_train_epochs": 1,
|
2643 |
+
"save_steps": 100,
|
2644 |
+
"stateful_callbacks": {
|
2645 |
+
"TrainerControl": {
|
2646 |
+
"args": {
|
2647 |
+
"should_epoch_stop": false,
|
2648 |
+
"should_evaluate": false,
|
2649 |
+
"should_log": false,
|
2650 |
+
"should_save": true,
|
2651 |
+
"should_training_stop": true
|
2652 |
+
},
|
2653 |
+
"attributes": {}
|
2654 |
+
}
|
2655 |
+
},
|
2656 |
+
"total_flos": 194932459438080.0,
|
2657 |
+
"train_batch_size": 4,
|
2658 |
+
"trial_name": null,
|
2659 |
+
"trial_params": null
|
2660 |
+
}
|