Commit
·
e49d74a
1
Parent(s):
01fb97f
Add evaluation results and output files to model card
Browse files- README.md +33 -0
- checkpoint-228/config.json +22 -0
- checkpoint-228/model.safetensors +3 -0
- checkpoint-228/optimizer.pt +3 -0
- checkpoint-228/preprocessor_config.json +22 -0
- checkpoint-228/rng_state.pth +3 -0
- checkpoint-228/scheduler.pt +3 -0
- checkpoint-228/trainer_state.json +1857 -0
- checkpoint-228/training_args.bin +3 -0
- config.json +22 -0
- model.safetensors +3 -0
- pr_curve.png +0 -0
- preprocessor_config.json +22 -0
- roc_curve.png +0 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language: code
|
3 |
+
tags:
|
4 |
+
- binary-classification
|
5 |
+
- model
|
6 |
+
- evaluation
|
7 |
+
metrics:
|
8 |
+
- average_precision: 0.74
|
9 |
+
- roc_auc: 0.84
|
10 |
+
---
|
11 |
+
|
12 |
+
# Binary Classification Model
|
13 |
+
|
14 |
+
## Evaluation Results
|
15 |
+
|
16 |
+
**Average Precision:** 0.74
|
17 |
+
**ROC AUC:** 0.84
|
18 |
+
|
19 |
+
## Visualizations
|
20 |
+
|
21 |
+
### Precision-Recall Curve
|
22 |
+

|
23 |
+
|
24 |
+
### ROC Curve
|
25 |
+

|
26 |
+
|
27 |
+
## Output Files and Directories
|
28 |
+
|
29 |
+
- 📂 `checkpoint-171/`
|
30 |
+
- `config.json`
|
31 |
+
- `model.safetensors`
|
32 |
+
- `preprocessor_config.json`
|
33 |
+
- `training_args.bin`
|
checkpoint-228/config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"ViTForImageClassification"
|
4 |
+
],
|
5 |
+
"attention_probs_dropout_prob": 0.0,
|
6 |
+
"encoder_stride": 16,
|
7 |
+
"hidden_act": "gelu",
|
8 |
+
"hidden_dropout_prob": 0.0,
|
9 |
+
"hidden_size": 256,
|
10 |
+
"image_size": 32,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 512,
|
13 |
+
"layer_norm_eps": 1e-12,
|
14 |
+
"model_type": "vit",
|
15 |
+
"num_attention_heads": 8,
|
16 |
+
"num_channels": 3,
|
17 |
+
"num_hidden_layers": 4,
|
18 |
+
"patch_size": 4,
|
19 |
+
"qkv_bias": true,
|
20 |
+
"torch_dtype": "float32",
|
21 |
+
"transformers_version": "4.46.1"
|
22 |
+
}
|
checkpoint-228/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fdb0e0ecad02f76f806820610e622dfa4e5ec25ccb22820e9030eb4aea09a13b
|
3 |
+
size 8563512
|
checkpoint-228/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1f15b02a3873cfd9da1b562e3528561911b7c8880303e68589c7e3771500297a
|
3 |
+
size 17153530
|
checkpoint-228/preprocessor_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"do_rescale": true,
|
4 |
+
"do_resize": false,
|
5 |
+
"image_mean": [
|
6 |
+
0.5,
|
7 |
+
0.5,
|
8 |
+
0.5
|
9 |
+
],
|
10 |
+
"image_processor_type": "ViTFeatureExtractor",
|
11 |
+
"image_std": [
|
12 |
+
0.5,
|
13 |
+
0.5,
|
14 |
+
0.5
|
15 |
+
],
|
16 |
+
"resample": 2,
|
17 |
+
"rescale_factor": 0.00392156862745098,
|
18 |
+
"size": {
|
19 |
+
"height": 32,
|
20 |
+
"width": 32
|
21 |
+
}
|
22 |
+
}
|
checkpoint-228/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d97a55cedc4d48798b577555ae4c3e3736f22ab399c66ba765115d5525fc356e
|
3 |
+
size 14244
|
checkpoint-228/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:507eea7b64305959b735e5710fe0927be568ba15ffe8cde09ad9936dd5f673e6
|
3 |
+
size 1000
|
checkpoint-228/trainer_state.json
ADDED
@@ -0,0 +1,1857 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.8388252148997135,
|
3 |
+
"best_model_checkpoint": "/tmp/logs/binary_classification_model_v3.1.3_junctions/checkpoint-228",
|
4 |
+
"epoch": 19.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 228,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.08333333333333333,
|
13 |
+
"grad_norm": 13.008352279663086,
|
14 |
+
"learning_rate": 2.0669698222405954e-07,
|
15 |
+
"loss": 1.4157,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.16666666666666666,
|
20 |
+
"grad_norm": 13.666772842407227,
|
21 |
+
"learning_rate": 4.133939644481191e-07,
|
22 |
+
"loss": 1.4228,
|
23 |
+
"step": 2
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.25,
|
27 |
+
"grad_norm": 18.011869430541992,
|
28 |
+
"learning_rate": 6.200909466721785e-07,
|
29 |
+
"loss": 1.4609,
|
30 |
+
"step": 3
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.3333333333333333,
|
34 |
+
"grad_norm": 16.066177368164062,
|
35 |
+
"learning_rate": 8.267879288962382e-07,
|
36 |
+
"loss": 1.4447,
|
37 |
+
"step": 4
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.4166666666666667,
|
41 |
+
"grad_norm": 17.739500045776367,
|
42 |
+
"learning_rate": 1.0334849111202977e-06,
|
43 |
+
"loss": 1.4561,
|
44 |
+
"step": 5
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.5,
|
48 |
+
"grad_norm": 12.701050758361816,
|
49 |
+
"learning_rate": 1.240181893344357e-06,
|
50 |
+
"loss": 1.4118,
|
51 |
+
"step": 6
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.5833333333333334,
|
55 |
+
"grad_norm": 12.726056098937988,
|
56 |
+
"learning_rate": 1.4468788755684167e-06,
|
57 |
+
"loss": 1.4086,
|
58 |
+
"step": 7
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.6666666666666666,
|
62 |
+
"grad_norm": 11.069748878479004,
|
63 |
+
"learning_rate": 1.6535758577924763e-06,
|
64 |
+
"loss": 1.3948,
|
65 |
+
"step": 8
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.75,
|
69 |
+
"grad_norm": 12.336809158325195,
|
70 |
+
"learning_rate": 1.860272840016536e-06,
|
71 |
+
"loss": 1.4037,
|
72 |
+
"step": 9
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.8333333333333334,
|
76 |
+
"grad_norm": 11.593267440795898,
|
77 |
+
"learning_rate": 2.0669698222405953e-06,
|
78 |
+
"loss": 1.3902,
|
79 |
+
"step": 10
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.9166666666666666,
|
83 |
+
"grad_norm": 5.577626705169678,
|
84 |
+
"learning_rate": 2.273666804464655e-06,
|
85 |
+
"loss": 1.3633,
|
86 |
+
"step": 11
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 1.0,
|
90 |
+
"grad_norm": 4.752732753753662,
|
91 |
+
"learning_rate": 2.480363786688714e-06,
|
92 |
+
"loss": 1.3461,
|
93 |
+
"step": 12
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 1.0,
|
97 |
+
"eval_accuracy": 0.4973137535816619,
|
98 |
+
"eval_f1": 0.002132954141485958,
|
99 |
+
"eval_loss": 0.67629474401474,
|
100 |
+
"eval_precision": 0.6,
|
101 |
+
"eval_recall": 0.0010683760683760685,
|
102 |
+
"eval_runtime": 0.7826,
|
103 |
+
"eval_samples_per_second": 7135.45,
|
104 |
+
"eval_steps_per_second": 14.056,
|
105 |
+
"step": 12
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 1.0833333333333333,
|
109 |
+
"grad_norm": 5.861985206604004,
|
110 |
+
"learning_rate": 2.6870607689127738e-06,
|
111 |
+
"loss": 1.36,
|
112 |
+
"step": 13
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 1.1666666666666667,
|
116 |
+
"grad_norm": 4.5783538818359375,
|
117 |
+
"learning_rate": 2.8937577511368334e-06,
|
118 |
+
"loss": 1.3491,
|
119 |
+
"step": 14
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 1.25,
|
123 |
+
"grad_norm": 4.950338363647461,
|
124 |
+
"learning_rate": 3.100454733360893e-06,
|
125 |
+
"loss": 1.3413,
|
126 |
+
"step": 15
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 1.3333333333333333,
|
130 |
+
"grad_norm": 4.519286155700684,
|
131 |
+
"learning_rate": 3.3071517155849526e-06,
|
132 |
+
"loss": 1.3398,
|
133 |
+
"step": 16
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 1.4166666666666667,
|
137 |
+
"grad_norm": 3.9378013610839844,
|
138 |
+
"learning_rate": 3.5138486978090123e-06,
|
139 |
+
"loss": 1.3361,
|
140 |
+
"step": 17
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 1.5,
|
144 |
+
"grad_norm": 7.235836029052734,
|
145 |
+
"learning_rate": 3.720545680033072e-06,
|
146 |
+
"loss": 1.3378,
|
147 |
+
"step": 18
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 1.5833333333333335,
|
151 |
+
"grad_norm": 5.351963996887207,
|
152 |
+
"learning_rate": 3.927242662257131e-06,
|
153 |
+
"loss": 1.318,
|
154 |
+
"step": 19
|
155 |
+
},
|
156 |
+
{
|
157 |
+
"epoch": 1.6666666666666665,
|
158 |
+
"grad_norm": 8.636580467224121,
|
159 |
+
"learning_rate": 4.133939644481191e-06,
|
160 |
+
"loss": 1.3278,
|
161 |
+
"step": 20
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 1.75,
|
165 |
+
"grad_norm": 5.724945545196533,
|
166 |
+
"learning_rate": 4.34063662670525e-06,
|
167 |
+
"loss": 1.3097,
|
168 |
+
"step": 21
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 1.8333333333333335,
|
172 |
+
"grad_norm": 5.427116870880127,
|
173 |
+
"learning_rate": 4.54733360892931e-06,
|
174 |
+
"loss": 1.297,
|
175 |
+
"step": 22
|
176 |
+
},
|
177 |
+
{
|
178 |
+
"epoch": 1.9166666666666665,
|
179 |
+
"grad_norm": 3.9907078742980957,
|
180 |
+
"learning_rate": 4.754030591153369e-06,
|
181 |
+
"loss": 1.2955,
|
182 |
+
"step": 23
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 2.0,
|
186 |
+
"grad_norm": 4.581727981567383,
|
187 |
+
"learning_rate": 4.960727573377428e-06,
|
188 |
+
"loss": 1.2784,
|
189 |
+
"step": 24
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 2.0,
|
193 |
+
"eval_accuracy": 0.5134312320916905,
|
194 |
+
"eval_f1": 0.1642571516456475,
|
195 |
+
"eval_loss": 0.6319519877433777,
|
196 |
+
"eval_precision": 0.6027088036117382,
|
197 |
+
"eval_recall": 0.09508547008547008,
|
198 |
+
"eval_runtime": 0.8299,
|
199 |
+
"eval_samples_per_second": 6728.655,
|
200 |
+
"eval_steps_per_second": 13.255,
|
201 |
+
"step": 24
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 2.0833333333333335,
|
205 |
+
"grad_norm": 3.7912609577178955,
|
206 |
+
"learning_rate": 5.167424555601488e-06,
|
207 |
+
"loss": 1.2751,
|
208 |
+
"step": 25
|
209 |
+
},
|
210 |
+
{
|
211 |
+
"epoch": 2.1666666666666665,
|
212 |
+
"grad_norm": 4.529625415802002,
|
213 |
+
"learning_rate": 5.3741215378255476e-06,
|
214 |
+
"loss": 1.2557,
|
215 |
+
"step": 26
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 2.25,
|
219 |
+
"grad_norm": 4.522603988647461,
|
220 |
+
"learning_rate": 5.580818520049608e-06,
|
221 |
+
"loss": 1.2458,
|
222 |
+
"step": 27
|
223 |
+
},
|
224 |
+
{
|
225 |
+
"epoch": 2.3333333333333335,
|
226 |
+
"grad_norm": 3.9087917804718018,
|
227 |
+
"learning_rate": 5.787515502273667e-06,
|
228 |
+
"loss": 1.2485,
|
229 |
+
"step": 28
|
230 |
+
},
|
231 |
+
{
|
232 |
+
"epoch": 2.4166666666666665,
|
233 |
+
"grad_norm": 3.348869800567627,
|
234 |
+
"learning_rate": 5.994212484497727e-06,
|
235 |
+
"loss": 1.2404,
|
236 |
+
"step": 29
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 2.5,
|
240 |
+
"grad_norm": 4.038131237030029,
|
241 |
+
"learning_rate": 6.200909466721786e-06,
|
242 |
+
"loss": 1.2055,
|
243 |
+
"step": 30
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 2.5833333333333335,
|
247 |
+
"grad_norm": 5.1803388595581055,
|
248 |
+
"learning_rate": 6.407606448945845e-06,
|
249 |
+
"loss": 1.1886,
|
250 |
+
"step": 31
|
251 |
+
},
|
252 |
+
{
|
253 |
+
"epoch": 2.6666666666666665,
|
254 |
+
"grad_norm": 4.889699935913086,
|
255 |
+
"learning_rate": 6.614303431169905e-06,
|
256 |
+
"loss": 1.1927,
|
257 |
+
"step": 32
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 2.75,
|
261 |
+
"grad_norm": 3.525738000869751,
|
262 |
+
"learning_rate": 6.8210004133939645e-06,
|
263 |
+
"loss": 1.1978,
|
264 |
+
"step": 33
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 2.8333333333333335,
|
268 |
+
"grad_norm": 5.528476715087891,
|
269 |
+
"learning_rate": 7.0276973956180245e-06,
|
270 |
+
"loss": 1.1413,
|
271 |
+
"step": 34
|
272 |
+
},
|
273 |
+
{
|
274 |
+
"epoch": 2.9166666666666665,
|
275 |
+
"grad_norm": 3.697988271713257,
|
276 |
+
"learning_rate": 7.234394377842084e-06,
|
277 |
+
"loss": 1.1407,
|
278 |
+
"step": 35
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 3.0,
|
282 |
+
"grad_norm": 3.308899164199829,
|
283 |
+
"learning_rate": 7.441091360066144e-06,
|
284 |
+
"loss": 1.1521,
|
285 |
+
"step": 36
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 3.0,
|
289 |
+
"eval_accuracy": 0.7224212034383954,
|
290 |
+
"eval_f1": 0.7198120028922632,
|
291 |
+
"eval_loss": 0.5532130599021912,
|
292 |
+
"eval_precision": 0.7309104258443465,
|
293 |
+
"eval_recall": 0.709045584045584,
|
294 |
+
"eval_runtime": 0.8201,
|
295 |
+
"eval_samples_per_second": 6808.741,
|
296 |
+
"eval_steps_per_second": 13.413,
|
297 |
+
"step": 36
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 3.0833333333333335,
|
301 |
+
"grad_norm": 3.082005500793457,
|
302 |
+
"learning_rate": 7.647788342290202e-06,
|
303 |
+
"loss": 1.1304,
|
304 |
+
"step": 37
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 3.1666666666666665,
|
308 |
+
"grad_norm": 5.744527339935303,
|
309 |
+
"learning_rate": 7.854485324514261e-06,
|
310 |
+
"loss": 1.1228,
|
311 |
+
"step": 38
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 3.25,
|
315 |
+
"grad_norm": 6.02752161026001,
|
316 |
+
"learning_rate": 8.061182306738322e-06,
|
317 |
+
"loss": 1.0696,
|
318 |
+
"step": 39
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 3.3333333333333335,
|
322 |
+
"grad_norm": 6.481032848358154,
|
323 |
+
"learning_rate": 8.267879288962381e-06,
|
324 |
+
"loss": 1.0761,
|
325 |
+
"step": 40
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 3.4166666666666665,
|
329 |
+
"grad_norm": 2.9695193767547607,
|
330 |
+
"learning_rate": 8.47457627118644e-06,
|
331 |
+
"loss": 1.0528,
|
332 |
+
"step": 41
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 3.5,
|
336 |
+
"grad_norm": 3.486065626144409,
|
337 |
+
"learning_rate": 8.6812732534105e-06,
|
338 |
+
"loss": 1.0842,
|
339 |
+
"step": 42
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 3.5833333333333335,
|
343 |
+
"grad_norm": 4.507181167602539,
|
344 |
+
"learning_rate": 8.88797023563456e-06,
|
345 |
+
"loss": 1.0332,
|
346 |
+
"step": 43
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 3.6666666666666665,
|
350 |
+
"grad_norm": 2.18536639213562,
|
351 |
+
"learning_rate": 9.09466721785862e-06,
|
352 |
+
"loss": 1.0743,
|
353 |
+
"step": 44
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 3.75,
|
357 |
+
"grad_norm": 2.1089820861816406,
|
358 |
+
"learning_rate": 9.301364200082679e-06,
|
359 |
+
"loss": 1.0433,
|
360 |
+
"step": 45
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 3.8333333333333335,
|
364 |
+
"grad_norm": 2.000920534133911,
|
365 |
+
"learning_rate": 9.508061182306738e-06,
|
366 |
+
"loss": 1.0042,
|
367 |
+
"step": 46
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 3.9166666666666665,
|
371 |
+
"grad_norm": 3.83160662651062,
|
372 |
+
"learning_rate": 9.714758164530797e-06,
|
373 |
+
"loss": 0.9925,
|
374 |
+
"step": 47
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 4.0,
|
378 |
+
"grad_norm": 5.499476432800293,
|
379 |
+
"learning_rate": 9.921455146754857e-06,
|
380 |
+
"loss": 0.9794,
|
381 |
+
"step": 48
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 4.0,
|
385 |
+
"eval_accuracy": 0.7816977077363897,
|
386 |
+
"eval_f1": 0.8033553798999838,
|
387 |
+
"eval_loss": 0.49034950137138367,
|
388 |
+
"eval_precision": 0.7342966676496608,
|
389 |
+
"eval_recall": 0.8867521367521367,
|
390 |
+
"eval_runtime": 0.8092,
|
391 |
+
"eval_samples_per_second": 6901.055,
|
392 |
+
"eval_steps_per_second": 13.594,
|
393 |
+
"step": 48
|
394 |
+
},
|
395 |
+
{
|
396 |
+
"epoch": 4.083333333333333,
|
397 |
+
"grad_norm": 4.450442314147949,
|
398 |
+
"learning_rate": 1.0128152128978916e-05,
|
399 |
+
"loss": 1.004,
|
400 |
+
"step": 49
|
401 |
+
},
|
402 |
+
{
|
403 |
+
"epoch": 4.166666666666667,
|
404 |
+
"grad_norm": 6.789775371551514,
|
405 |
+
"learning_rate": 1.0334849111202977e-05,
|
406 |
+
"loss": 0.9547,
|
407 |
+
"step": 50
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 4.25,
|
411 |
+
"grad_norm": 10.135536193847656,
|
412 |
+
"learning_rate": 1.0541546093427036e-05,
|
413 |
+
"loss": 1.0277,
|
414 |
+
"step": 51
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 4.333333333333333,
|
418 |
+
"grad_norm": 7.634149551391602,
|
419 |
+
"learning_rate": 1.0748243075651095e-05,
|
420 |
+
"loss": 0.9353,
|
421 |
+
"step": 52
|
422 |
+
},
|
423 |
+
{
|
424 |
+
"epoch": 4.416666666666667,
|
425 |
+
"grad_norm": 4.420949459075928,
|
426 |
+
"learning_rate": 1.0954940057875154e-05,
|
427 |
+
"loss": 0.9919,
|
428 |
+
"step": 53
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 4.5,
|
432 |
+
"grad_norm": 3.7844886779785156,
|
433 |
+
"learning_rate": 1.1161637040099215e-05,
|
434 |
+
"loss": 0.9805,
|
435 |
+
"step": 54
|
436 |
+
},
|
437 |
+
{
|
438 |
+
"epoch": 4.583333333333333,
|
439 |
+
"grad_norm": 1.2264916896820068,
|
440 |
+
"learning_rate": 1.1368334022323274e-05,
|
441 |
+
"loss": 0.9856,
|
442 |
+
"step": 55
|
443 |
+
},
|
444 |
+
{
|
445 |
+
"epoch": 4.666666666666667,
|
446 |
+
"grad_norm": 11.561905860900879,
|
447 |
+
"learning_rate": 1.1575031004547334e-05,
|
448 |
+
"loss": 1.0431,
|
449 |
+
"step": 56
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 4.75,
|
453 |
+
"grad_norm": 5.218517303466797,
|
454 |
+
"learning_rate": 1.1781727986771393e-05,
|
455 |
+
"loss": 0.9875,
|
456 |
+
"step": 57
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 4.833333333333333,
|
460 |
+
"grad_norm": 8.459822654724121,
|
461 |
+
"learning_rate": 1.1988424968995454e-05,
|
462 |
+
"loss": 0.9652,
|
463 |
+
"step": 58
|
464 |
+
},
|
465 |
+
{
|
466 |
+
"epoch": 4.916666666666667,
|
467 |
+
"grad_norm": 2.2815685272216797,
|
468 |
+
"learning_rate": 1.2195121951219513e-05,
|
469 |
+
"loss": 0.9794,
|
470 |
+
"step": 59
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 5.0,
|
474 |
+
"grad_norm": 9.657059669494629,
|
475 |
+
"learning_rate": 1.2401818933443572e-05,
|
476 |
+
"loss": 0.9521,
|
477 |
+
"step": 60
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 5.0,
|
481 |
+
"eval_accuracy": 0.7247492836676218,
|
482 |
+
"eval_f1": 0.723511422917791,
|
483 |
+
"eval_loss": 0.4798118770122528,
|
484 |
+
"eval_precision": 0.7310069065794257,
|
485 |
+
"eval_recall": 0.7161680911680912,
|
486 |
+
"eval_runtime": 0.8493,
|
487 |
+
"eval_samples_per_second": 6574.48,
|
488 |
+
"eval_steps_per_second": 12.951,
|
489 |
+
"step": 60
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 5.083333333333333,
|
493 |
+
"grad_norm": 12.710640907287598,
|
494 |
+
"learning_rate": 1.2608515915667631e-05,
|
495 |
+
"loss": 0.9839,
|
496 |
+
"step": 61
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 5.166666666666667,
|
500 |
+
"grad_norm": 12.185357093811035,
|
501 |
+
"learning_rate": 1.281521289789169e-05,
|
502 |
+
"loss": 1.0032,
|
503 |
+
"step": 62
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 5.25,
|
507 |
+
"grad_norm": 11.39303970336914,
|
508 |
+
"learning_rate": 1.3021909880115751e-05,
|
509 |
+
"loss": 0.962,
|
510 |
+
"step": 63
|
511 |
+
},
|
512 |
+
{
|
513 |
+
"epoch": 5.333333333333333,
|
514 |
+
"grad_norm": 1.437285304069519,
|
515 |
+
"learning_rate": 1.322860686233981e-05,
|
516 |
+
"loss": 0.9518,
|
517 |
+
"step": 64
|
518 |
+
},
|
519 |
+
{
|
520 |
+
"epoch": 5.416666666666667,
|
521 |
+
"grad_norm": 4.353344440460205,
|
522 |
+
"learning_rate": 1.343530384456387e-05,
|
523 |
+
"loss": 0.9594,
|
524 |
+
"step": 65
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 5.5,
|
528 |
+
"grad_norm": 4.012881755828857,
|
529 |
+
"learning_rate": 1.3642000826787929e-05,
|
530 |
+
"loss": 0.9457,
|
531 |
+
"step": 66
|
532 |
+
},
|
533 |
+
{
|
534 |
+
"epoch": 5.583333333333333,
|
535 |
+
"grad_norm": 2.2280287742614746,
|
536 |
+
"learning_rate": 1.384869780901199e-05,
|
537 |
+
"loss": 0.9613,
|
538 |
+
"step": 67
|
539 |
+
},
|
540 |
+
{
|
541 |
+
"epoch": 5.666666666666667,
|
542 |
+
"grad_norm": 1.5106911659240723,
|
543 |
+
"learning_rate": 1.4055394791236049e-05,
|
544 |
+
"loss": 0.9598,
|
545 |
+
"step": 68
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 5.75,
|
549 |
+
"grad_norm": 2.5880537033081055,
|
550 |
+
"learning_rate": 1.4262091773460108e-05,
|
551 |
+
"loss": 0.9843,
|
552 |
+
"step": 69
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 5.833333333333333,
|
556 |
+
"grad_norm": 1.996254563331604,
|
557 |
+
"learning_rate": 1.4468788755684167e-05,
|
558 |
+
"loss": 0.8948,
|
559 |
+
"step": 70
|
560 |
+
},
|
561 |
+
{
|
562 |
+
"epoch": 5.916666666666667,
|
563 |
+
"grad_norm": 4.193304061889648,
|
564 |
+
"learning_rate": 1.4675485737908228e-05,
|
565 |
+
"loss": 0.9226,
|
566 |
+
"step": 71
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 6.0,
|
570 |
+
"grad_norm": 3.2173025608062744,
|
571 |
+
"learning_rate": 1.4882182720132287e-05,
|
572 |
+
"loss": 0.9952,
|
573 |
+
"step": 72
|
574 |
+
},
|
575 |
+
{
|
576 |
+
"epoch": 6.0,
|
577 |
+
"eval_accuracy": 0.7809813753581661,
|
578 |
+
"eval_f1": 0.803218020917136,
|
579 |
+
"eval_loss": 0.4652039408683777,
|
580 |
+
"eval_precision": 0.7326093337246845,
|
581 |
+
"eval_recall": 0.8888888888888888,
|
582 |
+
"eval_runtime": 0.8886,
|
583 |
+
"eval_samples_per_second": 6284.084,
|
584 |
+
"eval_steps_per_second": 12.379,
|
585 |
+
"step": 72
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 6.083333333333333,
|
589 |
+
"grad_norm": 5.742236137390137,
|
590 |
+
"learning_rate": 1.5088879702356347e-05,
|
591 |
+
"loss": 0.9661,
|
592 |
+
"step": 73
|
593 |
+
},
|
594 |
+
{
|
595 |
+
"epoch": 6.166666666666667,
|
596 |
+
"grad_norm": 6.754814147949219,
|
597 |
+
"learning_rate": 1.5295576684580404e-05,
|
598 |
+
"loss": 0.9349,
|
599 |
+
"step": 74
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 6.25,
|
603 |
+
"grad_norm": 7.546127796173096,
|
604 |
+
"learning_rate": 1.5502273666804463e-05,
|
605 |
+
"loss": 0.9135,
|
606 |
+
"step": 75
|
607 |
+
},
|
608 |
+
{
|
609 |
+
"epoch": 6.333333333333333,
|
610 |
+
"grad_norm": 3.626211404800415,
|
611 |
+
"learning_rate": 1.5708970649028523e-05,
|
612 |
+
"loss": 0.9736,
|
613 |
+
"step": 76
|
614 |
+
},
|
615 |
+
{
|
616 |
+
"epoch": 6.416666666666667,
|
617 |
+
"grad_norm": 1.5129320621490479,
|
618 |
+
"learning_rate": 1.5915667631252582e-05,
|
619 |
+
"loss": 0.9112,
|
620 |
+
"step": 77
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 6.5,
|
624 |
+
"grad_norm": 5.451498031616211,
|
625 |
+
"learning_rate": 1.6122364613476644e-05,
|
626 |
+
"loss": 0.9227,
|
627 |
+
"step": 78
|
628 |
+
},
|
629 |
+
{
|
630 |
+
"epoch": 6.583333333333333,
|
631 |
+
"grad_norm": 1.7330838441848755,
|
632 |
+
"learning_rate": 1.6329061595700704e-05,
|
633 |
+
"loss": 0.934,
|
634 |
+
"step": 79
|
635 |
+
},
|
636 |
+
{
|
637 |
+
"epoch": 6.666666666666667,
|
638 |
+
"grad_norm": 7.836896896362305,
|
639 |
+
"learning_rate": 1.6535758577924763e-05,
|
640 |
+
"loss": 1.007,
|
641 |
+
"step": 80
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 6.75,
|
645 |
+
"grad_norm": 3.248494863510132,
|
646 |
+
"learning_rate": 1.6742455560148822e-05,
|
647 |
+
"loss": 0.9134,
|
648 |
+
"step": 81
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 6.833333333333333,
|
652 |
+
"grad_norm": 3.5367369651794434,
|
653 |
+
"learning_rate": 1.694915254237288e-05,
|
654 |
+
"loss": 0.9353,
|
655 |
+
"step": 82
|
656 |
+
},
|
657 |
+
{
|
658 |
+
"epoch": 6.916666666666667,
|
659 |
+
"grad_norm": 7.193138122558594,
|
660 |
+
"learning_rate": 1.715584952459694e-05,
|
661 |
+
"loss": 0.9518,
|
662 |
+
"step": 83
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 7.0,
|
666 |
+
"grad_norm": 1.2551543712615967,
|
667 |
+
"learning_rate": 1.7362546506821e-05,
|
668 |
+
"loss": 1.0059,
|
669 |
+
"step": 84
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 7.0,
|
673 |
+
"eval_accuracy": 0.7657593123209169,
|
674 |
+
"eval_f1": 0.7808310991957105,
|
675 |
+
"eval_loss": 0.46547508239746094,
|
676 |
+
"eval_precision": 0.7373417721518988,
|
677 |
+
"eval_recall": 0.8297720797720798,
|
678 |
+
"eval_runtime": 0.9148,
|
679 |
+
"eval_samples_per_second": 6104.108,
|
680 |
+
"eval_steps_per_second": 12.025,
|
681 |
+
"step": 84
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 7.083333333333333,
|
685 |
+
"grad_norm": 7.826042652130127,
|
686 |
+
"learning_rate": 1.756924348904506e-05,
|
687 |
+
"loss": 0.9174,
|
688 |
+
"step": 85
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 7.166666666666667,
|
692 |
+
"grad_norm": 6.515051364898682,
|
693 |
+
"learning_rate": 1.777594047126912e-05,
|
694 |
+
"loss": 0.9969,
|
695 |
+
"step": 86
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 7.25,
|
699 |
+
"grad_norm": 10.743474960327148,
|
700 |
+
"learning_rate": 1.798263745349318e-05,
|
701 |
+
"loss": 0.9585,
|
702 |
+
"step": 87
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 7.333333333333333,
|
706 |
+
"grad_norm": 1.4464409351348877,
|
707 |
+
"learning_rate": 1.818933443571724e-05,
|
708 |
+
"loss": 0.9702,
|
709 |
+
"step": 88
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 7.416666666666667,
|
713 |
+
"grad_norm": 9.397377014160156,
|
714 |
+
"learning_rate": 1.83960314179413e-05,
|
715 |
+
"loss": 0.9586,
|
716 |
+
"step": 89
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 7.5,
|
720 |
+
"grad_norm": 2.005628824234009,
|
721 |
+
"learning_rate": 1.8602728400165358e-05,
|
722 |
+
"loss": 0.9062,
|
723 |
+
"step": 90
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 7.583333333333333,
|
727 |
+
"grad_norm": 0.8444802761077881,
|
728 |
+
"learning_rate": 1.8809425382389417e-05,
|
729 |
+
"loss": 0.8903,
|
730 |
+
"step": 91
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 7.666666666666667,
|
734 |
+
"grad_norm": 3.0471880435943604,
|
735 |
+
"learning_rate": 1.9016122364613476e-05,
|
736 |
+
"loss": 0.9279,
|
737 |
+
"step": 92
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 7.75,
|
741 |
+
"grad_norm": 10.25209903717041,
|
742 |
+
"learning_rate": 1.9222819346837536e-05,
|
743 |
+
"loss": 0.9189,
|
744 |
+
"step": 93
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 7.833333333333333,
|
748 |
+
"grad_norm": 3.7932651042938232,
|
749 |
+
"learning_rate": 1.9429516329061595e-05,
|
750 |
+
"loss": 0.9612,
|
751 |
+
"step": 94
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 7.916666666666667,
|
755 |
+
"grad_norm": 8.85326862335205,
|
756 |
+
"learning_rate": 1.9636213311285657e-05,
|
757 |
+
"loss": 0.8939,
|
758 |
+
"step": 95
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 8.0,
|
762 |
+
"grad_norm": 4.815281867980957,
|
763 |
+
"learning_rate": 1.9842910293509713e-05,
|
764 |
+
"loss": 0.9353,
|
765 |
+
"step": 96
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 8.0,
|
769 |
+
"eval_accuracy": 0.7888610315186246,
|
770 |
+
"eval_f1": 0.8158675620802749,
|
771 |
+
"eval_loss": 0.45332759618759155,
|
772 |
+
"eval_precision": 0.7265646731571628,
|
773 |
+
"eval_recall": 0.9301994301994302,
|
774 |
+
"eval_runtime": 0.8586,
|
775 |
+
"eval_samples_per_second": 6503.642,
|
776 |
+
"eval_steps_per_second": 12.812,
|
777 |
+
"step": 96
|
778 |
+
},
|
779 |
+
{
|
780 |
+
"epoch": 8.083333333333334,
|
781 |
+
"grad_norm": 2.567741632461548,
|
782 |
+
"learning_rate": 2.0049607275733776e-05,
|
783 |
+
"loss": 0.9172,
|
784 |
+
"step": 97
|
785 |
+
},
|
786 |
+
{
|
787 |
+
"epoch": 8.166666666666666,
|
788 |
+
"grad_norm": 6.861955642700195,
|
789 |
+
"learning_rate": 2.025630425795783e-05,
|
790 |
+
"loss": 0.998,
|
791 |
+
"step": 98
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 8.25,
|
795 |
+
"grad_norm": 1.5927001237869263,
|
796 |
+
"learning_rate": 2.0463001240181894e-05,
|
797 |
+
"loss": 0.9389,
|
798 |
+
"step": 99
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 8.333333333333334,
|
802 |
+
"grad_norm": 7.267998695373535,
|
803 |
+
"learning_rate": 2.0669698222405953e-05,
|
804 |
+
"loss": 0.8969,
|
805 |
+
"step": 100
|
806 |
+
},
|
807 |
+
{
|
808 |
+
"epoch": 8.416666666666666,
|
809 |
+
"grad_norm": 6.102052211761475,
|
810 |
+
"learning_rate": 2.0876395204630013e-05,
|
811 |
+
"loss": 0.8746,
|
812 |
+
"step": 101
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 8.5,
|
816 |
+
"grad_norm": 7.087440490722656,
|
817 |
+
"learning_rate": 2.1083092186854072e-05,
|
818 |
+
"loss": 0.9124,
|
819 |
+
"step": 102
|
820 |
+
},
|
821 |
+
{
|
822 |
+
"epoch": 8.583333333333334,
|
823 |
+
"grad_norm": 4.715837001800537,
|
824 |
+
"learning_rate": 2.1289789169078134e-05,
|
825 |
+
"loss": 0.9194,
|
826 |
+
"step": 103
|
827 |
+
},
|
828 |
+
{
|
829 |
+
"epoch": 8.666666666666666,
|
830 |
+
"grad_norm": 8.213499069213867,
|
831 |
+
"learning_rate": 2.149648615130219e-05,
|
832 |
+
"loss": 0.8937,
|
833 |
+
"step": 104
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 8.75,
|
837 |
+
"grad_norm": 7.216043949127197,
|
838 |
+
"learning_rate": 2.1703183133526253e-05,
|
839 |
+
"loss": 0.9473,
|
840 |
+
"step": 105
|
841 |
+
},
|
842 |
+
{
|
843 |
+
"epoch": 8.833333333333334,
|
844 |
+
"grad_norm": 8.029379844665527,
|
845 |
+
"learning_rate": 2.190988011575031e-05,
|
846 |
+
"loss": 0.9368,
|
847 |
+
"step": 106
|
848 |
+
},
|
849 |
+
{
|
850 |
+
"epoch": 8.916666666666666,
|
851 |
+
"grad_norm": 26.224346160888672,
|
852 |
+
"learning_rate": 2.211657709797437e-05,
|
853 |
+
"loss": 0.9433,
|
854 |
+
"step": 107
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 9.0,
|
858 |
+
"grad_norm": 4.666304111480713,
|
859 |
+
"learning_rate": 2.232327408019843e-05,
|
860 |
+
"loss": 0.958,
|
861 |
+
"step": 108
|
862 |
+
},
|
863 |
+
{
|
864 |
+
"epoch": 9.0,
|
865 |
+
"eval_accuracy": 0.7901146131805158,
|
866 |
+
"eval_f1": 0.8189681804139636,
|
867 |
+
"eval_loss": 0.44851964712142944,
|
868 |
+
"eval_precision": 0.7231314784506274,
|
869 |
+
"eval_recall": 0.9440883190883191,
|
870 |
+
"eval_runtime": 0.8565,
|
871 |
+
"eval_samples_per_second": 6519.192,
|
872 |
+
"eval_steps_per_second": 12.842,
|
873 |
+
"step": 108
|
874 |
+
},
|
875 |
+
{
|
876 |
+
"epoch": 9.083333333333334,
|
877 |
+
"grad_norm": 5.73352575302124,
|
878 |
+
"learning_rate": 2.252997106242249e-05,
|
879 |
+
"loss": 0.9322,
|
880 |
+
"step": 109
|
881 |
+
},
|
882 |
+
{
|
883 |
+
"epoch": 9.166666666666666,
|
884 |
+
"grad_norm": 5.408687591552734,
|
885 |
+
"learning_rate": 2.273666804464655e-05,
|
886 |
+
"loss": 0.918,
|
887 |
+
"step": 110
|
888 |
+
},
|
889 |
+
{
|
890 |
+
"epoch": 9.25,
|
891 |
+
"grad_norm": 4.718300819396973,
|
892 |
+
"learning_rate": 2.294336502687061e-05,
|
893 |
+
"loss": 0.9055,
|
894 |
+
"step": 111
|
895 |
+
},
|
896 |
+
{
|
897 |
+
"epoch": 9.333333333333334,
|
898 |
+
"grad_norm": 4.96096658706665,
|
899 |
+
"learning_rate": 2.3150062009094667e-05,
|
900 |
+
"loss": 0.8929,
|
901 |
+
"step": 112
|
902 |
+
},
|
903 |
+
{
|
904 |
+
"epoch": 9.416666666666666,
|
905 |
+
"grad_norm": 5.030294418334961,
|
906 |
+
"learning_rate": 2.335675899131873e-05,
|
907 |
+
"loss": 0.9204,
|
908 |
+
"step": 113
|
909 |
+
},
|
910 |
+
{
|
911 |
+
"epoch": 9.5,
|
912 |
+
"grad_norm": 10.890329360961914,
|
913 |
+
"learning_rate": 2.3563455973542786e-05,
|
914 |
+
"loss": 0.8786,
|
915 |
+
"step": 114
|
916 |
+
},
|
917 |
+
{
|
918 |
+
"epoch": 9.583333333333334,
|
919 |
+
"grad_norm": 4.339248180389404,
|
920 |
+
"learning_rate": 2.3770152955766848e-05,
|
921 |
+
"loss": 0.8955,
|
922 |
+
"step": 115
|
923 |
+
},
|
924 |
+
{
|
925 |
+
"epoch": 9.666666666666666,
|
926 |
+
"grad_norm": 5.493417263031006,
|
927 |
+
"learning_rate": 2.3976849937990907e-05,
|
928 |
+
"loss": 0.9566,
|
929 |
+
"step": 116
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 9.75,
|
933 |
+
"grad_norm": 14.134764671325684,
|
934 |
+
"learning_rate": 2.4183546920214967e-05,
|
935 |
+
"loss": 0.8951,
|
936 |
+
"step": 117
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 9.833333333333334,
|
940 |
+
"grad_norm": 9.192184448242188,
|
941 |
+
"learning_rate": 2.4390243902439026e-05,
|
942 |
+
"loss": 0.9366,
|
943 |
+
"step": 118
|
944 |
+
},
|
945 |
+
{
|
946 |
+
"epoch": 9.916666666666666,
|
947 |
+
"grad_norm": 5.674501419067383,
|
948 |
+
"learning_rate": 2.459694088466308e-05,
|
949 |
+
"loss": 0.9481,
|
950 |
+
"step": 119
|
951 |
+
},
|
952 |
+
{
|
953 |
+
"epoch": 10.0,
|
954 |
+
"grad_norm": 7.524171352386475,
|
955 |
+
"learning_rate": 2.4803637866887144e-05,
|
956 |
+
"loss": 0.8831,
|
957 |
+
"step": 120
|
958 |
+
},
|
959 |
+
{
|
960 |
+
"epoch": 10.0,
|
961 |
+
"eval_accuracy": 0.7877865329512894,
|
962 |
+
"eval_f1": 0.8118749007779013,
|
963 |
+
"eval_loss": 0.441786527633667,
|
964 |
+
"eval_precision": 0.7324548839873961,
|
965 |
+
"eval_recall": 0.9106125356125356,
|
966 |
+
"eval_runtime": 0.8845,
|
967 |
+
"eval_samples_per_second": 6313.336,
|
968 |
+
"eval_steps_per_second": 12.437,
|
969 |
+
"step": 120
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 10.083333333333334,
|
973 |
+
"grad_norm": 4.1865315437316895,
|
974 |
+
"learning_rate": 2.50103348491112e-05,
|
975 |
+
"loss": 0.8577,
|
976 |
+
"step": 121
|
977 |
+
},
|
978 |
+
{
|
979 |
+
"epoch": 10.166666666666666,
|
980 |
+
"grad_norm": 3.912219524383545,
|
981 |
+
"learning_rate": 2.5217031831335263e-05,
|
982 |
+
"loss": 0.8968,
|
983 |
+
"step": 122
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 10.25,
|
987 |
+
"grad_norm": 2.462045431137085,
|
988 |
+
"learning_rate": 2.5423728813559322e-05,
|
989 |
+
"loss": 0.873,
|
990 |
+
"step": 123
|
991 |
+
},
|
992 |
+
{
|
993 |
+
"epoch": 10.333333333333334,
|
994 |
+
"grad_norm": 9.20329761505127,
|
995 |
+
"learning_rate": 2.563042579578338e-05,
|
996 |
+
"loss": 0.9218,
|
997 |
+
"step": 124
|
998 |
+
},
|
999 |
+
{
|
1000 |
+
"epoch": 10.416666666666666,
|
1001 |
+
"grad_norm": 10.658252716064453,
|
1002 |
+
"learning_rate": 2.583712277800744e-05,
|
1003 |
+
"loss": 0.8754,
|
1004 |
+
"step": 125
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 10.5,
|
1008 |
+
"grad_norm": 12.687186241149902,
|
1009 |
+
"learning_rate": 2.6043819760231503e-05,
|
1010 |
+
"loss": 0.8939,
|
1011 |
+
"step": 126
|
1012 |
+
},
|
1013 |
+
{
|
1014 |
+
"epoch": 10.583333333333334,
|
1015 |
+
"grad_norm": 15.834625244140625,
|
1016 |
+
"learning_rate": 2.625051674245556e-05,
|
1017 |
+
"loss": 0.9424,
|
1018 |
+
"step": 127
|
1019 |
+
},
|
1020 |
+
{
|
1021 |
+
"epoch": 10.666666666666666,
|
1022 |
+
"grad_norm": 2.268646717071533,
|
1023 |
+
"learning_rate": 2.645721372467962e-05,
|
1024 |
+
"loss": 0.9092,
|
1025 |
+
"step": 128
|
1026 |
+
},
|
1027 |
+
{
|
1028 |
+
"epoch": 10.75,
|
1029 |
+
"grad_norm": 11.147329330444336,
|
1030 |
+
"learning_rate": 2.6663910706903677e-05,
|
1031 |
+
"loss": 0.8842,
|
1032 |
+
"step": 129
|
1033 |
+
},
|
1034 |
+
{
|
1035 |
+
"epoch": 10.833333333333334,
|
1036 |
+
"grad_norm": 5.504317283630371,
|
1037 |
+
"learning_rate": 2.687060768912774e-05,
|
1038 |
+
"loss": 0.8608,
|
1039 |
+
"step": 130
|
1040 |
+
},
|
1041 |
+
{
|
1042 |
+
"epoch": 10.916666666666666,
|
1043 |
+
"grad_norm": 15.80197811126709,
|
1044 |
+
"learning_rate": 2.70773046713518e-05,
|
1045 |
+
"loss": 0.9409,
|
1046 |
+
"step": 131
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 11.0,
|
1050 |
+
"grad_norm": 8.165436744689941,
|
1051 |
+
"learning_rate": 2.7284001653575858e-05,
|
1052 |
+
"loss": 0.8816,
|
1053 |
+
"step": 132
|
1054 |
+
},
|
1055 |
+
{
|
1056 |
+
"epoch": 11.0,
|
1057 |
+
"eval_accuracy": 0.7978151862464183,
|
1058 |
+
"eval_f1": 0.8197924980047885,
|
1059 |
+
"eval_loss": 0.4275962710380554,
|
1060 |
+
"eval_precision": 0.7428406132484814,
|
1061 |
+
"eval_recall": 0.9145299145299145,
|
1062 |
+
"eval_runtime": 1.1147,
|
1063 |
+
"eval_samples_per_second": 5009.26,
|
1064 |
+
"eval_steps_per_second": 9.868,
|
1065 |
+
"step": 132
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 11.083333333333334,
|
1069 |
+
"grad_norm": 2.223496913909912,
|
1070 |
+
"learning_rate": 2.7490698635799917e-05,
|
1071 |
+
"loss": 0.8692,
|
1072 |
+
"step": 133
|
1073 |
+
},
|
1074 |
+
{
|
1075 |
+
"epoch": 11.166666666666666,
|
1076 |
+
"grad_norm": 17.665945053100586,
|
1077 |
+
"learning_rate": 2.769739561802398e-05,
|
1078 |
+
"loss": 0.9118,
|
1079 |
+
"step": 134
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 11.25,
|
1083 |
+
"grad_norm": 33.71101379394531,
|
1084 |
+
"learning_rate": 2.7904092600248035e-05,
|
1085 |
+
"loss": 0.8861,
|
1086 |
+
"step": 135
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 11.333333333333334,
|
1090 |
+
"grad_norm": 23.88389015197754,
|
1091 |
+
"learning_rate": 2.8110789582472098e-05,
|
1092 |
+
"loss": 0.8556,
|
1093 |
+
"step": 136
|
1094 |
+
},
|
1095 |
+
{
|
1096 |
+
"epoch": 11.416666666666666,
|
1097 |
+
"grad_norm": 1.9535236358642578,
|
1098 |
+
"learning_rate": 2.8317486564696154e-05,
|
1099 |
+
"loss": 0.8316,
|
1100 |
+
"step": 137
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 11.5,
|
1104 |
+
"grad_norm": 14.455438613891602,
|
1105 |
+
"learning_rate": 2.8524183546920216e-05,
|
1106 |
+
"loss": 0.9293,
|
1107 |
+
"step": 138
|
1108 |
+
},
|
1109 |
+
{
|
1110 |
+
"epoch": 11.583333333333334,
|
1111 |
+
"grad_norm": 9.921207427978516,
|
1112 |
+
"learning_rate": 2.8730880529144276e-05,
|
1113 |
+
"loss": 0.8619,
|
1114 |
+
"step": 139
|
1115 |
+
},
|
1116 |
+
{
|
1117 |
+
"epoch": 11.666666666666666,
|
1118 |
+
"grad_norm": 12.480018615722656,
|
1119 |
+
"learning_rate": 2.8937577511368335e-05,
|
1120 |
+
"loss": 0.9273,
|
1121 |
+
"step": 140
|
1122 |
+
},
|
1123 |
+
{
|
1124 |
+
"epoch": 11.75,
|
1125 |
+
"grad_norm": 14.018143653869629,
|
1126 |
+
"learning_rate": 2.9144274493592394e-05,
|
1127 |
+
"loss": 0.809,
|
1128 |
+
"step": 141
|
1129 |
+
},
|
1130 |
+
{
|
1131 |
+
"epoch": 11.833333333333334,
|
1132 |
+
"grad_norm": 6.436577320098877,
|
1133 |
+
"learning_rate": 2.9350971475816457e-05,
|
1134 |
+
"loss": 0.8092,
|
1135 |
+
"step": 142
|
1136 |
+
},
|
1137 |
+
{
|
1138 |
+
"epoch": 11.916666666666666,
|
1139 |
+
"grad_norm": 21.0689640045166,
|
1140 |
+
"learning_rate": 2.9557668458040512e-05,
|
1141 |
+
"loss": 0.8925,
|
1142 |
+
"step": 143
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 12.0,
|
1146 |
+
"grad_norm": 21.12411117553711,
|
1147 |
+
"learning_rate": 2.9764365440264575e-05,
|
1148 |
+
"loss": 0.9238,
|
1149 |
+
"step": 144
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 12.0,
|
1153 |
+
"eval_accuracy": 0.8042621776504298,
|
1154 |
+
"eval_f1": 0.8249799839871897,
|
1155 |
+
"eval_loss": 0.4249096214771271,
|
1156 |
+
"eval_precision": 0.7494908350305499,
|
1157 |
+
"eval_recall": 0.9173789173789174,
|
1158 |
+
"eval_runtime": 0.9206,
|
1159 |
+
"eval_samples_per_second": 6065.673,
|
1160 |
+
"eval_steps_per_second": 11.949,
|
1161 |
+
"step": 144
|
1162 |
+
},
|
1163 |
+
{
|
1164 |
+
"epoch": 12.083333333333334,
|
1165 |
+
"grad_norm": 16.12071418762207,
|
1166 |
+
"learning_rate": 2.997106242248863e-05,
|
1167 |
+
"loss": 0.8618,
|
1168 |
+
"step": 145
|
1169 |
+
},
|
1170 |
+
{
|
1171 |
+
"epoch": 12.166666666666666,
|
1172 |
+
"grad_norm": 9.940985679626465,
|
1173 |
+
"learning_rate": 3.0177759404712693e-05,
|
1174 |
+
"loss": 0.8383,
|
1175 |
+
"step": 146
|
1176 |
+
},
|
1177 |
+
{
|
1178 |
+
"epoch": 12.25,
|
1179 |
+
"grad_norm": 7.171441078186035,
|
1180 |
+
"learning_rate": 3.0384456386936753e-05,
|
1181 |
+
"loss": 0.8693,
|
1182 |
+
"step": 147
|
1183 |
+
},
|
1184 |
+
{
|
1185 |
+
"epoch": 12.333333333333334,
|
1186 |
+
"grad_norm": 17.261926651000977,
|
1187 |
+
"learning_rate": 3.059115336916081e-05,
|
1188 |
+
"loss": 0.8513,
|
1189 |
+
"step": 148
|
1190 |
+
},
|
1191 |
+
{
|
1192 |
+
"epoch": 12.416666666666666,
|
1193 |
+
"grad_norm": 16.04341697692871,
|
1194 |
+
"learning_rate": 3.079785035138487e-05,
|
1195 |
+
"loss": 0.8327,
|
1196 |
+
"step": 149
|
1197 |
+
},
|
1198 |
+
{
|
1199 |
+
"epoch": 12.5,
|
1200 |
+
"grad_norm": 6.637321472167969,
|
1201 |
+
"learning_rate": 3.100454733360893e-05,
|
1202 |
+
"loss": 0.828,
|
1203 |
+
"step": 150
|
1204 |
+
},
|
1205 |
+
{
|
1206 |
+
"epoch": 12.583333333333334,
|
1207 |
+
"grad_norm": 13.405370712280273,
|
1208 |
+
"learning_rate": 3.121124431583299e-05,
|
1209 |
+
"loss": 0.8454,
|
1210 |
+
"step": 151
|
1211 |
+
},
|
1212 |
+
{
|
1213 |
+
"epoch": 12.666666666666666,
|
1214 |
+
"grad_norm": 14.123351097106934,
|
1215 |
+
"learning_rate": 3.1417941298057045e-05,
|
1216 |
+
"loss": 0.8108,
|
1217 |
+
"step": 152
|
1218 |
+
},
|
1219 |
+
{
|
1220 |
+
"epoch": 12.75,
|
1221 |
+
"grad_norm": 3.29341459274292,
|
1222 |
+
"learning_rate": 3.162463828028111e-05,
|
1223 |
+
"loss": 0.8067,
|
1224 |
+
"step": 153
|
1225 |
+
},
|
1226 |
+
{
|
1227 |
+
"epoch": 12.833333333333334,
|
1228 |
+
"grad_norm": 10.833609580993652,
|
1229 |
+
"learning_rate": 3.1831335262505164e-05,
|
1230 |
+
"loss": 0.8219,
|
1231 |
+
"step": 154
|
1232 |
+
},
|
1233 |
+
{
|
1234 |
+
"epoch": 12.916666666666666,
|
1235 |
+
"grad_norm": 11.29476261138916,
|
1236 |
+
"learning_rate": 3.2038032244729226e-05,
|
1237 |
+
"loss": 0.8237,
|
1238 |
+
"step": 155
|
1239 |
+
},
|
1240 |
+
{
|
1241 |
+
"epoch": 13.0,
|
1242 |
+
"grad_norm": 14.614105224609375,
|
1243 |
+
"learning_rate": 3.224472922695329e-05,
|
1244 |
+
"loss": 0.8667,
|
1245 |
+
"step": 156
|
1246 |
+
},
|
1247 |
+
{
|
1248 |
+
"epoch": 13.0,
|
1249 |
+
"eval_accuracy": 0.8202005730659025,
|
1250 |
+
"eval_f1": 0.8284347231715653,
|
1251 |
+
"eval_loss": 0.4048195779323578,
|
1252 |
+
"eval_precision": 0.7963206307490145,
|
1253 |
+
"eval_recall": 0.8632478632478633,
|
1254 |
+
"eval_runtime": 0.9164,
|
1255 |
+
"eval_samples_per_second": 6093.191,
|
1256 |
+
"eval_steps_per_second": 12.003,
|
1257 |
+
"step": 156
|
1258 |
+
},
|
1259 |
+
{
|
1260 |
+
"epoch": 13.083333333333334,
|
1261 |
+
"grad_norm": 14.725034713745117,
|
1262 |
+
"learning_rate": 3.2451426209177345e-05,
|
1263 |
+
"loss": 0.8001,
|
1264 |
+
"step": 157
|
1265 |
+
},
|
1266 |
+
{
|
1267 |
+
"epoch": 13.166666666666666,
|
1268 |
+
"grad_norm": 7.957756042480469,
|
1269 |
+
"learning_rate": 3.265812319140141e-05,
|
1270 |
+
"loss": 0.7783,
|
1271 |
+
"step": 158
|
1272 |
+
},
|
1273 |
+
{
|
1274 |
+
"epoch": 13.25,
|
1275 |
+
"grad_norm": 10.417937278747559,
|
1276 |
+
"learning_rate": 3.286482017362547e-05,
|
1277 |
+
"loss": 0.8804,
|
1278 |
+
"step": 159
|
1279 |
+
},
|
1280 |
+
{
|
1281 |
+
"epoch": 13.333333333333334,
|
1282 |
+
"grad_norm": 19.055553436279297,
|
1283 |
+
"learning_rate": 3.3071517155849525e-05,
|
1284 |
+
"loss": 0.8248,
|
1285 |
+
"step": 160
|
1286 |
+
},
|
1287 |
+
{
|
1288 |
+
"epoch": 13.416666666666666,
|
1289 |
+
"grad_norm": 8.582987785339355,
|
1290 |
+
"learning_rate": 3.327821413807359e-05,
|
1291 |
+
"loss": 0.8331,
|
1292 |
+
"step": 161
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 13.5,
|
1296 |
+
"grad_norm": 9.180889129638672,
|
1297 |
+
"learning_rate": 3.3484911120297644e-05,
|
1298 |
+
"loss": 0.8187,
|
1299 |
+
"step": 162
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 13.583333333333334,
|
1303 |
+
"grad_norm": 10.075421333312988,
|
1304 |
+
"learning_rate": 3.3691608102521706e-05,
|
1305 |
+
"loss": 0.7689,
|
1306 |
+
"step": 163
|
1307 |
+
},
|
1308 |
+
{
|
1309 |
+
"epoch": 13.666666666666666,
|
1310 |
+
"grad_norm": 9.318410873413086,
|
1311 |
+
"learning_rate": 3.389830508474576e-05,
|
1312 |
+
"loss": 0.7934,
|
1313 |
+
"step": 164
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 13.75,
|
1317 |
+
"grad_norm": 2.9320974349975586,
|
1318 |
+
"learning_rate": 3.4105002066969825e-05,
|
1319 |
+
"loss": 0.8437,
|
1320 |
+
"step": 165
|
1321 |
+
},
|
1322 |
+
{
|
1323 |
+
"epoch": 13.833333333333334,
|
1324 |
+
"grad_norm": 14.561713218688965,
|
1325 |
+
"learning_rate": 3.431169904919388e-05,
|
1326 |
+
"loss": 0.8451,
|
1327 |
+
"step": 166
|
1328 |
+
},
|
1329 |
+
{
|
1330 |
+
"epoch": 13.916666666666666,
|
1331 |
+
"grad_norm": 21.375545501708984,
|
1332 |
+
"learning_rate": 3.451839603141794e-05,
|
1333 |
+
"loss": 0.8189,
|
1334 |
+
"step": 167
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 14.0,
|
1338 |
+
"grad_norm": 22.70510482788086,
|
1339 |
+
"learning_rate": 3.4725093013642e-05,
|
1340 |
+
"loss": 0.7486,
|
1341 |
+
"step": 168
|
1342 |
+
},
|
1343 |
+
{
|
1344 |
+
"epoch": 14.0,
|
1345 |
+
"eval_accuracy": 0.8305873925501432,
|
1346 |
+
"eval_f1": 0.8388964577656676,
|
1347 |
+
"eval_loss": 0.3896721303462982,
|
1348 |
+
"eval_precision": 0.8038511749347258,
|
1349 |
+
"eval_recall": 0.8771367521367521,
|
1350 |
+
"eval_runtime": 1.0222,
|
1351 |
+
"eval_samples_per_second": 5462.668,
|
1352 |
+
"eval_steps_per_second": 10.761,
|
1353 |
+
"step": 168
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 14.083333333333334,
|
1357 |
+
"grad_norm": 11.696023941040039,
|
1358 |
+
"learning_rate": 3.493178999586606e-05,
|
1359 |
+
"loss": 0.8159,
|
1360 |
+
"step": 169
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 14.166666666666666,
|
1364 |
+
"grad_norm": 8.56267261505127,
|
1365 |
+
"learning_rate": 3.513848697809012e-05,
|
1366 |
+
"loss": 0.8037,
|
1367 |
+
"step": 170
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 14.25,
|
1371 |
+
"grad_norm": 25.44778823852539,
|
1372 |
+
"learning_rate": 3.534518396031418e-05,
|
1373 |
+
"loss": 0.8222,
|
1374 |
+
"step": 171
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 14.333333333333334,
|
1378 |
+
"grad_norm": 17.102243423461914,
|
1379 |
+
"learning_rate": 3.555188094253824e-05,
|
1380 |
+
"loss": 0.8194,
|
1381 |
+
"step": 172
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 14.416666666666666,
|
1385 |
+
"grad_norm": 4.854645729064941,
|
1386 |
+
"learning_rate": 3.57585779247623e-05,
|
1387 |
+
"loss": 0.7582,
|
1388 |
+
"step": 173
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 14.5,
|
1392 |
+
"grad_norm": 16.25187873840332,
|
1393 |
+
"learning_rate": 3.596527490698636e-05,
|
1394 |
+
"loss": 0.8399,
|
1395 |
+
"step": 174
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 14.583333333333334,
|
1399 |
+
"grad_norm": 17.408138275146484,
|
1400 |
+
"learning_rate": 3.6171971889210424e-05,
|
1401 |
+
"loss": 0.8771,
|
1402 |
+
"step": 175
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 14.666666666666666,
|
1406 |
+
"grad_norm": 13.063253402709961,
|
1407 |
+
"learning_rate": 3.637866887143448e-05,
|
1408 |
+
"loss": 0.8353,
|
1409 |
+
"step": 176
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 14.75,
|
1413 |
+
"grad_norm": 9.531180381774902,
|
1414 |
+
"learning_rate": 3.658536585365854e-05,
|
1415 |
+
"loss": 0.7752,
|
1416 |
+
"step": 177
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 14.833333333333334,
|
1420 |
+
"grad_norm": 9.24107837677002,
|
1421 |
+
"learning_rate": 3.67920628358826e-05,
|
1422 |
+
"loss": 0.8098,
|
1423 |
+
"step": 178
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 14.916666666666666,
|
1427 |
+
"grad_norm": 23.467121124267578,
|
1428 |
+
"learning_rate": 3.6998759818106654e-05,
|
1429 |
+
"loss": 0.8073,
|
1430 |
+
"step": 179
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 15.0,
|
1434 |
+
"grad_norm": 16.87409019470215,
|
1435 |
+
"learning_rate": 3.7205456800330716e-05,
|
1436 |
+
"loss": 0.8507,
|
1437 |
+
"step": 180
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 15.0,
|
1441 |
+
"eval_accuracy": 0.8114255014326648,
|
1442 |
+
"eval_f1": 0.8217972584193602,
|
1443 |
+
"eval_loss": 0.4059450030326843,
|
1444 |
+
"eval_precision": 0.782973234440503,
|
1445 |
+
"eval_recall": 0.8646723646723646,
|
1446 |
+
"eval_runtime": 1.0507,
|
1447 |
+
"eval_samples_per_second": 5314.388,
|
1448 |
+
"eval_steps_per_second": 10.469,
|
1449 |
+
"step": 180
|
1450 |
+
},
|
1451 |
+
{
|
1452 |
+
"epoch": 15.083333333333334,
|
1453 |
+
"grad_norm": 16.232149124145508,
|
1454 |
+
"learning_rate": 3.741215378255477e-05,
|
1455 |
+
"loss": 0.8486,
|
1456 |
+
"step": 181
|
1457 |
+
},
|
1458 |
+
{
|
1459 |
+
"epoch": 15.166666666666666,
|
1460 |
+
"grad_norm": 12.089692115783691,
|
1461 |
+
"learning_rate": 3.7618850764778835e-05,
|
1462 |
+
"loss": 0.8296,
|
1463 |
+
"step": 182
|
1464 |
+
},
|
1465 |
+
{
|
1466 |
+
"epoch": 15.25,
|
1467 |
+
"grad_norm": 13.09214973449707,
|
1468 |
+
"learning_rate": 3.782554774700289e-05,
|
1469 |
+
"loss": 0.8211,
|
1470 |
+
"step": 183
|
1471 |
+
},
|
1472 |
+
{
|
1473 |
+
"epoch": 15.333333333333334,
|
1474 |
+
"grad_norm": 31.21879768371582,
|
1475 |
+
"learning_rate": 3.803224472922695e-05,
|
1476 |
+
"loss": 0.8287,
|
1477 |
+
"step": 184
|
1478 |
+
},
|
1479 |
+
{
|
1480 |
+
"epoch": 15.416666666666666,
|
1481 |
+
"grad_norm": 23.260948181152344,
|
1482 |
+
"learning_rate": 3.823894171145101e-05,
|
1483 |
+
"loss": 0.8406,
|
1484 |
+
"step": 185
|
1485 |
+
},
|
1486 |
+
{
|
1487 |
+
"epoch": 15.5,
|
1488 |
+
"grad_norm": 8.325723648071289,
|
1489 |
+
"learning_rate": 3.844563869367507e-05,
|
1490 |
+
"loss": 0.7944,
|
1491 |
+
"step": 186
|
1492 |
+
},
|
1493 |
+
{
|
1494 |
+
"epoch": 15.583333333333334,
|
1495 |
+
"grad_norm": 11.76101016998291,
|
1496 |
+
"learning_rate": 3.8652335675899134e-05,
|
1497 |
+
"loss": 0.8481,
|
1498 |
+
"step": 187
|
1499 |
+
},
|
1500 |
+
{
|
1501 |
+
"epoch": 15.666666666666666,
|
1502 |
+
"grad_norm": 16.549983978271484,
|
1503 |
+
"learning_rate": 3.885903265812319e-05,
|
1504 |
+
"loss": 0.82,
|
1505 |
+
"step": 188
|
1506 |
+
},
|
1507 |
+
{
|
1508 |
+
"epoch": 15.75,
|
1509 |
+
"grad_norm": 19.037729263305664,
|
1510 |
+
"learning_rate": 3.906572964034725e-05,
|
1511 |
+
"loss": 0.8206,
|
1512 |
+
"step": 189
|
1513 |
+
},
|
1514 |
+
{
|
1515 |
+
"epoch": 15.833333333333334,
|
1516 |
+
"grad_norm": 9.386899948120117,
|
1517 |
+
"learning_rate": 3.9272426622571315e-05,
|
1518 |
+
"loss": 0.7478,
|
1519 |
+
"step": 190
|
1520 |
+
},
|
1521 |
+
{
|
1522 |
+
"epoch": 15.916666666666666,
|
1523 |
+
"grad_norm": 9.824003219604492,
|
1524 |
+
"learning_rate": 3.947912360479537e-05,
|
1525 |
+
"loss": 0.747,
|
1526 |
+
"step": 191
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 16.0,
|
1530 |
+
"grad_norm": 11.112277030944824,
|
1531 |
+
"learning_rate": 3.9685820587019427e-05,
|
1532 |
+
"loss": 0.7267,
|
1533 |
+
"step": 192
|
1534 |
+
},
|
1535 |
+
{
|
1536 |
+
"epoch": 16.0,
|
1537 |
+
"eval_accuracy": 0.832378223495702,
|
1538 |
+
"eval_f1": 0.8414096916299559,
|
1539 |
+
"eval_loss": 0.38714471459388733,
|
1540 |
+
"eval_precision": 0.8025210084033614,
|
1541 |
+
"eval_recall": 0.8842592592592593,
|
1542 |
+
"eval_runtime": 0.9871,
|
1543 |
+
"eval_samples_per_second": 5657.228,
|
1544 |
+
"eval_steps_per_second": 11.144,
|
1545 |
+
"step": 192
|
1546 |
+
},
|
1547 |
+
{
|
1548 |
+
"epoch": 16.083333333333332,
|
1549 |
+
"grad_norm": 10.595818519592285,
|
1550 |
+
"learning_rate": 3.989251756924349e-05,
|
1551 |
+
"loss": 0.759,
|
1552 |
+
"step": 193
|
1553 |
+
},
|
1554 |
+
{
|
1555 |
+
"epoch": 16.166666666666668,
|
1556 |
+
"grad_norm": 11.839064598083496,
|
1557 |
+
"learning_rate": 4.009921455146755e-05,
|
1558 |
+
"loss": 0.825,
|
1559 |
+
"step": 194
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 16.25,
|
1563 |
+
"grad_norm": 3.4257161617279053,
|
1564 |
+
"learning_rate": 4.0305911533691614e-05,
|
1565 |
+
"loss": 0.7992,
|
1566 |
+
"step": 195
|
1567 |
+
},
|
1568 |
+
{
|
1569 |
+
"epoch": 16.333333333333332,
|
1570 |
+
"grad_norm": 18.654296875,
|
1571 |
+
"learning_rate": 4.051260851591566e-05,
|
1572 |
+
"loss": 0.8204,
|
1573 |
+
"step": 196
|
1574 |
+
},
|
1575 |
+
{
|
1576 |
+
"epoch": 16.416666666666668,
|
1577 |
+
"grad_norm": 15.579732894897461,
|
1578 |
+
"learning_rate": 4.0719305498139726e-05,
|
1579 |
+
"loss": 0.8364,
|
1580 |
+
"step": 197
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 16.5,
|
1584 |
+
"grad_norm": 9.09702205657959,
|
1585 |
+
"learning_rate": 4.092600248036379e-05,
|
1586 |
+
"loss": 0.7636,
|
1587 |
+
"step": 198
|
1588 |
+
},
|
1589 |
+
{
|
1590 |
+
"epoch": 16.583333333333332,
|
1591 |
+
"grad_norm": 7.024932384490967,
|
1592 |
+
"learning_rate": 4.113269946258785e-05,
|
1593 |
+
"loss": 0.743,
|
1594 |
+
"step": 199
|
1595 |
+
},
|
1596 |
+
{
|
1597 |
+
"epoch": 16.666666666666668,
|
1598 |
+
"grad_norm": 11.396801948547363,
|
1599 |
+
"learning_rate": 4.133939644481191e-05,
|
1600 |
+
"loss": 0.835,
|
1601 |
+
"step": 200
|
1602 |
+
},
|
1603 |
+
{
|
1604 |
+
"epoch": 16.75,
|
1605 |
+
"grad_norm": 7.783175945281982,
|
1606 |
+
"learning_rate": 4.154609342703596e-05,
|
1607 |
+
"loss": 0.7866,
|
1608 |
+
"step": 201
|
1609 |
+
},
|
1610 |
+
{
|
1611 |
+
"epoch": 16.833333333333332,
|
1612 |
+
"grad_norm": 9.490063667297363,
|
1613 |
+
"learning_rate": 4.1752790409260025e-05,
|
1614 |
+
"loss": 0.8196,
|
1615 |
+
"step": 202
|
1616 |
+
},
|
1617 |
+
{
|
1618 |
+
"epoch": 16.916666666666668,
|
1619 |
+
"grad_norm": 18.039331436157227,
|
1620 |
+
"learning_rate": 4.195948739148409e-05,
|
1621 |
+
"loss": 0.7984,
|
1622 |
+
"step": 203
|
1623 |
+
},
|
1624 |
+
{
|
1625 |
+
"epoch": 17.0,
|
1626 |
+
"grad_norm": 7.622076034545898,
|
1627 |
+
"learning_rate": 4.2166184373708144e-05,
|
1628 |
+
"loss": 0.7222,
|
1629 |
+
"step": 204
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 17.0,
|
1633 |
+
"eval_accuracy": 0.8334527220630372,
|
1634 |
+
"eval_f1": 0.8446374874707652,
|
1635 |
+
"eval_loss": 0.3847378194332123,
|
1636 |
+
"eval_precision": 0.7954688483322845,
|
1637 |
+
"eval_recall": 0.9002849002849003,
|
1638 |
+
"eval_runtime": 0.9936,
|
1639 |
+
"eval_samples_per_second": 5620.079,
|
1640 |
+
"eval_steps_per_second": 11.071,
|
1641 |
+
"step": 204
|
1642 |
+
},
|
1643 |
+
{
|
1644 |
+
"epoch": 17.083333333333332,
|
1645 |
+
"grad_norm": 13.529067993164062,
|
1646 |
+
"learning_rate": 4.2372881355932206e-05,
|
1647 |
+
"loss": 0.8177,
|
1648 |
+
"step": 205
|
1649 |
+
},
|
1650 |
+
{
|
1651 |
+
"epoch": 17.166666666666668,
|
1652 |
+
"grad_norm": 13.250212669372559,
|
1653 |
+
"learning_rate": 4.257957833815627e-05,
|
1654 |
+
"loss": 0.8353,
|
1655 |
+
"step": 206
|
1656 |
+
},
|
1657 |
+
{
|
1658 |
+
"epoch": 17.25,
|
1659 |
+
"grad_norm": 7.602872848510742,
|
1660 |
+
"learning_rate": 4.2786275320380325e-05,
|
1661 |
+
"loss": 0.793,
|
1662 |
+
"step": 207
|
1663 |
+
},
|
1664 |
+
{
|
1665 |
+
"epoch": 17.333333333333332,
|
1666 |
+
"grad_norm": 19.804988861083984,
|
1667 |
+
"learning_rate": 4.299297230260438e-05,
|
1668 |
+
"loss": 0.8069,
|
1669 |
+
"step": 208
|
1670 |
+
},
|
1671 |
+
{
|
1672 |
+
"epoch": 17.416666666666668,
|
1673 |
+
"grad_norm": 20.907503128051758,
|
1674 |
+
"learning_rate": 4.319966928482844e-05,
|
1675 |
+
"loss": 0.8482,
|
1676 |
+
"step": 209
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 17.5,
|
1680 |
+
"grad_norm": 9.696860313415527,
|
1681 |
+
"learning_rate": 4.3406366267052506e-05,
|
1682 |
+
"loss": 0.8227,
|
1683 |
+
"step": 210
|
1684 |
+
},
|
1685 |
+
{
|
1686 |
+
"epoch": 17.583333333333332,
|
1687 |
+
"grad_norm": 9.85171890258789,
|
1688 |
+
"learning_rate": 4.3613063249276555e-05,
|
1689 |
+
"loss": 0.7604,
|
1690 |
+
"step": 211
|
1691 |
+
},
|
1692 |
+
{
|
1693 |
+
"epoch": 17.666666666666668,
|
1694 |
+
"grad_norm": 10.699975967407227,
|
1695 |
+
"learning_rate": 4.381976023150062e-05,
|
1696 |
+
"loss": 0.7899,
|
1697 |
+
"step": 212
|
1698 |
+
},
|
1699 |
+
{
|
1700 |
+
"epoch": 17.75,
|
1701 |
+
"grad_norm": 3.391948938369751,
|
1702 |
+
"learning_rate": 4.402645721372468e-05,
|
1703 |
+
"loss": 0.779,
|
1704 |
+
"step": 213
|
1705 |
+
},
|
1706 |
+
{
|
1707 |
+
"epoch": 17.833333333333332,
|
1708 |
+
"grad_norm": 5.576776027679443,
|
1709 |
+
"learning_rate": 4.423315419594874e-05,
|
1710 |
+
"loss": 0.7847,
|
1711 |
+
"step": 214
|
1712 |
+
},
|
1713 |
+
{
|
1714 |
+
"epoch": 17.916666666666668,
|
1715 |
+
"grad_norm": 6.727151870727539,
|
1716 |
+
"learning_rate": 4.44398511781728e-05,
|
1717 |
+
"loss": 0.7116,
|
1718 |
+
"step": 215
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 18.0,
|
1722 |
+
"grad_norm": 4.030666351318359,
|
1723 |
+
"learning_rate": 4.464654816039686e-05,
|
1724 |
+
"loss": 0.7302,
|
1725 |
+
"step": 216
|
1726 |
+
},
|
1727 |
+
{
|
1728 |
+
"epoch": 18.0,
|
1729 |
+
"eval_accuracy": 0.8053366762177651,
|
1730 |
+
"eval_f1": 0.8097321897426921,
|
1731 |
+
"eval_loss": 0.40027686953544617,
|
1732 |
+
"eval_precision": 0.7962134251290878,
|
1733 |
+
"eval_recall": 0.8237179487179487,
|
1734 |
+
"eval_runtime": 1.0057,
|
1735 |
+
"eval_samples_per_second": 5552.404,
|
1736 |
+
"eval_steps_per_second": 10.938,
|
1737 |
+
"step": 216
|
1738 |
+
},
|
1739 |
+
{
|
1740 |
+
"epoch": 18.083333333333332,
|
1741 |
+
"grad_norm": 14.245407104492188,
|
1742 |
+
"learning_rate": 4.4853245142620917e-05,
|
1743 |
+
"loss": 0.7765,
|
1744 |
+
"step": 217
|
1745 |
+
},
|
1746 |
+
{
|
1747 |
+
"epoch": 18.166666666666668,
|
1748 |
+
"grad_norm": 19.02879524230957,
|
1749 |
+
"learning_rate": 4.505994212484498e-05,
|
1750 |
+
"loss": 0.8181,
|
1751 |
+
"step": 218
|
1752 |
+
},
|
1753 |
+
{
|
1754 |
+
"epoch": 18.25,
|
1755 |
+
"grad_norm": 5.276630878448486,
|
1756 |
+
"learning_rate": 4.5266639107069035e-05,
|
1757 |
+
"loss": 0.7047,
|
1758 |
+
"step": 219
|
1759 |
+
},
|
1760 |
+
{
|
1761 |
+
"epoch": 18.333333333333332,
|
1762 |
+
"grad_norm": 8.71511173248291,
|
1763 |
+
"learning_rate": 4.54733360892931e-05,
|
1764 |
+
"loss": 0.8127,
|
1765 |
+
"step": 220
|
1766 |
+
},
|
1767 |
+
{
|
1768 |
+
"epoch": 18.416666666666668,
|
1769 |
+
"grad_norm": 13.701212882995605,
|
1770 |
+
"learning_rate": 4.568003307151716e-05,
|
1771 |
+
"loss": 0.8178,
|
1772 |
+
"step": 221
|
1773 |
+
},
|
1774 |
+
{
|
1775 |
+
"epoch": 18.5,
|
1776 |
+
"grad_norm": 5.985068321228027,
|
1777 |
+
"learning_rate": 4.588673005374122e-05,
|
1778 |
+
"loss": 0.7874,
|
1779 |
+
"step": 222
|
1780 |
+
},
|
1781 |
+
{
|
1782 |
+
"epoch": 18.583333333333332,
|
1783 |
+
"grad_norm": 6.030764579772949,
|
1784 |
+
"learning_rate": 4.609342703596527e-05,
|
1785 |
+
"loss": 0.7476,
|
1786 |
+
"step": 223
|
1787 |
+
},
|
1788 |
+
{
|
1789 |
+
"epoch": 18.666666666666668,
|
1790 |
+
"grad_norm": 8.99073314666748,
|
1791 |
+
"learning_rate": 4.6300124018189334e-05,
|
1792 |
+
"loss": 0.7563,
|
1793 |
+
"step": 224
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 18.75,
|
1797 |
+
"grad_norm": 11.01559829711914,
|
1798 |
+
"learning_rate": 4.65068210004134e-05,
|
1799 |
+
"loss": 0.7773,
|
1800 |
+
"step": 225
|
1801 |
+
},
|
1802 |
+
{
|
1803 |
+
"epoch": 18.833333333333332,
|
1804 |
+
"grad_norm": 14.062512397766113,
|
1805 |
+
"learning_rate": 4.671351798263746e-05,
|
1806 |
+
"loss": 0.8552,
|
1807 |
+
"step": 226
|
1808 |
+
},
|
1809 |
+
{
|
1810 |
+
"epoch": 18.916666666666668,
|
1811 |
+
"grad_norm": 8.626992225646973,
|
1812 |
+
"learning_rate": 4.692021496486151e-05,
|
1813 |
+
"loss": 0.7621,
|
1814 |
+
"step": 227
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 19.0,
|
1818 |
+
"grad_norm": 7.3263092041015625,
|
1819 |
+
"learning_rate": 4.712691194708557e-05,
|
1820 |
+
"loss": 0.7701,
|
1821 |
+
"step": 228
|
1822 |
+
},
|
1823 |
+
{
|
1824 |
+
"epoch": 19.0,
|
1825 |
+
"eval_accuracy": 0.8388252148997135,
|
1826 |
+
"eval_f1": 0.8501997336884154,
|
1827 |
+
"eval_loss": 0.3791760504245758,
|
1828 |
+
"eval_precision": 0.798125,
|
1829 |
+
"eval_recall": 0.9095441595441596,
|
1830 |
+
"eval_runtime": 1.0638,
|
1831 |
+
"eval_samples_per_second": 5248.936,
|
1832 |
+
"eval_steps_per_second": 10.34,
|
1833 |
+
"step": 228
|
1834 |
+
}
|
1835 |
+
],
|
1836 |
+
"logging_steps": 1,
|
1837 |
+
"max_steps": 240,
|
1838 |
+
"num_input_tokens_seen": 0,
|
1839 |
+
"num_train_epochs": 20,
|
1840 |
+
"save_steps": 500,
|
1841 |
+
"stateful_callbacks": {
|
1842 |
+
"TrainerControl": {
|
1843 |
+
"args": {
|
1844 |
+
"should_epoch_stop": false,
|
1845 |
+
"should_evaluate": false,
|
1846 |
+
"should_log": false,
|
1847 |
+
"should_save": true,
|
1848 |
+
"should_training_stop": false
|
1849 |
+
},
|
1850 |
+
"attributes": {}
|
1851 |
+
}
|
1852 |
+
},
|
1853 |
+
"total_flos": 9059803140280320.0,
|
1854 |
+
"train_batch_size": 512,
|
1855 |
+
"trial_name": null,
|
1856 |
+
"trial_params": null
|
1857 |
+
}
|
checkpoint-228/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:06d7f0ea24f0d299e0ee2b35fddd8f36d0fb779c6c2dcb1fa3fff86ad67f86c0
|
3 |
+
size 5368
|
config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"ViTForImageClassification"
|
4 |
+
],
|
5 |
+
"attention_probs_dropout_prob": 0.0,
|
6 |
+
"encoder_stride": 16,
|
7 |
+
"hidden_act": "gelu",
|
8 |
+
"hidden_dropout_prob": 0.0,
|
9 |
+
"hidden_size": 256,
|
10 |
+
"image_size": 32,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 512,
|
13 |
+
"layer_norm_eps": 1e-12,
|
14 |
+
"model_type": "vit",
|
15 |
+
"num_attention_heads": 8,
|
16 |
+
"num_channels": 3,
|
17 |
+
"num_hidden_layers": 4,
|
18 |
+
"patch_size": 4,
|
19 |
+
"qkv_bias": true,
|
20 |
+
"torch_dtype": "float32",
|
21 |
+
"transformers_version": "4.46.1"
|
22 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3fa9944531baa929a509197683aeb130fea2405d15cdfe2d93eb632f1a1d9c27
|
3 |
+
size 8563512
|
pr_curve.png
ADDED
![]() |
preprocessor_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"do_rescale": true,
|
4 |
+
"do_resize": false,
|
5 |
+
"image_mean": [
|
6 |
+
0.5,
|
7 |
+
0.5,
|
8 |
+
0.5
|
9 |
+
],
|
10 |
+
"image_processor_type": "ViTFeatureExtractor",
|
11 |
+
"image_std": [
|
12 |
+
0.5,
|
13 |
+
0.5,
|
14 |
+
0.5
|
15 |
+
],
|
16 |
+
"resample": 2,
|
17 |
+
"rescale_factor": 0.00392156862745098,
|
18 |
+
"size": {
|
19 |
+
"height": 32,
|
20 |
+
"width": 32
|
21 |
+
}
|
22 |
+
}
|
roc_curve.png
ADDED
![]() |
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:06d7f0ea24f0d299e0ee2b35fddd8f36d0fb779c6c2dcb1fa3fff86ad67f86c0
|
3 |
+
size 5368
|