Model save
Browse files- README.md +67 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Aug28_02-15-27_COE-CS-sv003/events.out.tfevents.1724811721.COE-CS-sv003.3122798.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1751 -0
README.md
ADDED
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
+
datasets:
|
9 |
+
- generator
|
10 |
+
model-index:
|
11 |
+
- name: preference_p0.1_seed42_level3_style
|
12 |
+
results: []
|
13 |
+
---
|
14 |
+
|
15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
16 |
+
should probably proofread and complete it, then remove this comment. -->
|
17 |
+
|
18 |
+
# preference_p0.1_seed42_level3_style
|
19 |
+
|
20 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset.
|
21 |
+
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.3492
|
23 |
+
|
24 |
+
## Model description
|
25 |
+
|
26 |
+
More information needed
|
27 |
+
|
28 |
+
## Intended uses & limitations
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Training and evaluation data
|
33 |
+
|
34 |
+
More information needed
|
35 |
+
|
36 |
+
## Training procedure
|
37 |
+
|
38 |
+
### Training hyperparameters
|
39 |
+
|
40 |
+
The following hyperparameters were used during training:
|
41 |
+
- learning_rate: 1e-05
|
42 |
+
- train_batch_size: 2
|
43 |
+
- eval_batch_size: 1
|
44 |
+
- seed: 42
|
45 |
+
- distributed_type: multi-GPU
|
46 |
+
- num_devices: 4
|
47 |
+
- gradient_accumulation_steps: 4
|
48 |
+
- total_train_batch_size: 32
|
49 |
+
- total_eval_batch_size: 4
|
50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
51 |
+
- lr_scheduler_type: cosine
|
52 |
+
- lr_scheduler_warmup_ratio: 0.1
|
53 |
+
- num_epochs: 1
|
54 |
+
|
55 |
+
### Training results
|
56 |
+
|
57 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
58 |
+
|:-------------:|:------:|:----:|:---------------:|
|
59 |
+
| 0.3717 | 0.9998 | 1210 | 0.3492 |
|
60 |
+
|
61 |
+
|
62 |
+
### Framework versions
|
63 |
+
|
64 |
+
- Transformers 4.43.4
|
65 |
+
- Pytorch 2.3.1+cu121
|
66 |
+
- Datasets 2.19.1
|
67 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.999793431109275,
|
3 |
+
"total_flos": 253297038458880.0,
|
4 |
+
"train_loss": 0.5434080246066259,
|
5 |
+
"train_runtime": 27449.6475,
|
6 |
+
"train_samples": 98793,
|
7 |
+
"train_samples_per_second": 1.411,
|
8 |
+
"train_steps_per_second": 0.044
|
9 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"eos_token_id": 2,
|
5 |
+
"transformers_version": "4.43.4"
|
6 |
+
}
|
runs/Aug28_02-15-27_COE-CS-sv003/events.out.tfevents.1724811721.COE-CS-sv003.3122798.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6240043602408294dbee96beacbe82c34637050ea3f8f58d36d0b4e113089511
|
3 |
+
size 57026
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.999793431109275,
|
3 |
+
"total_flos": 253297038458880.0,
|
4 |
+
"train_loss": 0.5434080246066259,
|
5 |
+
"train_runtime": 27449.6475,
|
6 |
+
"train_samples": 98793,
|
7 |
+
"train_samples_per_second": 1.411,
|
8 |
+
"train_steps_per_second": 0.044
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1751 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.999793431109275,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1210,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0008262755629002272,
|
13 |
+
"grad_norm": 23.554044233313164,
|
14 |
+
"learning_rate": 8.264462809917357e-08,
|
15 |
+
"loss": 1.4094,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.004131377814501136,
|
20 |
+
"grad_norm": 23.487881173685214,
|
21 |
+
"learning_rate": 4.132231404958678e-07,
|
22 |
+
"loss": 1.44,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.008262755629002272,
|
27 |
+
"grad_norm": 8.324804089201631,
|
28 |
+
"learning_rate": 8.264462809917356e-07,
|
29 |
+
"loss": 1.3352,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.01239413344350341,
|
34 |
+
"grad_norm": 10.482872108095675,
|
35 |
+
"learning_rate": 1.2396694214876035e-06,
|
36 |
+
"loss": 1.1699,
|
37 |
+
"step": 15
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.016525511258004544,
|
41 |
+
"grad_norm": 3.0017247032336667,
|
42 |
+
"learning_rate": 1.6528925619834712e-06,
|
43 |
+
"loss": 1.0235,
|
44 |
+
"step": 20
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.02065688907250568,
|
48 |
+
"grad_norm": 2.760976256425166,
|
49 |
+
"learning_rate": 2.066115702479339e-06,
|
50 |
+
"loss": 0.9675,
|
51 |
+
"step": 25
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.02478826688700682,
|
55 |
+
"grad_norm": 2.376505582612684,
|
56 |
+
"learning_rate": 2.479338842975207e-06,
|
57 |
+
"loss": 0.9406,
|
58 |
+
"step": 30
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.028919644701507954,
|
62 |
+
"grad_norm": 2.2154227436349765,
|
63 |
+
"learning_rate": 2.8925619834710743e-06,
|
64 |
+
"loss": 0.9302,
|
65 |
+
"step": 35
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.03305102251600909,
|
69 |
+
"grad_norm": 2.3739150674432445,
|
70 |
+
"learning_rate": 3.3057851239669424e-06,
|
71 |
+
"loss": 0.8994,
|
72 |
+
"step": 40
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.03718240033051023,
|
76 |
+
"grad_norm": 2.2109809527940554,
|
77 |
+
"learning_rate": 3.71900826446281e-06,
|
78 |
+
"loss": 0.8982,
|
79 |
+
"step": 45
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.04131377814501136,
|
83 |
+
"grad_norm": 2.28822301545997,
|
84 |
+
"learning_rate": 4.132231404958678e-06,
|
85 |
+
"loss": 0.8837,
|
86 |
+
"step": 50
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.0454451559595125,
|
90 |
+
"grad_norm": 2.2559023546514343,
|
91 |
+
"learning_rate": 4.5454545454545455e-06,
|
92 |
+
"loss": 0.8754,
|
93 |
+
"step": 55
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.04957653377401364,
|
97 |
+
"grad_norm": 2.2137771037984684,
|
98 |
+
"learning_rate": 4.958677685950414e-06,
|
99 |
+
"loss": 0.8582,
|
100 |
+
"step": 60
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.05370791158851477,
|
104 |
+
"grad_norm": 2.373934998656087,
|
105 |
+
"learning_rate": 5.371900826446281e-06,
|
106 |
+
"loss": 0.8593,
|
107 |
+
"step": 65
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.05783928940301591,
|
111 |
+
"grad_norm": 2.467630003052232,
|
112 |
+
"learning_rate": 5.785123966942149e-06,
|
113 |
+
"loss": 0.856,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.06197066721751704,
|
118 |
+
"grad_norm": 2.424099124838778,
|
119 |
+
"learning_rate": 6.198347107438017e-06,
|
120 |
+
"loss": 0.853,
|
121 |
+
"step": 75
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.06610204503201818,
|
125 |
+
"grad_norm": 2.362990499630114,
|
126 |
+
"learning_rate": 6.611570247933885e-06,
|
127 |
+
"loss": 0.8383,
|
128 |
+
"step": 80
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.07023342284651932,
|
132 |
+
"grad_norm": 2.399473277862941,
|
133 |
+
"learning_rate": 7.0247933884297525e-06,
|
134 |
+
"loss": 0.8318,
|
135 |
+
"step": 85
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.07436480066102046,
|
139 |
+
"grad_norm": 2.5700793481674755,
|
140 |
+
"learning_rate": 7.43801652892562e-06,
|
141 |
+
"loss": 0.8303,
|
142 |
+
"step": 90
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.07849617847552158,
|
146 |
+
"grad_norm": 2.3911560791916466,
|
147 |
+
"learning_rate": 7.851239669421489e-06,
|
148 |
+
"loss": 0.8283,
|
149 |
+
"step": 95
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.08262755629002272,
|
153 |
+
"grad_norm": 2.5970628827400057,
|
154 |
+
"learning_rate": 8.264462809917356e-06,
|
155 |
+
"loss": 0.8249,
|
156 |
+
"step": 100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.08675893410452386,
|
160 |
+
"grad_norm": 2.3382293960099143,
|
161 |
+
"learning_rate": 8.677685950413224e-06,
|
162 |
+
"loss": 0.8186,
|
163 |
+
"step": 105
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.090890311919025,
|
167 |
+
"grad_norm": 2.423365953954326,
|
168 |
+
"learning_rate": 9.090909090909091e-06,
|
169 |
+
"loss": 0.8202,
|
170 |
+
"step": 110
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.09502168973352614,
|
174 |
+
"grad_norm": 2.5282725418752694,
|
175 |
+
"learning_rate": 9.50413223140496e-06,
|
176 |
+
"loss": 0.8122,
|
177 |
+
"step": 115
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.09915306754802727,
|
181 |
+
"grad_norm": 2.3639158309512522,
|
182 |
+
"learning_rate": 9.917355371900828e-06,
|
183 |
+
"loss": 0.8127,
|
184 |
+
"step": 120
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.1032844453625284,
|
188 |
+
"grad_norm": 2.844448120421469,
|
189 |
+
"learning_rate": 9.999667111219573e-06,
|
190 |
+
"loss": 0.7962,
|
191 |
+
"step": 125
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.10741582317702954,
|
195 |
+
"grad_norm": 3.443216907128371,
|
196 |
+
"learning_rate": 9.998314826517564e-06,
|
197 |
+
"loss": 0.8031,
|
198 |
+
"step": 130
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.11154720099153068,
|
202 |
+
"grad_norm": 2.681450835682187,
|
203 |
+
"learning_rate": 9.995922621477252e-06,
|
204 |
+
"loss": 0.805,
|
205 |
+
"step": 135
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.11567857880603181,
|
209 |
+
"grad_norm": 2.41859239043235,
|
210 |
+
"learning_rate": 9.99249099380692e-06,
|
211 |
+
"loss": 0.7891,
|
212 |
+
"step": 140
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.11980995662053295,
|
216 |
+
"grad_norm": 2.2320583844472375,
|
217 |
+
"learning_rate": 9.988020657471078e-06,
|
218 |
+
"loss": 0.7945,
|
219 |
+
"step": 145
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.12394133443503408,
|
223 |
+
"grad_norm": 2.6404700133557064,
|
224 |
+
"learning_rate": 9.98251254254193e-06,
|
225 |
+
"loss": 0.7789,
|
226 |
+
"step": 150
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.12807271224953523,
|
230 |
+
"grad_norm": 2.4700012252546855,
|
231 |
+
"learning_rate": 9.97596779500586e-06,
|
232 |
+
"loss": 0.7776,
|
233 |
+
"step": 155
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.13220409006403636,
|
237 |
+
"grad_norm": 2.534567680863635,
|
238 |
+
"learning_rate": 9.968387776525009e-06,
|
239 |
+
"loss": 0.7665,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.13633546787853748,
|
244 |
+
"grad_norm": 2.3195785880081274,
|
245 |
+
"learning_rate": 9.959774064153977e-06,
|
246 |
+
"loss": 0.7633,
|
247 |
+
"step": 165
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.14046684569303863,
|
251 |
+
"grad_norm": 2.6303462252799124,
|
252 |
+
"learning_rate": 9.950128450011706e-06,
|
253 |
+
"loss": 0.7587,
|
254 |
+
"step": 170
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.14459822350753976,
|
258 |
+
"grad_norm": 2.4452240563950687,
|
259 |
+
"learning_rate": 9.939452940908627e-06,
|
260 |
+
"loss": 0.7627,
|
261 |
+
"step": 175
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.1487296013220409,
|
265 |
+
"grad_norm": 2.3656117475844294,
|
266 |
+
"learning_rate": 9.927749757929125e-06,
|
267 |
+
"loss": 0.7546,
|
268 |
+
"step": 180
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.15286097913654204,
|
272 |
+
"grad_norm": 2.307155404533025,
|
273 |
+
"learning_rate": 9.915021335969452e-06,
|
274 |
+
"loss": 0.744,
|
275 |
+
"step": 185
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.15699235695104316,
|
279 |
+
"grad_norm": 2.3398594948394127,
|
280 |
+
"learning_rate": 9.901270323231114e-06,
|
281 |
+
"loss": 0.7423,
|
282 |
+
"step": 190
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.1611237347655443,
|
286 |
+
"grad_norm": 2.3156292065612214,
|
287 |
+
"learning_rate": 9.886499580669917e-06,
|
288 |
+
"loss": 0.7396,
|
289 |
+
"step": 195
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.16525511258004544,
|
293 |
+
"grad_norm": 2.574743453191487,
|
294 |
+
"learning_rate": 9.870712181400726e-06,
|
295 |
+
"loss": 0.7335,
|
296 |
+
"step": 200
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.1693864903945466,
|
300 |
+
"grad_norm": 2.339577898310028,
|
301 |
+
"learning_rate": 9.853911410058097e-06,
|
302 |
+
"loss": 0.7398,
|
303 |
+
"step": 205
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.17351786820904772,
|
307 |
+
"grad_norm": 2.489900169912843,
|
308 |
+
"learning_rate": 9.836100762112887e-06,
|
309 |
+
"loss": 0.7423,
|
310 |
+
"step": 210
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.17764924602354884,
|
314 |
+
"grad_norm": 2.5114661447059365,
|
315 |
+
"learning_rate": 9.817283943145014e-06,
|
316 |
+
"loss": 0.7273,
|
317 |
+
"step": 215
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.18178062383805,
|
321 |
+
"grad_norm": 2.4765147278897097,
|
322 |
+
"learning_rate": 9.797464868072489e-06,
|
323 |
+
"loss": 0.7171,
|
324 |
+
"step": 220
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.18591200165255112,
|
328 |
+
"grad_norm": 2.35952144225102,
|
329 |
+
"learning_rate": 9.776647660336905e-06,
|
330 |
+
"loss": 0.7165,
|
331 |
+
"step": 225
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.19004337946705227,
|
335 |
+
"grad_norm": 2.3928640553968115,
|
336 |
+
"learning_rate": 9.754836651045538e-06,
|
337 |
+
"loss": 0.7228,
|
338 |
+
"step": 230
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.1941747572815534,
|
342 |
+
"grad_norm": 2.359655168950183,
|
343 |
+
"learning_rate": 9.732036378070243e-06,
|
344 |
+
"loss": 0.6999,
|
345 |
+
"step": 235
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.19830613509605455,
|
349 |
+
"grad_norm": 2.7166804592398854,
|
350 |
+
"learning_rate": 9.708251585103324e-06,
|
351 |
+
"loss": 0.6924,
|
352 |
+
"step": 240
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.20243751291055567,
|
356 |
+
"grad_norm": 2.6360301220005415,
|
357 |
+
"learning_rate": 9.683487220670595e-06,
|
358 |
+
"loss": 0.7038,
|
359 |
+
"step": 245
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.2065688907250568,
|
363 |
+
"grad_norm": 2.358814454829314,
|
364 |
+
"learning_rate": 9.657748437101819e-06,
|
365 |
+
"loss": 0.7076,
|
366 |
+
"step": 250
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.21070026853955795,
|
370 |
+
"grad_norm": 2.250240362766412,
|
371 |
+
"learning_rate": 9.631040589458742e-06,
|
372 |
+
"loss": 0.6833,
|
373 |
+
"step": 255
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.21483164635405907,
|
377 |
+
"grad_norm": 2.4709639248779083,
|
378 |
+
"learning_rate": 9.603369234420944e-06,
|
379 |
+
"loss": 0.6751,
|
380 |
+
"step": 260
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.21896302416856023,
|
384 |
+
"grad_norm": 2.315474134768072,
|
385 |
+
"learning_rate": 9.574740129129767e-06,
|
386 |
+
"loss": 0.6988,
|
387 |
+
"step": 265
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.22309440198306135,
|
391 |
+
"grad_norm": 2.2090103807004975,
|
392 |
+
"learning_rate": 9.545159229990493e-06,
|
393 |
+
"loss": 0.6854,
|
394 |
+
"step": 270
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.22722577979756248,
|
398 |
+
"grad_norm": 2.366702578463947,
|
399 |
+
"learning_rate": 9.514632691433108e-06,
|
400 |
+
"loss": 0.6657,
|
401 |
+
"step": 275
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.23135715761206363,
|
405 |
+
"grad_norm": 2.298899028385364,
|
406 |
+
"learning_rate": 9.483166864631837e-06,
|
407 |
+
"loss": 0.663,
|
408 |
+
"step": 280
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.23548853542656475,
|
412 |
+
"grad_norm": 2.2714246758736394,
|
413 |
+
"learning_rate": 9.450768296183764e-06,
|
414 |
+
"loss": 0.6655,
|
415 |
+
"step": 285
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.2396199132410659,
|
419 |
+
"grad_norm": 2.1782463852503438,
|
420 |
+
"learning_rate": 9.417443726746776e-06,
|
421 |
+
"loss": 0.6678,
|
422 |
+
"step": 290
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.24375129105556703,
|
426 |
+
"grad_norm": 2.206125943355958,
|
427 |
+
"learning_rate": 9.383200089637143e-06,
|
428 |
+
"loss": 0.6579,
|
429 |
+
"step": 295
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.24788266887006816,
|
433 |
+
"grad_norm": 2.4822099384962133,
|
434 |
+
"learning_rate": 9.348044509387022e-06,
|
435 |
+
"loss": 0.6688,
|
436 |
+
"step": 300
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.2520140466845693,
|
440 |
+
"grad_norm": 2.6347462586981827,
|
441 |
+
"learning_rate": 9.311984300262151e-06,
|
442 |
+
"loss": 0.6723,
|
443 |
+
"step": 305
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.25614542449907046,
|
447 |
+
"grad_norm": 2.3969886300206915,
|
448 |
+
"learning_rate": 9.275026964740101e-06,
|
449 |
+
"loss": 0.6495,
|
450 |
+
"step": 310
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.2602768023135716,
|
454 |
+
"grad_norm": 2.2255060581251733,
|
455 |
+
"learning_rate": 9.237180191949347e-06,
|
456 |
+
"loss": 0.6548,
|
457 |
+
"step": 315
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.2644081801280727,
|
461 |
+
"grad_norm": 2.4470079917886585,
|
462 |
+
"learning_rate": 9.198451856069514e-06,
|
463 |
+
"loss": 0.6402,
|
464 |
+
"step": 320
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.26853955794257384,
|
468 |
+
"grad_norm": 2.232598978818489,
|
469 |
+
"learning_rate": 9.158850014693123e-06,
|
470 |
+
"loss": 0.6428,
|
471 |
+
"step": 325
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.27267093575707496,
|
475 |
+
"grad_norm": 2.2780042831807807,
|
476 |
+
"learning_rate": 9.118382907149164e-06,
|
477 |
+
"loss": 0.6449,
|
478 |
+
"step": 330
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.27680231357157614,
|
482 |
+
"grad_norm": 2.250070488730356,
|
483 |
+
"learning_rate": 9.077058952788888e-06,
|
484 |
+
"loss": 0.6248,
|
485 |
+
"step": 335
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.28093369138607727,
|
489 |
+
"grad_norm": 2.257817990210915,
|
490 |
+
"learning_rate": 9.034886749234112e-06,
|
491 |
+
"loss": 0.6287,
|
492 |
+
"step": 340
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.2850650692005784,
|
496 |
+
"grad_norm": 2.2753575010897373,
|
497 |
+
"learning_rate": 8.991875070588449e-06,
|
498 |
+
"loss": 0.631,
|
499 |
+
"step": 345
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.2891964470150795,
|
503 |
+
"grad_norm": 2.5652909583991534,
|
504 |
+
"learning_rate": 8.948032865611823e-06,
|
505 |
+
"loss": 0.6232,
|
506 |
+
"step": 350
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.29332782482958064,
|
510 |
+
"grad_norm": 2.3365773173303843,
|
511 |
+
"learning_rate": 8.90336925585864e-06,
|
512 |
+
"loss": 0.608,
|
513 |
+
"step": 355
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.2974592026440818,
|
517 |
+
"grad_norm": 2.27141916228481,
|
518 |
+
"learning_rate": 8.857893533780015e-06,
|
519 |
+
"loss": 0.6048,
|
520 |
+
"step": 360
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.30159058045858295,
|
524 |
+
"grad_norm": 2.121899092032811,
|
525 |
+
"learning_rate": 8.811615160790428e-06,
|
526 |
+
"loss": 0.6192,
|
527 |
+
"step": 365
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.30572195827308407,
|
531 |
+
"grad_norm": 2.210463830947181,
|
532 |
+
"learning_rate": 8.764543765299245e-06,
|
533 |
+
"loss": 0.611,
|
534 |
+
"step": 370
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.3098533360875852,
|
538 |
+
"grad_norm": 2.2581243380183835,
|
539 |
+
"learning_rate": 8.716689140707488e-06,
|
540 |
+
"loss": 0.6116,
|
541 |
+
"step": 375
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.3139847139020863,
|
545 |
+
"grad_norm": 2.291459985719934,
|
546 |
+
"learning_rate": 8.668061243370273e-06,
|
547 |
+
"loss": 0.6094,
|
548 |
+
"step": 380
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.3181160917165875,
|
552 |
+
"grad_norm": 2.242064952311342,
|
553 |
+
"learning_rate": 8.61867019052535e-06,
|
554 |
+
"loss": 0.6076,
|
555 |
+
"step": 385
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.3222474695310886,
|
559 |
+
"grad_norm": 2.3770181909891304,
|
560 |
+
"learning_rate": 8.568526258188172e-06,
|
561 |
+
"loss": 0.6047,
|
562 |
+
"step": 390
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.32637884734558975,
|
566 |
+
"grad_norm": 2.4304957747194593,
|
567 |
+
"learning_rate": 8.517639879013918e-06,
|
568 |
+
"loss": 0.6037,
|
569 |
+
"step": 395
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.3305102251600909,
|
573 |
+
"grad_norm": 2.159929931604691,
|
574 |
+
"learning_rate": 8.466021640126946e-06,
|
575 |
+
"loss": 0.5924,
|
576 |
+
"step": 400
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.334641602974592,
|
580 |
+
"grad_norm": 2.181588993237923,
|
581 |
+
"learning_rate": 8.413682280918094e-06,
|
582 |
+
"loss": 0.5859,
|
583 |
+
"step": 405
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.3387729807890932,
|
587 |
+
"grad_norm": 2.239848195768108,
|
588 |
+
"learning_rate": 8.3606326908103e-06,
|
589 |
+
"loss": 0.5808,
|
590 |
+
"step": 410
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.3429043586035943,
|
594 |
+
"grad_norm": 2.2395305817254165,
|
595 |
+
"learning_rate": 8.306883906993022e-06,
|
596 |
+
"loss": 0.5805,
|
597 |
+
"step": 415
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.34703573641809543,
|
601 |
+
"grad_norm": 2.276771540411031,
|
602 |
+
"learning_rate": 8.25244711212589e-06,
|
603 |
+
"loss": 0.5771,
|
604 |
+
"step": 420
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.35116711423259656,
|
608 |
+
"grad_norm": 2.2443733616231674,
|
609 |
+
"learning_rate": 8.197333632012123e-06,
|
610 |
+
"loss": 0.5724,
|
611 |
+
"step": 425
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.3552984920470977,
|
615 |
+
"grad_norm": 2.266086543452868,
|
616 |
+
"learning_rate": 8.141554933242135e-06,
|
617 |
+
"loss": 0.5766,
|
618 |
+
"step": 430
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.35942986986159886,
|
622 |
+
"grad_norm": 2.3440174185212426,
|
623 |
+
"learning_rate": 8.08512262080787e-06,
|
624 |
+
"loss": 0.5704,
|
625 |
+
"step": 435
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.3635612476761,
|
629 |
+
"grad_norm": 2.231383713024198,
|
630 |
+
"learning_rate": 8.028048435688333e-06,
|
631 |
+
"loss": 0.5672,
|
632 |
+
"step": 440
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.3676926254906011,
|
636 |
+
"grad_norm": 2.210345864732212,
|
637 |
+
"learning_rate": 7.970344252406832e-06,
|
638 |
+
"loss": 0.5654,
|
639 |
+
"step": 445
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.37182400330510224,
|
643 |
+
"grad_norm": 2.201487752141703,
|
644 |
+
"learning_rate": 7.912022076560426e-06,
|
645 |
+
"loss": 0.5608,
|
646 |
+
"step": 450
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.37595538111960336,
|
650 |
+
"grad_norm": 2.2126765922615017,
|
651 |
+
"learning_rate": 7.853094042322121e-06,
|
652 |
+
"loss": 0.5676,
|
653 |
+
"step": 455
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.38008675893410454,
|
657 |
+
"grad_norm": 2.368648057415258,
|
658 |
+
"learning_rate": 7.7935724099163e-06,
|
659 |
+
"loss": 0.5611,
|
660 |
+
"step": 460
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.38421813674860567,
|
664 |
+
"grad_norm": 2.249897659892324,
|
665 |
+
"learning_rate": 7.733469563067928e-06,
|
666 |
+
"loss": 0.5632,
|
667 |
+
"step": 465
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.3883495145631068,
|
671 |
+
"grad_norm": 2.257196050360634,
|
672 |
+
"learning_rate": 7.67279800642607e-06,
|
673 |
+
"loss": 0.5414,
|
674 |
+
"step": 470
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.3924808923776079,
|
678 |
+
"grad_norm": 2.1223073675520125,
|
679 |
+
"learning_rate": 7.611570362962247e-06,
|
680 |
+
"loss": 0.5421,
|
681 |
+
"step": 475
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.3966122701921091,
|
685 |
+
"grad_norm": 2.209118984554094,
|
686 |
+
"learning_rate": 7.549799371344175e-06,
|
687 |
+
"loss": 0.5475,
|
688 |
+
"step": 480
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.4007436480066102,
|
692 |
+
"grad_norm": 2.070603624522537,
|
693 |
+
"learning_rate": 7.487497883285428e-06,
|
694 |
+
"loss": 0.5477,
|
695 |
+
"step": 485
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.40487502582111135,
|
699 |
+
"grad_norm": 2.2718749779596097,
|
700 |
+
"learning_rate": 7.424678860871584e-06,
|
701 |
+
"loss": 0.5384,
|
702 |
+
"step": 490
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.40900640363561247,
|
706 |
+
"grad_norm": 2.347289860613089,
|
707 |
+
"learning_rate": 7.361355373863415e-06,
|
708 |
+
"loss": 0.5407,
|
709 |
+
"step": 495
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.4131377814501136,
|
713 |
+
"grad_norm": 2.491186132317311,
|
714 |
+
"learning_rate": 7.297540596977663e-06,
|
715 |
+
"loss": 0.5393,
|
716 |
+
"step": 500
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.4172691592646148,
|
720 |
+
"grad_norm": 2.3320063032579177,
|
721 |
+
"learning_rate": 7.233247807145989e-06,
|
722 |
+
"loss": 0.5311,
|
723 |
+
"step": 505
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.4214005370791159,
|
727 |
+
"grad_norm": 2.122601227745636,
|
728 |
+
"learning_rate": 7.168490380752648e-06,
|
729 |
+
"loss": 0.537,
|
730 |
+
"step": 510
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.425531914893617,
|
734 |
+
"grad_norm": 2.2214499757842865,
|
735 |
+
"learning_rate": 7.103281790851483e-06,
|
736 |
+
"loss": 0.5338,
|
737 |
+
"step": 515
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.42966329270811815,
|
741 |
+
"grad_norm": 2.1654002355826494,
|
742 |
+
"learning_rate": 7.037635604362786e-06,
|
743 |
+
"loss": 0.5188,
|
744 |
+
"step": 520
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.4337946705226193,
|
748 |
+
"grad_norm": 2.1599888645905847,
|
749 |
+
"learning_rate": 6.971565479250659e-06,
|
750 |
+
"loss": 0.5321,
|
751 |
+
"step": 525
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.43792604833712045,
|
755 |
+
"grad_norm": 2.4521064066537,
|
756 |
+
"learning_rate": 6.905085161681408e-06,
|
757 |
+
"loss": 0.5137,
|
758 |
+
"step": 530
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.4420574261516216,
|
762 |
+
"grad_norm": 2.0735269808191563,
|
763 |
+
"learning_rate": 6.838208483163601e-06,
|
764 |
+
"loss": 0.5181,
|
765 |
+
"step": 535
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.4461888039661227,
|
769 |
+
"grad_norm": 2.1964532327451307,
|
770 |
+
"learning_rate": 6.770949357670358e-06,
|
771 |
+
"loss": 0.5079,
|
772 |
+
"step": 540
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.45032018178062383,
|
776 |
+
"grad_norm": 2.1721038727423996,
|
777 |
+
"learning_rate": 6.703321778744495e-06,
|
778 |
+
"loss": 0.5169,
|
779 |
+
"step": 545
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.45445155959512495,
|
783 |
+
"grad_norm": 2.083001885275941,
|
784 |
+
"learning_rate": 6.635339816587109e-06,
|
785 |
+
"loss": 0.5189,
|
786 |
+
"step": 550
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.45858293740962613,
|
790 |
+
"grad_norm": 2.2406805358253523,
|
791 |
+
"learning_rate": 6.5670176151302136e-06,
|
792 |
+
"loss": 0.5056,
|
793 |
+
"step": 555
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.46271431522412726,
|
797 |
+
"grad_norm": 2.362515256614418,
|
798 |
+
"learning_rate": 6.4983693890940335e-06,
|
799 |
+
"loss": 0.5068,
|
800 |
+
"step": 560
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.4668456930386284,
|
804 |
+
"grad_norm": 2.073329405987291,
|
805 |
+
"learning_rate": 6.4294094210295725e-06,
|
806 |
+
"loss": 0.5146,
|
807 |
+
"step": 565
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.4709770708531295,
|
811 |
+
"grad_norm": 2.0802359451245276,
|
812 |
+
"learning_rate": 6.360152058347068e-06,
|
813 |
+
"loss": 0.4993,
|
814 |
+
"step": 570
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.47510844866763063,
|
818 |
+
"grad_norm": 2.205382348988315,
|
819 |
+
"learning_rate": 6.290611710330957e-06,
|
820 |
+
"loss": 0.5056,
|
821 |
+
"step": 575
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.4792398264821318,
|
825 |
+
"grad_norm": 2.106317834822583,
|
826 |
+
"learning_rate": 6.2208028451419575e-06,
|
827 |
+
"loss": 0.4939,
|
828 |
+
"step": 580
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.48337120429663294,
|
832 |
+
"grad_norm": 2.0997265512356953,
|
833 |
+
"learning_rate": 6.150739986806911e-06,
|
834 |
+
"loss": 0.4904,
|
835 |
+
"step": 585
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.48750258211113406,
|
839 |
+
"grad_norm": 2.087397537314153,
|
840 |
+
"learning_rate": 6.0804377121969985e-06,
|
841 |
+
"loss": 0.4903,
|
842 |
+
"step": 590
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.4916339599256352,
|
846 |
+
"grad_norm": 2.1772365057315834,
|
847 |
+
"learning_rate": 6.009910647994956e-06,
|
848 |
+
"loss": 0.5043,
|
849 |
+
"step": 595
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.4957653377401363,
|
853 |
+
"grad_norm": 2.0357765671221872,
|
854 |
+
"learning_rate": 5.939173467651942e-06,
|
855 |
+
"loss": 0.4916,
|
856 |
+
"step": 600
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.4998967155546375,
|
860 |
+
"grad_norm": 2.140953972933528,
|
861 |
+
"learning_rate": 5.8682408883346535e-06,
|
862 |
+
"loss": 0.4881,
|
863 |
+
"step": 605
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.5040280933691386,
|
867 |
+
"grad_norm": 2.133259953600996,
|
868 |
+
"learning_rate": 5.7971276678633625e-06,
|
869 |
+
"loss": 0.4925,
|
870 |
+
"step": 610
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.5081594711836398,
|
874 |
+
"grad_norm": 2.0897840178036753,
|
875 |
+
"learning_rate": 5.725848601641492e-06,
|
876 |
+
"loss": 0.4909,
|
877 |
+
"step": 615
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.5122908489981409,
|
881 |
+
"grad_norm": 2.1033344267926553,
|
882 |
+
"learning_rate": 5.654418519577369e-06,
|
883 |
+
"loss": 0.4808,
|
884 |
+
"step": 620
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.516422226812642,
|
888 |
+
"grad_norm": 2.0931483592499927,
|
889 |
+
"learning_rate": 5.5828522829987965e-06,
|
890 |
+
"loss": 0.4697,
|
891 |
+
"step": 625
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.5205536046271432,
|
895 |
+
"grad_norm": 2.0194919084584444,
|
896 |
+
"learning_rate": 5.511164781561096e-06,
|
897 |
+
"loss": 0.479,
|
898 |
+
"step": 630
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.5246849824416443,
|
902 |
+
"grad_norm": 2.070247481858063,
|
903 |
+
"learning_rate": 5.439370930149252e-06,
|
904 |
+
"loss": 0.4705,
|
905 |
+
"step": 635
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.5288163602561454,
|
909 |
+
"grad_norm": 2.060542153916691,
|
910 |
+
"learning_rate": 5.367485665774802e-06,
|
911 |
+
"loss": 0.4689,
|
912 |
+
"step": 640
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.5329477380706465,
|
916 |
+
"grad_norm": 2.171769190139748,
|
917 |
+
"learning_rate": 5.295523944468137e-06,
|
918 |
+
"loss": 0.4678,
|
919 |
+
"step": 645
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.5370791158851477,
|
923 |
+
"grad_norm": 2.437387629266701,
|
924 |
+
"learning_rate": 5.223500738166837e-06,
|
925 |
+
"loss": 0.4723,
|
926 |
+
"step": 650
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.5412104936996488,
|
930 |
+
"grad_norm": 2.139213174417897,
|
931 |
+
"learning_rate": 5.1514310316006835e-06,
|
932 |
+
"loss": 0.472,
|
933 |
+
"step": 655
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.5453418715141499,
|
937 |
+
"grad_norm": 2.083008247659044,
|
938 |
+
"learning_rate": 5.07932981917404e-06,
|
939 |
+
"loss": 0.4684,
|
940 |
+
"step": 660
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.5494732493286512,
|
944 |
+
"grad_norm": 2.161242770803335,
|
945 |
+
"learning_rate": 5.007212101846194e-06,
|
946 |
+
"loss": 0.4587,
|
947 |
+
"step": 665
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.5536046271431523,
|
951 |
+
"grad_norm": 2.0500783532272604,
|
952 |
+
"learning_rate": 4.935092884010347e-06,
|
953 |
+
"loss": 0.4611,
|
954 |
+
"step": 670
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.5577360049576534,
|
958 |
+
"grad_norm": 2.1439435869325116,
|
959 |
+
"learning_rate": 4.8629871703718844e-06,
|
960 |
+
"loss": 0.4622,
|
961 |
+
"step": 675
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.5618673827721545,
|
965 |
+
"grad_norm": 2.066743306657874,
|
966 |
+
"learning_rate": 4.7909099628265946e-06,
|
967 |
+
"loss": 0.4672,
|
968 |
+
"step": 680
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.5659987605866557,
|
972 |
+
"grad_norm": 2.0461947863660574,
|
973 |
+
"learning_rate": 4.718876257339444e-06,
|
974 |
+
"loss": 0.4512,
|
975 |
+
"step": 685
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.5701301384011568,
|
979 |
+
"grad_norm": 2.13397891500544,
|
980 |
+
"learning_rate": 4.646901040824622e-06,
|
981 |
+
"loss": 0.4568,
|
982 |
+
"step": 690
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.5742615162156579,
|
986 |
+
"grad_norm": 2.038909744378958,
|
987 |
+
"learning_rate": 4.574999288027437e-06,
|
988 |
+
"loss": 0.4505,
|
989 |
+
"step": 695
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.578392894030159,
|
993 |
+
"grad_norm": 2.061119461177199,
|
994 |
+
"learning_rate": 4.503185958408767e-06,
|
995 |
+
"loss": 0.4536,
|
996 |
+
"step": 700
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.5825242718446602,
|
1000 |
+
"grad_norm": 2.147709730965502,
|
1001 |
+
"learning_rate": 4.431475993032673e-06,
|
1002 |
+
"loss": 0.4567,
|
1003 |
+
"step": 705
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.5866556496591613,
|
1007 |
+
"grad_norm": 2.0471799173895078,
|
1008 |
+
"learning_rate": 4.359884311457857e-06,
|
1009 |
+
"loss": 0.452,
|
1010 |
+
"step": 710
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.5907870274736625,
|
1014 |
+
"grad_norm": 2.096762090340259,
|
1015 |
+
"learning_rate": 4.2884258086335755e-06,
|
1016 |
+
"loss": 0.4407,
|
1017 |
+
"step": 715
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.5949184052881636,
|
1021 |
+
"grad_norm": 2.1776523990414436,
|
1022 |
+
"learning_rate": 4.217115351800693e-06,
|
1023 |
+
"loss": 0.4364,
|
1024 |
+
"step": 720
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.5990497831026648,
|
1028 |
+
"grad_norm": 2.148545496231866,
|
1029 |
+
"learning_rate": 4.145967777398481e-06,
|
1030 |
+
"loss": 0.4407,
|
1031 |
+
"step": 725
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.6031811609171659,
|
1035 |
+
"grad_norm": 2.0382917606214326,
|
1036 |
+
"learning_rate": 4.074997887977843e-06,
|
1037 |
+
"loss": 0.4339,
|
1038 |
+
"step": 730
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.607312538731667,
|
1042 |
+
"grad_norm": 2.096587532898138,
|
1043 |
+
"learning_rate": 4.004220449121574e-06,
|
1044 |
+
"loss": 0.4337,
|
1045 |
+
"step": 735
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.6114439165461681,
|
1049 |
+
"grad_norm": 2.0260734306608272,
|
1050 |
+
"learning_rate": 3.933650186372329e-06,
|
1051 |
+
"loss": 0.4297,
|
1052 |
+
"step": 740
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.6155752943606693,
|
1056 |
+
"grad_norm": 2.0747974830236116,
|
1057 |
+
"learning_rate": 3.863301782168896e-06,
|
1058 |
+
"loss": 0.4362,
|
1059 |
+
"step": 745
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.6197066721751704,
|
1063 |
+
"grad_norm": 2.1542278120116154,
|
1064 |
+
"learning_rate": 3.7931898727914723e-06,
|
1065 |
+
"loss": 0.4413,
|
1066 |
+
"step": 750
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.6238380499896715,
|
1070 |
+
"grad_norm": 1.9775710411271903,
|
1071 |
+
"learning_rate": 3.7233290453165127e-06,
|
1072 |
+
"loss": 0.4273,
|
1073 |
+
"step": 755
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.6279694278041726,
|
1077 |
+
"grad_norm": 2.034389521374345,
|
1078 |
+
"learning_rate": 3.6537338345818273e-06,
|
1079 |
+
"loss": 0.4312,
|
1080 |
+
"step": 760
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.6321008056186739,
|
1084 |
+
"grad_norm": 2.059316170572532,
|
1085 |
+
"learning_rate": 3.5844187201625567e-06,
|
1086 |
+
"loss": 0.4301,
|
1087 |
+
"step": 765
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.636232183433175,
|
1091 |
+
"grad_norm": 2.0806349809281235,
|
1092 |
+
"learning_rate": 3.5153981233586277e-06,
|
1093 |
+
"loss": 0.4234,
|
1094 |
+
"step": 770
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.6403635612476761,
|
1098 |
+
"grad_norm": 2.1474742343641933,
|
1099 |
+
"learning_rate": 3.446686404194337e-06,
|
1100 |
+
"loss": 0.4228,
|
1101 |
+
"step": 775
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.6444949390621773,
|
1105 |
+
"grad_norm": 2.141080032663376,
|
1106 |
+
"learning_rate": 3.3782978584307035e-06,
|
1107 |
+
"loss": 0.4179,
|
1108 |
+
"step": 780
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.6486263168766784,
|
1112 |
+
"grad_norm": 2.0266388142571876,
|
1113 |
+
"learning_rate": 3.310246714591162e-06,
|
1114 |
+
"loss": 0.4197,
|
1115 |
+
"step": 785
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.6527576946911795,
|
1119 |
+
"grad_norm": 2.1015063906736873,
|
1120 |
+
"learning_rate": 3.2425471310012645e-06,
|
1121 |
+
"loss": 0.4224,
|
1122 |
+
"step": 790
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.6568890725056806,
|
1126 |
+
"grad_norm": 2.0688211913933126,
|
1127 |
+
"learning_rate": 3.1752131928429787e-06,
|
1128 |
+
"loss": 0.4148,
|
1129 |
+
"step": 795
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.6610204503201818,
|
1133 |
+
"grad_norm": 2.020342446426012,
|
1134 |
+
"learning_rate": 3.1082589092242116e-06,
|
1135 |
+
"loss": 0.4209,
|
1136 |
+
"step": 800
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.6651518281346829,
|
1140 |
+
"grad_norm": 2.050265450999902,
|
1141 |
+
"learning_rate": 3.041698210264149e-06,
|
1142 |
+
"loss": 0.4086,
|
1143 |
+
"step": 805
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.669283205949184,
|
1147 |
+
"grad_norm": 2.0414704146915117,
|
1148 |
+
"learning_rate": 2.9755449441950434e-06,
|
1149 |
+
"loss": 0.424,
|
1150 |
+
"step": 810
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.6734145837636852,
|
1154 |
+
"grad_norm": 2.0309738852976924,
|
1155 |
+
"learning_rate": 2.9098128744810245e-06,
|
1156 |
+
"loss": 0.4136,
|
1157 |
+
"step": 815
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.6775459615781864,
|
1161 |
+
"grad_norm": 2.123876449787772,
|
1162 |
+
"learning_rate": 2.84451567695456e-06,
|
1163 |
+
"loss": 0.4213,
|
1164 |
+
"step": 820
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.6816773393926875,
|
1168 |
+
"grad_norm": 2.1356187469686043,
|
1169 |
+
"learning_rate": 2.7796669369711294e-06,
|
1170 |
+
"loss": 0.4216,
|
1171 |
+
"step": 825
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.6858087172071886,
|
1175 |
+
"grad_norm": 2.06757675819696,
|
1176 |
+
"learning_rate": 2.715280146582752e-06,
|
1177 |
+
"loss": 0.4192,
|
1178 |
+
"step": 830
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.6899400950216897,
|
1182 |
+
"grad_norm": 2.0954350411098916,
|
1183 |
+
"learning_rate": 2.651368701730889e-06,
|
1184 |
+
"loss": 0.4077,
|
1185 |
+
"step": 835
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.6940714728361909,
|
1189 |
+
"grad_norm": 1.9490264079119275,
|
1190 |
+
"learning_rate": 2.5879458994593786e-06,
|
1191 |
+
"loss": 0.4018,
|
1192 |
+
"step": 840
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.698202850650692,
|
1196 |
+
"grad_norm": 1.9862703131789983,
|
1197 |
+
"learning_rate": 2.5250249351479206e-06,
|
1198 |
+
"loss": 0.4127,
|
1199 |
+
"step": 845
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.7023342284651931,
|
1203 |
+
"grad_norm": 1.9907644914298972,
|
1204 |
+
"learning_rate": 2.4626188997667224e-06,
|
1205 |
+
"loss": 0.4057,
|
1206 |
+
"step": 850
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.7064656062796942,
|
1210 |
+
"grad_norm": 2.0178385478283,
|
1211 |
+
"learning_rate": 2.400740777152874e-06,
|
1212 |
+
"loss": 0.405,
|
1213 |
+
"step": 855
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.7105969840941954,
|
1217 |
+
"grad_norm": 2.0040936929718627,
|
1218 |
+
"learning_rate": 2.3394034413090015e-06,
|
1219 |
+
"loss": 0.4054,
|
1220 |
+
"step": 860
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.7147283619086966,
|
1224 |
+
"grad_norm": 2.0032870121182516,
|
1225 |
+
"learning_rate": 2.278619653724781e-06,
|
1226 |
+
"loss": 0.4025,
|
1227 |
+
"step": 865
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.7188597397231977,
|
1231 |
+
"grad_norm": 2.0315969629289867,
|
1232 |
+
"learning_rate": 2.218402060721845e-06,
|
1233 |
+
"loss": 0.3929,
|
1234 |
+
"step": 870
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.7229911175376988,
|
1238 |
+
"grad_norm": 2.0727282353845626,
|
1239 |
+
"learning_rate": 2.1587631908226812e-06,
|
1240 |
+
"loss": 0.3986,
|
1241 |
+
"step": 875
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.7271224953522,
|
1245 |
+
"grad_norm": 1.9152876057461066,
|
1246 |
+
"learning_rate": 2.09971545214401e-06,
|
1247 |
+
"loss": 0.3915,
|
1248 |
+
"step": 880
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.7312538731667011,
|
1252 |
+
"grad_norm": 2.006764866181322,
|
1253 |
+
"learning_rate": 2.04127112981522e-06,
|
1254 |
+
"loss": 0.3972,
|
1255 |
+
"step": 885
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.7353852509812022,
|
1259 |
+
"grad_norm": 2.009533139724546,
|
1260 |
+
"learning_rate": 1.9834423834224014e-06,
|
1261 |
+
"loss": 0.3952,
|
1262 |
+
"step": 890
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.7395166287957033,
|
1266 |
+
"grad_norm": 1.9136982154678623,
|
1267 |
+
"learning_rate": 1.926241244478496e-06,
|
1268 |
+
"loss": 0.3894,
|
1269 |
+
"step": 895
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.7436480066102045,
|
1273 |
+
"grad_norm": 1.9940424884511188,
|
1274 |
+
"learning_rate": 1.8696796139200852e-06,
|
1275 |
+
"loss": 0.3964,
|
1276 |
+
"step": 900
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.7477793844247056,
|
1280 |
+
"grad_norm": 1.9232507318610614,
|
1281 |
+
"learning_rate": 1.8137692596313528e-06,
|
1282 |
+
"loss": 0.39,
|
1283 |
+
"step": 905
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.7519107622392067,
|
1287 |
+
"grad_norm": 1.9669358048444339,
|
1288 |
+
"learning_rate": 1.7585218139957205e-06,
|
1289 |
+
"loss": 0.4089,
|
1290 |
+
"step": 910
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.756042140053708,
|
1294 |
+
"grad_norm": 2.0275135426829007,
|
1295 |
+
"learning_rate": 1.7039487714756953e-06,
|
1296 |
+
"loss": 0.3897,
|
1297 |
+
"step": 915
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.7601735178682091,
|
1301 |
+
"grad_norm": 1.9145558326260135,
|
1302 |
+
"learning_rate": 1.6500614862213866e-06,
|
1303 |
+
"loss": 0.3845,
|
1304 |
+
"step": 920
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.7643048956827102,
|
1308 |
+
"grad_norm": 1.946476089351215,
|
1309 |
+
"learning_rate": 1.596871169708235e-06,
|
1310 |
+
"loss": 0.3827,
|
1311 |
+
"step": 925
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.7684362734972113,
|
1315 |
+
"grad_norm": 2.061234713938133,
|
1316 |
+
"learning_rate": 1.54438888840441e-06,
|
1317 |
+
"loss": 0.3865,
|
1318 |
+
"step": 930
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.7725676513117125,
|
1322 |
+
"grad_norm": 2.0046087896210993,
|
1323 |
+
"learning_rate": 1.4926255614683931e-06,
|
1324 |
+
"loss": 0.3919,
|
1325 |
+
"step": 935
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.7766990291262136,
|
1329 |
+
"grad_norm": 2.013100573926538,
|
1330 |
+
"learning_rate": 1.4415919584771999e-06,
|
1331 |
+
"loss": 0.3791,
|
1332 |
+
"step": 940
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.7808304069407147,
|
1336 |
+
"grad_norm": 2.052665994161596,
|
1337 |
+
"learning_rate": 1.3912986971857168e-06,
|
1338 |
+
"loss": 0.3813,
|
1339 |
+
"step": 945
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.7849617847552158,
|
1343 |
+
"grad_norm": 1.9665758334089818,
|
1344 |
+
"learning_rate": 1.3417562413176405e-06,
|
1345 |
+
"loss": 0.3743,
|
1346 |
+
"step": 950
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.789093162569717,
|
1350 |
+
"grad_norm": 1.9653082426188921,
|
1351 |
+
"learning_rate": 1.292974898388456e-06,
|
1352 |
+
"loss": 0.3884,
|
1353 |
+
"step": 955
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.7932245403842182,
|
1357 |
+
"grad_norm": 2.1523771874487707,
|
1358 |
+
"learning_rate": 1.2449648175609115e-06,
|
1359 |
+
"loss": 0.383,
|
1360 |
+
"step": 960
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.7973559181987193,
|
1364 |
+
"grad_norm": 1.8532885315782601,
|
1365 |
+
"learning_rate": 1.1977359875334454e-06,
|
1366 |
+
"loss": 0.3812,
|
1367 |
+
"step": 965
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.8014872960132204,
|
1371 |
+
"grad_norm": 2.0616460511941477,
|
1372 |
+
"learning_rate": 1.1512982344619904e-06,
|
1373 |
+
"loss": 0.3742,
|
1374 |
+
"step": 970
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.8056186738277216,
|
1378 |
+
"grad_norm": 2.000103259130858,
|
1379 |
+
"learning_rate": 1.1056612199156093e-06,
|
1380 |
+
"loss": 0.3795,
|
1381 |
+
"step": 975
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.8097500516422227,
|
1385 |
+
"grad_norm": 2.0764477975903404,
|
1386 |
+
"learning_rate": 1.0608344388663576e-06,
|
1387 |
+
"loss": 0.3732,
|
1388 |
+
"step": 980
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.8138814294567238,
|
1392 |
+
"grad_norm": 2.0844020934084315,
|
1393 |
+
"learning_rate": 1.01682721771382e-06,
|
1394 |
+
"loss": 0.3738,
|
1395 |
+
"step": 985
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.8180128072712249,
|
1399 |
+
"grad_norm": 2.050710610881507,
|
1400 |
+
"learning_rate": 9.73648712344707e-07,
|
1401 |
+
"loss": 0.373,
|
1402 |
+
"step": 990
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.8221441850857261,
|
1406 |
+
"grad_norm": 2.068871529971891,
|
1407 |
+
"learning_rate": 9.313079062279429e-07,
|
1408 |
+
"loss": 0.3852,
|
1409 |
+
"step": 995
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.8262755629002272,
|
1413 |
+
"grad_norm": 1.9622479084933468,
|
1414 |
+
"learning_rate": 8.898136085456127e-07,
|
1415 |
+
"loss": 0.3839,
|
1416 |
+
"step": 1000
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.8304069407147283,
|
1420 |
+
"grad_norm": 2.0489239754877913,
|
1421 |
+
"learning_rate": 8.491744523601741e-07,
|
1422 |
+
"loss": 0.3789,
|
1423 |
+
"step": 1005
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.8345383185292296,
|
1427 |
+
"grad_norm": 1.97405744731925,
|
1428 |
+
"learning_rate": 8.093988928183127e-07,
|
1429 |
+
"loss": 0.3659,
|
1430 |
+
"step": 1010
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.8386696963437307,
|
1434 |
+
"grad_norm": 2.0649700701955966,
|
1435 |
+
"learning_rate": 7.70495205391818e-07,
|
1436 |
+
"loss": 0.3684,
|
1437 |
+
"step": 1015
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.8428010741582318,
|
1441 |
+
"grad_norm": 1.9711762707705944,
|
1442 |
+
"learning_rate": 7.324714841558322e-07,
|
1443 |
+
"loss": 0.3728,
|
1444 |
+
"step": 1020
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 0.8469324519727329,
|
1448 |
+
"grad_norm": 2.089515078127623,
|
1449 |
+
"learning_rate": 6.953356401048466e-07,
|
1450 |
+
"loss": 0.361,
|
1451 |
+
"step": 1025
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 0.851063829787234,
|
1455 |
+
"grad_norm": 2.0969140211573536,
|
1456 |
+
"learning_rate": 6.590953995067812e-07,
|
1457 |
+
"loss": 0.3621,
|
1458 |
+
"step": 1030
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 0.8551952076017352,
|
1462 |
+
"grad_norm": 2.014897877209678,
|
1463 |
+
"learning_rate": 6.237583022955079e-07,
|
1464 |
+
"loss": 0.377,
|
1465 |
+
"step": 1035
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 0.8593265854162363,
|
1469 |
+
"grad_norm": 2.0616530107992617,
|
1470 |
+
"learning_rate": 5.89331700502136e-07,
|
1471 |
+
"loss": 0.373,
|
1472 |
+
"step": 1040
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 0.8634579632307374,
|
1476 |
+
"grad_norm": 2.0017552200191164,
|
1477 |
+
"learning_rate": 5.558227567253832e-07,
|
1478 |
+
"loss": 0.3738,
|
1479 |
+
"step": 1045
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 0.8675893410452385,
|
1483 |
+
"grad_norm": 2.0884355733341415,
|
1484 |
+
"learning_rate": 5.232384426413772e-07,
|
1485 |
+
"loss": 0.3625,
|
1486 |
+
"step": 1050
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 0.8717207188597397,
|
1490 |
+
"grad_norm": 2.111748293695631,
|
1491 |
+
"learning_rate": 4.915855375531642e-07,
|
1492 |
+
"loss": 0.3789,
|
1493 |
+
"step": 1055
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.8758520966742409,
|
1497 |
+
"grad_norm": 2.0363025293506305,
|
1498 |
+
"learning_rate": 4.608706269802471e-07,
|
1499 |
+
"loss": 0.3615,
|
1500 |
+
"step": 1060
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 0.879983474488742,
|
1504 |
+
"grad_norm": 2.0861363712497742,
|
1505 |
+
"learning_rate": 4.3110010128843747e-07,
|
1506 |
+
"loss": 0.3803,
|
1507 |
+
"step": 1065
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 0.8841148523032432,
|
1511 |
+
"grad_norm": 1.9794223337927541,
|
1512 |
+
"learning_rate": 4.022801543603194e-07,
|
1513 |
+
"loss": 0.37,
|
1514 |
+
"step": 1070
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 0.8882462301177443,
|
1518 |
+
"grad_norm": 1.976468279453303,
|
1519 |
+
"learning_rate": 3.744167823065814e-07,
|
1520 |
+
"loss": 0.3511,
|
1521 |
+
"step": 1075
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 0.8923776079322454,
|
1525 |
+
"grad_norm": 2.0814002530931583,
|
1526 |
+
"learning_rate": 3.4751578221850377e-07,
|
1527 |
+
"loss": 0.3611,
|
1528 |
+
"step": 1080
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 0.8965089857467465,
|
1532 |
+
"grad_norm": 2.0984819483954844,
|
1533 |
+
"learning_rate": 3.2158275096184744e-07,
|
1534 |
+
"loss": 0.3562,
|
1535 |
+
"step": 1085
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 0.9006403635612477,
|
1539 |
+
"grad_norm": 2.144295712901401,
|
1540 |
+
"learning_rate": 2.966230840124007e-07,
|
1541 |
+
"loss": 0.3674,
|
1542 |
+
"step": 1090
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 0.9047717413757488,
|
1546 |
+
"grad_norm": 2.118969417696939,
|
1547 |
+
"learning_rate": 2.72641974333433e-07,
|
1548 |
+
"loss": 0.3622,
|
1549 |
+
"step": 1095
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 0.9089031191902499,
|
1553 |
+
"grad_norm": 2.0003217874891694,
|
1554 |
+
"learning_rate": 2.4964441129527337e-07,
|
1555 |
+
"loss": 0.3697,
|
1556 |
+
"step": 1100
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.913034497004751,
|
1560 |
+
"grad_norm": 2.052789066548848,
|
1561 |
+
"learning_rate": 2.2763517963725169e-07,
|
1562 |
+
"loss": 0.3709,
|
1563 |
+
"step": 1105
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 0.9171658748192523,
|
1567 |
+
"grad_norm": 1.9424275069967558,
|
1568 |
+
"learning_rate": 2.0661885847221398e-07,
|
1569 |
+
"loss": 0.3672,
|
1570 |
+
"step": 1110
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 0.9212972526337534,
|
1574 |
+
"grad_norm": 2.030858945131136,
|
1575 |
+
"learning_rate": 1.8659982033381928e-07,
|
1576 |
+
"loss": 0.3675,
|
1577 |
+
"step": 1115
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 0.9254286304482545,
|
1581 |
+
"grad_norm": 2.0600092854113132,
|
1582 |
+
"learning_rate": 1.6758223026681507e-07,
|
1583 |
+
"loss": 0.3636,
|
1584 |
+
"step": 1120
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 0.9295600082627556,
|
1588 |
+
"grad_norm": 2.0750384756960374,
|
1589 |
+
"learning_rate": 1.4957004496048256e-07,
|
1590 |
+
"loss": 0.3656,
|
1591 |
+
"step": 1125
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 0.9336913860772568,
|
1595 |
+
"grad_norm": 2.006693676746082,
|
1596 |
+
"learning_rate": 1.3256701192542853e-07,
|
1597 |
+
"loss": 0.3676,
|
1598 |
+
"step": 1130
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 0.9378227638917579,
|
1602 |
+
"grad_norm": 2.114218265869402,
|
1603 |
+
"learning_rate": 1.1657666871390471e-07,
|
1604 |
+
"loss": 0.3658,
|
1605 |
+
"step": 1135
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 0.941954141706259,
|
1609 |
+
"grad_norm": 2.0597089907860537,
|
1610 |
+
"learning_rate": 1.0160234218380305e-07,
|
1611 |
+
"loss": 0.3573,
|
1612 |
+
"step": 1140
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 0.9460855195207601,
|
1616 |
+
"grad_norm": 2.0898079361017334,
|
1617 |
+
"learning_rate": 8.76471478064872e-08,
|
1618 |
+
"loss": 0.3724,
|
1619 |
+
"step": 1145
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 0.9502168973352613,
|
1623 |
+
"grad_norm": 2.0371025770189632,
|
1624 |
+
"learning_rate": 7.471398901860772e-08,
|
1625 |
+
"loss": 0.3625,
|
1626 |
+
"step": 1150
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 0.9543482751497624,
|
1630 |
+
"grad_norm": 1.938944230136909,
|
1631 |
+
"learning_rate": 6.280555661802857e-08,
|
1632 |
+
"loss": 0.3634,
|
1633 |
+
"step": 1155
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 0.9584796529642636,
|
1637 |
+
"grad_norm": 2.1345661558712967,
|
1638 |
+
"learning_rate": 5.192432820399718e-08,
|
1639 |
+
"loss": 0.3674,
|
1640 |
+
"step": 1160
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 0.9626110307787648,
|
1644 |
+
"grad_norm": 1.9759349002447268,
|
1645 |
+
"learning_rate": 4.207256766166845e-08,
|
1646 |
+
"loss": 0.3621,
|
1647 |
+
"step": 1165
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 0.9667424085932659,
|
1651 |
+
"grad_norm": 2.0037876312026213,
|
1652 |
+
"learning_rate": 3.3252324691093185e-08,
|
1653 |
+
"loss": 0.3645,
|
1654 |
+
"step": 1170
|
1655 |
+
},
|
1656 |
+
{
|
1657 |
+
"epoch": 0.970873786407767,
|
1658 |
+
"grad_norm": 2.059268789116945,
|
1659 |
+
"learning_rate": 2.546543438077087e-08,
|
1660 |
+
"loss": 0.3569,
|
1661 |
+
"step": 1175
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 0.9750051642222681,
|
1665 |
+
"grad_norm": 2.1019547322975916,
|
1666 |
+
"learning_rate": 1.8713516825851207e-08,
|
1667 |
+
"loss": 0.3604,
|
1668 |
+
"step": 1180
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 0.9791365420367693,
|
1672 |
+
"grad_norm": 2.067476136669,
|
1673 |
+
"learning_rate": 1.2997976791065403e-08,
|
1674 |
+
"loss": 0.3611,
|
1675 |
+
"step": 1185
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 0.9832679198512704,
|
1679 |
+
"grad_norm": 1.970947575751658,
|
1680 |
+
"learning_rate": 8.32000341846162e-09,
|
1681 |
+
"loss": 0.3626,
|
1682 |
+
"step": 1190
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.9873992976657715,
|
1686 |
+
"grad_norm": 2.0213770050103816,
|
1687 |
+
"learning_rate": 4.6805699799967744e-09,
|
1688 |
+
"loss": 0.3577,
|
1689 |
+
"step": 1195
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 0.9915306754802726,
|
1693 |
+
"grad_norm": 2.0031463180853915,
|
1694 |
+
"learning_rate": 2.0804336750429588e-09,
|
1695 |
+
"loss": 0.3594,
|
1696 |
+
"step": 1200
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 0.9956620532947738,
|
1700 |
+
"grad_norm": 2.0391505951802165,
|
1701 |
+
"learning_rate": 5.201354728517905e-10,
|
1702 |
+
"loss": 0.3624,
|
1703 |
+
"step": 1205
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 0.999793431109275,
|
1707 |
+
"grad_norm": 2.0674337207917235,
|
1708 |
+
"learning_rate": 0.0,
|
1709 |
+
"loss": 0.3717,
|
1710 |
+
"step": 1210
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 0.999793431109275,
|
1714 |
+
"eval_loss": 0.3491726517677307,
|
1715 |
+
"eval_runtime": 125.1809,
|
1716 |
+
"eval_samples_per_second": 3.115,
|
1717 |
+
"eval_steps_per_second": 0.783,
|
1718 |
+
"step": 1210
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 0.999793431109275,
|
1722 |
+
"step": 1210,
|
1723 |
+
"total_flos": 253297038458880.0,
|
1724 |
+
"train_loss": 0.5434080246066259,
|
1725 |
+
"train_runtime": 27449.6475,
|
1726 |
+
"train_samples_per_second": 1.411,
|
1727 |
+
"train_steps_per_second": 0.044
|
1728 |
+
}
|
1729 |
+
],
|
1730 |
+
"logging_steps": 5,
|
1731 |
+
"max_steps": 1210,
|
1732 |
+
"num_input_tokens_seen": 0,
|
1733 |
+
"num_train_epochs": 1,
|
1734 |
+
"save_steps": 100,
|
1735 |
+
"stateful_callbacks": {
|
1736 |
+
"TrainerControl": {
|
1737 |
+
"args": {
|
1738 |
+
"should_epoch_stop": false,
|
1739 |
+
"should_evaluate": false,
|
1740 |
+
"should_log": false,
|
1741 |
+
"should_save": true,
|
1742 |
+
"should_training_stop": true
|
1743 |
+
},
|
1744 |
+
"attributes": {}
|
1745 |
+
}
|
1746 |
+
},
|
1747 |
+
"total_flos": 253297038458880.0,
|
1748 |
+
"train_batch_size": 2,
|
1749 |
+
"trial_name": null,
|
1750 |
+
"trial_params": null
|
1751 |
+
}
|