jamescalam
commited on
Commit
•
b363d25
1
Parent(s):
37ac69a
first model
Browse files- all_results.json +13 -0
- config.json +48 -0
- eval_results.json +8 -0
- preprocessor_config.json +17 -0
- pytorch_model.bin +3 -0
- train_results.json +8 -0
- trainer_state.json +4333 -0
- training_args.bin +3 -0
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9812,
|
4 |
+
"eval_loss": 0.07883724570274353,
|
5 |
+
"eval_runtime": 61.8224,
|
6 |
+
"eval_samples_per_second": 161.754,
|
7 |
+
"eval_steps_per_second": 10.11,
|
8 |
+
"total_flos": 1.54995091808256e+19,
|
9 |
+
"train_loss": 0.08309991182349008,
|
10 |
+
"train_runtime": 5212.8854,
|
11 |
+
"train_samples_per_second": 38.366,
|
12 |
+
"train_steps_per_second": 1.199
|
13 |
+
}
|
config.json
ADDED
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/vit-base-patch16-224-in21k",
|
3 |
+
"architectures": [
|
4 |
+
"ViTForImageClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.0,
|
7 |
+
"encoder_stride": 16,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_dropout_prob": 0.0,
|
10 |
+
"hidden_size": 768,
|
11 |
+
"id2label": {
|
12 |
+
"0": "airplane",
|
13 |
+
"1": "automobile",
|
14 |
+
"2": "bird",
|
15 |
+
"3": "cat",
|
16 |
+
"4": "deer",
|
17 |
+
"5": "dog",
|
18 |
+
"6": "frog",
|
19 |
+
"7": "horse",
|
20 |
+
"8": "ship",
|
21 |
+
"9": "truck"
|
22 |
+
},
|
23 |
+
"image_size": 224,
|
24 |
+
"initializer_range": 0.02,
|
25 |
+
"intermediate_size": 3072,
|
26 |
+
"label2id": {
|
27 |
+
"airplane": "0",
|
28 |
+
"automobile": "1",
|
29 |
+
"bird": "2",
|
30 |
+
"cat": "3",
|
31 |
+
"deer": "4",
|
32 |
+
"dog": "5",
|
33 |
+
"frog": "6",
|
34 |
+
"horse": "7",
|
35 |
+
"ship": "8",
|
36 |
+
"truck": "9"
|
37 |
+
},
|
38 |
+
"layer_norm_eps": 1e-12,
|
39 |
+
"model_type": "vit",
|
40 |
+
"num_attention_heads": 12,
|
41 |
+
"num_channels": 3,
|
42 |
+
"num_hidden_layers": 12,
|
43 |
+
"patch_size": 16,
|
44 |
+
"problem_type": "single_label_classification",
|
45 |
+
"qkv_bias": true,
|
46 |
+
"torch_dtype": "float32",
|
47 |
+
"transformers_version": "4.21.3"
|
48 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9812,
|
4 |
+
"eval_loss": 0.07883724570274353,
|
5 |
+
"eval_runtime": 61.8224,
|
6 |
+
"eval_samples_per_second": 161.754,
|
7 |
+
"eval_steps_per_second": 10.11
|
8 |
+
}
|
preprocessor_config.json
ADDED
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"do_resize": true,
|
4 |
+
"feature_extractor_type": "ViTFeatureExtractor",
|
5 |
+
"image_mean": [
|
6 |
+
0.5,
|
7 |
+
0.5,
|
8 |
+
0.5
|
9 |
+
],
|
10 |
+
"image_std": [
|
11 |
+
0.5,
|
12 |
+
0.5,
|
13 |
+
0.5
|
14 |
+
],
|
15 |
+
"resample": 2,
|
16 |
+
"size": 224
|
17 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f935fa81e1693f5b6508224376a63ddac62a9689f5760cc50c8e0bb6d50c6061
|
3 |
+
size 343291569
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"total_flos": 1.54995091808256e+19,
|
4 |
+
"train_loss": 0.08309991182349008,
|
5 |
+
"train_runtime": 5212.8854,
|
6 |
+
"train_samples_per_second": 38.366,
|
7 |
+
"train_steps_per_second": 1.199
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,4333 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.07883724570274353,
|
3 |
+
"best_model_checkpoint": "./cifar/checkpoint-6200",
|
4 |
+
"epoch": 4.0,
|
5 |
+
"global_step": 6252,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 0.0001996801023672425,
|
13 |
+
"loss": 2.0965,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.01,
|
18 |
+
"learning_rate": 0.00019936020473448497,
|
19 |
+
"loss": 1.4515,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.02,
|
24 |
+
"learning_rate": 0.00019904030710172745,
|
25 |
+
"loss": 0.9389,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.03,
|
30 |
+
"learning_rate": 0.00019872040946896996,
|
31 |
+
"loss": 0.689,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.03,
|
36 |
+
"learning_rate": 0.0001984005118362124,
|
37 |
+
"loss": 0.4917,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.04,
|
42 |
+
"learning_rate": 0.0001980806142034549,
|
43 |
+
"loss": 0.4551,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.04,
|
48 |
+
"learning_rate": 0.0001977607165706974,
|
49 |
+
"loss": 0.4832,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.05,
|
54 |
+
"learning_rate": 0.00019744081893793988,
|
55 |
+
"loss": 0.4072,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.06,
|
60 |
+
"learning_rate": 0.00019712092130518236,
|
61 |
+
"loss": 0.3861,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.06,
|
66 |
+
"learning_rate": 0.00019680102367242483,
|
67 |
+
"loss": 0.3212,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.06,
|
72 |
+
"eval_accuracy": 0.8785,
|
73 |
+
"eval_loss": 0.47495898604393005,
|
74 |
+
"eval_runtime": 47.4814,
|
75 |
+
"eval_samples_per_second": 210.609,
|
76 |
+
"eval_steps_per_second": 13.163,
|
77 |
+
"step": 100
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 0.07,
|
81 |
+
"learning_rate": 0.00019648112603966731,
|
82 |
+
"loss": 0.3596,
|
83 |
+
"step": 110
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 0.08,
|
87 |
+
"learning_rate": 0.0001961612284069098,
|
88 |
+
"loss": 0.3605,
|
89 |
+
"step": 120
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.08,
|
93 |
+
"learning_rate": 0.00019584133077415227,
|
94 |
+
"loss": 0.3013,
|
95 |
+
"step": 130
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 0.09,
|
99 |
+
"learning_rate": 0.00019552143314139478,
|
100 |
+
"loss": 0.3512,
|
101 |
+
"step": 140
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 0.1,
|
105 |
+
"learning_rate": 0.00019520153550863723,
|
106 |
+
"loss": 0.3202,
|
107 |
+
"step": 150
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.1,
|
111 |
+
"learning_rate": 0.00019488163787587971,
|
112 |
+
"loss": 0.2308,
|
113 |
+
"step": 160
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 0.11,
|
117 |
+
"learning_rate": 0.00019456174024312222,
|
118 |
+
"loss": 0.4124,
|
119 |
+
"step": 170
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 0.12,
|
123 |
+
"learning_rate": 0.0001942418426103647,
|
124 |
+
"loss": 0.2991,
|
125 |
+
"step": 180
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 0.12,
|
129 |
+
"learning_rate": 0.00019392194497760718,
|
130 |
+
"loss": 0.2891,
|
131 |
+
"step": 190
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 0.13,
|
135 |
+
"learning_rate": 0.00019360204734484966,
|
136 |
+
"loss": 0.3145,
|
137 |
+
"step": 200
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.13,
|
141 |
+
"eval_accuracy": 0.9347,
|
142 |
+
"eval_loss": 0.2536068856716156,
|
143 |
+
"eval_runtime": 48.1227,
|
144 |
+
"eval_samples_per_second": 207.802,
|
145 |
+
"eval_steps_per_second": 12.988,
|
146 |
+
"step": 200
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.13,
|
150 |
+
"learning_rate": 0.00019328214971209214,
|
151 |
+
"loss": 0.1861,
|
152 |
+
"step": 210
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.14,
|
156 |
+
"learning_rate": 0.00019296225207933462,
|
157 |
+
"loss": 0.3056,
|
158 |
+
"step": 220
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.15,
|
162 |
+
"learning_rate": 0.0001926423544465771,
|
163 |
+
"loss": 0.3428,
|
164 |
+
"step": 230
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.15,
|
168 |
+
"learning_rate": 0.0001923224568138196,
|
169 |
+
"loss": 0.1819,
|
170 |
+
"step": 240
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.16,
|
174 |
+
"learning_rate": 0.00019200255918106206,
|
175 |
+
"loss": 0.2286,
|
176 |
+
"step": 250
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.17,
|
180 |
+
"learning_rate": 0.00019168266154830454,
|
181 |
+
"loss": 0.1687,
|
182 |
+
"step": 260
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.17,
|
186 |
+
"learning_rate": 0.00019136276391554705,
|
187 |
+
"loss": 0.2,
|
188 |
+
"step": 270
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.18,
|
192 |
+
"learning_rate": 0.0001910428662827895,
|
193 |
+
"loss": 0.2408,
|
194 |
+
"step": 280
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.19,
|
198 |
+
"learning_rate": 0.000190722968650032,
|
199 |
+
"loss": 0.2047,
|
200 |
+
"step": 290
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.19,
|
204 |
+
"learning_rate": 0.00019040307101727449,
|
205 |
+
"loss": 0.3278,
|
206 |
+
"step": 300
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.19,
|
210 |
+
"eval_accuracy": 0.9314,
|
211 |
+
"eval_loss": 0.2535996735095978,
|
212 |
+
"eval_runtime": 47.8164,
|
213 |
+
"eval_samples_per_second": 209.133,
|
214 |
+
"eval_steps_per_second": 13.071,
|
215 |
+
"step": 300
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 0.2,
|
219 |
+
"learning_rate": 0.00019008317338451697,
|
220 |
+
"loss": 0.2192,
|
221 |
+
"step": 310
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 0.2,
|
225 |
+
"learning_rate": 0.00018976327575175944,
|
226 |
+
"loss": 0.244,
|
227 |
+
"step": 320
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 0.21,
|
231 |
+
"learning_rate": 0.00018944337811900192,
|
232 |
+
"loss": 0.2785,
|
233 |
+
"step": 330
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.22,
|
237 |
+
"learning_rate": 0.00018912348048624443,
|
238 |
+
"loss": 0.1992,
|
239 |
+
"step": 340
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 0.22,
|
243 |
+
"learning_rate": 0.00018880358285348688,
|
244 |
+
"loss": 0.2039,
|
245 |
+
"step": 350
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 0.23,
|
249 |
+
"learning_rate": 0.00018848368522072936,
|
250 |
+
"loss": 0.2919,
|
251 |
+
"step": 360
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 0.24,
|
255 |
+
"learning_rate": 0.00018816378758797187,
|
256 |
+
"loss": 0.2075,
|
257 |
+
"step": 370
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 0.24,
|
261 |
+
"learning_rate": 0.00018784388995521432,
|
262 |
+
"loss": 0.2507,
|
263 |
+
"step": 380
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 0.25,
|
267 |
+
"learning_rate": 0.00018752399232245683,
|
268 |
+
"loss": 0.1471,
|
269 |
+
"step": 390
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.26,
|
273 |
+
"learning_rate": 0.0001872040946896993,
|
274 |
+
"loss": 0.2429,
|
275 |
+
"step": 400
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.26,
|
279 |
+
"eval_accuracy": 0.9265,
|
280 |
+
"eval_loss": 0.2561478912830353,
|
281 |
+
"eval_runtime": 48.1994,
|
282 |
+
"eval_samples_per_second": 207.472,
|
283 |
+
"eval_steps_per_second": 12.967,
|
284 |
+
"step": 400
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 0.26,
|
288 |
+
"learning_rate": 0.0001868841970569418,
|
289 |
+
"loss": 0.3039,
|
290 |
+
"step": 410
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 0.27,
|
294 |
+
"learning_rate": 0.00018656429942418427,
|
295 |
+
"loss": 0.3438,
|
296 |
+
"step": 420
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.28,
|
300 |
+
"learning_rate": 0.00018624440179142675,
|
301 |
+
"loss": 0.208,
|
302 |
+
"step": 430
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 0.28,
|
306 |
+
"learning_rate": 0.00018592450415866926,
|
307 |
+
"loss": 0.1909,
|
308 |
+
"step": 440
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 0.29,
|
312 |
+
"learning_rate": 0.0001856046065259117,
|
313 |
+
"loss": 0.2136,
|
314 |
+
"step": 450
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 0.29,
|
318 |
+
"learning_rate": 0.0001852847088931542,
|
319 |
+
"loss": 0.2753,
|
320 |
+
"step": 460
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 0.3,
|
324 |
+
"learning_rate": 0.0001849648112603967,
|
325 |
+
"loss": 0.2285,
|
326 |
+
"step": 470
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 0.31,
|
330 |
+
"learning_rate": 0.00018464491362763915,
|
331 |
+
"loss": 0.2331,
|
332 |
+
"step": 480
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 0.31,
|
336 |
+
"learning_rate": 0.00018432501599488166,
|
337 |
+
"loss": 0.2163,
|
338 |
+
"step": 490
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.32,
|
342 |
+
"learning_rate": 0.00018400511836212414,
|
343 |
+
"loss": 0.2622,
|
344 |
+
"step": 500
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 0.32,
|
348 |
+
"eval_accuracy": 0.9447,
|
349 |
+
"eval_loss": 0.19128669798374176,
|
350 |
+
"eval_runtime": 48.5832,
|
351 |
+
"eval_samples_per_second": 205.832,
|
352 |
+
"eval_steps_per_second": 12.865,
|
353 |
+
"step": 500
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 0.33,
|
357 |
+
"learning_rate": 0.00018368522072936662,
|
358 |
+
"loss": 0.1883,
|
359 |
+
"step": 510
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.33,
|
363 |
+
"learning_rate": 0.0001833653230966091,
|
364 |
+
"loss": 0.2193,
|
365 |
+
"step": 520
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 0.34,
|
369 |
+
"learning_rate": 0.00018304542546385158,
|
370 |
+
"loss": 0.1819,
|
371 |
+
"step": 530
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 0.35,
|
375 |
+
"learning_rate": 0.00018272552783109406,
|
376 |
+
"loss": 0.2348,
|
377 |
+
"step": 540
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 0.35,
|
381 |
+
"learning_rate": 0.00018240563019833653,
|
382 |
+
"loss": 0.1955,
|
383 |
+
"step": 550
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 0.36,
|
387 |
+
"learning_rate": 0.00018208573256557901,
|
388 |
+
"loss": 0.1497,
|
389 |
+
"step": 560
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 0.36,
|
393 |
+
"learning_rate": 0.00018176583493282152,
|
394 |
+
"loss": 0.1614,
|
395 |
+
"step": 570
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 0.37,
|
399 |
+
"learning_rate": 0.00018144593730006397,
|
400 |
+
"loss": 0.1747,
|
401 |
+
"step": 580
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.38,
|
405 |
+
"learning_rate": 0.00018112603966730648,
|
406 |
+
"loss": 0.1957,
|
407 |
+
"step": 590
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 0.38,
|
411 |
+
"learning_rate": 0.00018080614203454896,
|
412 |
+
"loss": 0.1714,
|
413 |
+
"step": 600
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 0.38,
|
417 |
+
"eval_accuracy": 0.939,
|
418 |
+
"eval_loss": 0.21139121055603027,
|
419 |
+
"eval_runtime": 48.6064,
|
420 |
+
"eval_samples_per_second": 205.734,
|
421 |
+
"eval_steps_per_second": 12.858,
|
422 |
+
"step": 600
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.39,
|
426 |
+
"learning_rate": 0.00018048624440179141,
|
427 |
+
"loss": 0.2064,
|
428 |
+
"step": 610
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.4,
|
432 |
+
"learning_rate": 0.00018016634676903392,
|
433 |
+
"loss": 0.2239,
|
434 |
+
"step": 620
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.4,
|
438 |
+
"learning_rate": 0.0001798464491362764,
|
439 |
+
"loss": 0.1753,
|
440 |
+
"step": 630
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 0.41,
|
444 |
+
"learning_rate": 0.00017952655150351888,
|
445 |
+
"loss": 0.2009,
|
446 |
+
"step": 640
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 0.42,
|
450 |
+
"learning_rate": 0.00017920665387076136,
|
451 |
+
"loss": 0.1754,
|
452 |
+
"step": 650
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 0.42,
|
456 |
+
"learning_rate": 0.00017888675623800384,
|
457 |
+
"loss": 0.1365,
|
458 |
+
"step": 660
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.43,
|
462 |
+
"learning_rate": 0.00017856685860524635,
|
463 |
+
"loss": 0.142,
|
464 |
+
"step": 670
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.44,
|
468 |
+
"learning_rate": 0.0001782469609724888,
|
469 |
+
"loss": 0.2181,
|
470 |
+
"step": 680
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 0.44,
|
474 |
+
"learning_rate": 0.0001779270633397313,
|
475 |
+
"loss": 0.1755,
|
476 |
+
"step": 690
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.45,
|
480 |
+
"learning_rate": 0.00017760716570697379,
|
481 |
+
"loss": 0.1966,
|
482 |
+
"step": 700
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 0.45,
|
486 |
+
"eval_accuracy": 0.9348,
|
487 |
+
"eval_loss": 0.21518036723136902,
|
488 |
+
"eval_runtime": 48.7309,
|
489 |
+
"eval_samples_per_second": 205.209,
|
490 |
+
"eval_steps_per_second": 12.826,
|
491 |
+
"step": 700
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 0.45,
|
495 |
+
"learning_rate": 0.00017728726807421624,
|
496 |
+
"loss": 0.1925,
|
497 |
+
"step": 710
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 0.46,
|
501 |
+
"learning_rate": 0.00017696737044145875,
|
502 |
+
"loss": 0.2256,
|
503 |
+
"step": 720
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 0.47,
|
507 |
+
"learning_rate": 0.00017664747280870123,
|
508 |
+
"loss": 0.1714,
|
509 |
+
"step": 730
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 0.47,
|
513 |
+
"learning_rate": 0.0001763275751759437,
|
514 |
+
"loss": 0.2855,
|
515 |
+
"step": 740
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 0.48,
|
519 |
+
"learning_rate": 0.00017600767754318619,
|
520 |
+
"loss": 0.2259,
|
521 |
+
"step": 750
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 0.49,
|
525 |
+
"learning_rate": 0.00017568777991042867,
|
526 |
+
"loss": 0.1718,
|
527 |
+
"step": 760
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.49,
|
531 |
+
"learning_rate": 0.00017536788227767115,
|
532 |
+
"loss": 0.1441,
|
533 |
+
"step": 770
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 0.5,
|
537 |
+
"learning_rate": 0.00017504798464491362,
|
538 |
+
"loss": 0.3405,
|
539 |
+
"step": 780
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 0.51,
|
543 |
+
"learning_rate": 0.00017472808701215613,
|
544 |
+
"loss": 0.1524,
|
545 |
+
"step": 790
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 0.51,
|
549 |
+
"learning_rate": 0.0001744081893793986,
|
550 |
+
"loss": 0.1395,
|
551 |
+
"step": 800
|
552 |
+
},
|
553 |
+
{
|
554 |
+
"epoch": 0.51,
|
555 |
+
"eval_accuracy": 0.9453,
|
556 |
+
"eval_loss": 0.18169045448303223,
|
557 |
+
"eval_runtime": 47.7658,
|
558 |
+
"eval_samples_per_second": 209.355,
|
559 |
+
"eval_steps_per_second": 13.085,
|
560 |
+
"step": 800
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 0.52,
|
564 |
+
"learning_rate": 0.00017408829174664106,
|
565 |
+
"loss": 0.134,
|
566 |
+
"step": 810
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 0.52,
|
570 |
+
"learning_rate": 0.00017376839411388357,
|
571 |
+
"loss": 0.1626,
|
572 |
+
"step": 820
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 0.53,
|
576 |
+
"learning_rate": 0.00017344849648112605,
|
577 |
+
"loss": 0.2228,
|
578 |
+
"step": 830
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 0.54,
|
582 |
+
"learning_rate": 0.00017312859884836853,
|
583 |
+
"loss": 0.204,
|
584 |
+
"step": 840
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 0.54,
|
588 |
+
"learning_rate": 0.000172808701215611,
|
589 |
+
"loss": 0.0642,
|
590 |
+
"step": 850
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.55,
|
594 |
+
"learning_rate": 0.0001724888035828535,
|
595 |
+
"loss": 0.215,
|
596 |
+
"step": 860
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 0.56,
|
600 |
+
"learning_rate": 0.00017216890595009597,
|
601 |
+
"loss": 0.1824,
|
602 |
+
"step": 870
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 0.56,
|
606 |
+
"learning_rate": 0.00017184900831733845,
|
607 |
+
"loss": 0.1739,
|
608 |
+
"step": 880
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 0.57,
|
612 |
+
"learning_rate": 0.00017152911068458096,
|
613 |
+
"loss": 0.1884,
|
614 |
+
"step": 890
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 0.58,
|
618 |
+
"learning_rate": 0.00017120921305182344,
|
619 |
+
"loss": 0.1323,
|
620 |
+
"step": 900
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 0.58,
|
624 |
+
"eval_accuracy": 0.9324,
|
625 |
+
"eval_loss": 0.2200065404176712,
|
626 |
+
"eval_runtime": 47.5971,
|
627 |
+
"eval_samples_per_second": 210.097,
|
628 |
+
"eval_steps_per_second": 13.131,
|
629 |
+
"step": 900
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 0.58,
|
633 |
+
"learning_rate": 0.0001708893154190659,
|
634 |
+
"loss": 0.1711,
|
635 |
+
"step": 910
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 0.59,
|
639 |
+
"learning_rate": 0.0001705694177863084,
|
640 |
+
"loss": 0.1293,
|
641 |
+
"step": 920
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 0.6,
|
645 |
+
"learning_rate": 0.00017024952015355088,
|
646 |
+
"loss": 0.2275,
|
647 |
+
"step": 930
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 0.6,
|
651 |
+
"learning_rate": 0.00016992962252079336,
|
652 |
+
"loss": 0.1531,
|
653 |
+
"step": 940
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.61,
|
657 |
+
"learning_rate": 0.00016960972488803584,
|
658 |
+
"loss": 0.1187,
|
659 |
+
"step": 950
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 0.61,
|
663 |
+
"learning_rate": 0.00016928982725527832,
|
664 |
+
"loss": 0.158,
|
665 |
+
"step": 960
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 0.62,
|
669 |
+
"learning_rate": 0.0001689699296225208,
|
670 |
+
"loss": 0.1857,
|
671 |
+
"step": 970
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 0.63,
|
675 |
+
"learning_rate": 0.00016865003198976328,
|
676 |
+
"loss": 0.142,
|
677 |
+
"step": 980
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 0.63,
|
681 |
+
"learning_rate": 0.00016833013435700578,
|
682 |
+
"loss": 0.2363,
|
683 |
+
"step": 990
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 0.64,
|
687 |
+
"learning_rate": 0.00016801023672424826,
|
688 |
+
"loss": 0.1707,
|
689 |
+
"step": 1000
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 0.64,
|
693 |
+
"eval_accuracy": 0.945,
|
694 |
+
"eval_loss": 0.18686933815479279,
|
695 |
+
"eval_runtime": 47.6282,
|
696 |
+
"eval_samples_per_second": 209.96,
|
697 |
+
"eval_steps_per_second": 13.122,
|
698 |
+
"step": 1000
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 0.65,
|
702 |
+
"learning_rate": 0.00016769033909149071,
|
703 |
+
"loss": 0.2039,
|
704 |
+
"step": 1010
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 0.65,
|
708 |
+
"learning_rate": 0.00016737044145873322,
|
709 |
+
"loss": 0.146,
|
710 |
+
"step": 1020
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 0.66,
|
714 |
+
"learning_rate": 0.0001670505438259757,
|
715 |
+
"loss": 0.1763,
|
716 |
+
"step": 1030
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.67,
|
720 |
+
"learning_rate": 0.00016673064619321818,
|
721 |
+
"loss": 0.1645,
|
722 |
+
"step": 1040
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 0.67,
|
726 |
+
"learning_rate": 0.00016641074856046066,
|
727 |
+
"loss": 0.1368,
|
728 |
+
"step": 1050
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 0.68,
|
732 |
+
"learning_rate": 0.00016609085092770314,
|
733 |
+
"loss": 0.1442,
|
734 |
+
"step": 1060
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 0.68,
|
738 |
+
"learning_rate": 0.00016577095329494562,
|
739 |
+
"loss": 0.1986,
|
740 |
+
"step": 1070
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 0.69,
|
744 |
+
"learning_rate": 0.0001654510556621881,
|
745 |
+
"loss": 0.1744,
|
746 |
+
"step": 1080
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 0.7,
|
750 |
+
"learning_rate": 0.0001651311580294306,
|
751 |
+
"loss": 0.1307,
|
752 |
+
"step": 1090
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 0.7,
|
756 |
+
"learning_rate": 0.00016481126039667306,
|
757 |
+
"loss": 0.1799,
|
758 |
+
"step": 1100
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.7,
|
762 |
+
"eval_accuracy": 0.9447,
|
763 |
+
"eval_loss": 0.18500135838985443,
|
764 |
+
"eval_runtime": 47.9607,
|
765 |
+
"eval_samples_per_second": 208.504,
|
766 |
+
"eval_steps_per_second": 13.032,
|
767 |
+
"step": 1100
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 0.71,
|
771 |
+
"learning_rate": 0.00016449136276391554,
|
772 |
+
"loss": 0.1933,
|
773 |
+
"step": 1110
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 0.72,
|
777 |
+
"learning_rate": 0.00016417146513115805,
|
778 |
+
"loss": 0.1567,
|
779 |
+
"step": 1120
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.72,
|
783 |
+
"learning_rate": 0.00016385156749840053,
|
784 |
+
"loss": 0.2155,
|
785 |
+
"step": 1130
|
786 |
+
},
|
787 |
+
{
|
788 |
+
"epoch": 0.73,
|
789 |
+
"learning_rate": 0.000163531669865643,
|
790 |
+
"loss": 0.2179,
|
791 |
+
"step": 1140
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 0.74,
|
795 |
+
"learning_rate": 0.0001632117722328855,
|
796 |
+
"loss": 0.2175,
|
797 |
+
"step": 1150
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 0.74,
|
801 |
+
"learning_rate": 0.00016289187460012797,
|
802 |
+
"loss": 0.1873,
|
803 |
+
"step": 1160
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 0.75,
|
807 |
+
"learning_rate": 0.00016257197696737045,
|
808 |
+
"loss": 0.1778,
|
809 |
+
"step": 1170
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 0.75,
|
813 |
+
"learning_rate": 0.00016225207933461293,
|
814 |
+
"loss": 0.2057,
|
815 |
+
"step": 1180
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 0.76,
|
819 |
+
"learning_rate": 0.00016193218170185543,
|
820 |
+
"loss": 0.1469,
|
821 |
+
"step": 1190
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.77,
|
825 |
+
"learning_rate": 0.00016161228406909789,
|
826 |
+
"loss": 0.1104,
|
827 |
+
"step": 1200
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 0.77,
|
831 |
+
"eval_accuracy": 0.9376,
|
832 |
+
"eval_loss": 0.20156049728393555,
|
833 |
+
"eval_runtime": 48.4092,
|
834 |
+
"eval_samples_per_second": 206.572,
|
835 |
+
"eval_steps_per_second": 12.911,
|
836 |
+
"step": 1200
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 0.77,
|
840 |
+
"learning_rate": 0.00016129238643634037,
|
841 |
+
"loss": 0.22,
|
842 |
+
"step": 1210
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.78,
|
846 |
+
"learning_rate": 0.00016097248880358287,
|
847 |
+
"loss": 0.1224,
|
848 |
+
"step": 1220
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 0.79,
|
852 |
+
"learning_rate": 0.00016065259117082535,
|
853 |
+
"loss": 0.1591,
|
854 |
+
"step": 1230
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 0.79,
|
858 |
+
"learning_rate": 0.00016033269353806783,
|
859 |
+
"loss": 0.1492,
|
860 |
+
"step": 1240
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 0.8,
|
864 |
+
"learning_rate": 0.0001600127959053103,
|
865 |
+
"loss": 0.1605,
|
866 |
+
"step": 1250
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 0.81,
|
870 |
+
"learning_rate": 0.0001596928982725528,
|
871 |
+
"loss": 0.1457,
|
872 |
+
"step": 1260
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 0.81,
|
876 |
+
"learning_rate": 0.00015937300063979527,
|
877 |
+
"loss": 0.0612,
|
878 |
+
"step": 1270
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 0.82,
|
882 |
+
"learning_rate": 0.00015905310300703775,
|
883 |
+
"loss": 0.082,
|
884 |
+
"step": 1280
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.83,
|
888 |
+
"learning_rate": 0.00015873320537428026,
|
889 |
+
"loss": 0.0897,
|
890 |
+
"step": 1290
|
891 |
+
},
|
892 |
+
{
|
893 |
+
"epoch": 0.83,
|
894 |
+
"learning_rate": 0.0001584133077415227,
|
895 |
+
"loss": 0.1298,
|
896 |
+
"step": 1300
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 0.83,
|
900 |
+
"eval_accuracy": 0.9577,
|
901 |
+
"eval_loss": 0.13800279796123505,
|
902 |
+
"eval_runtime": 62.6571,
|
903 |
+
"eval_samples_per_second": 159.599,
|
904 |
+
"eval_steps_per_second": 9.975,
|
905 |
+
"step": 1300
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.84,
|
909 |
+
"learning_rate": 0.0001580934101087652,
|
910 |
+
"loss": 0.1759,
|
911 |
+
"step": 1310
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 0.84,
|
915 |
+
"learning_rate": 0.0001577735124760077,
|
916 |
+
"loss": 0.1883,
|
917 |
+
"step": 1320
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 0.85,
|
921 |
+
"learning_rate": 0.00015745361484325015,
|
922 |
+
"loss": 0.1002,
|
923 |
+
"step": 1330
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 0.86,
|
927 |
+
"learning_rate": 0.00015713371721049266,
|
928 |
+
"loss": 0.1287,
|
929 |
+
"step": 1340
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 0.86,
|
933 |
+
"learning_rate": 0.00015681381957773514,
|
934 |
+
"loss": 0.1834,
|
935 |
+
"step": 1350
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 0.87,
|
939 |
+
"learning_rate": 0.00015649392194497762,
|
940 |
+
"loss": 0.1503,
|
941 |
+
"step": 1360
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 0.88,
|
945 |
+
"learning_rate": 0.0001561740243122201,
|
946 |
+
"loss": 0.2471,
|
947 |
+
"step": 1370
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.88,
|
951 |
+
"learning_rate": 0.00015585412667946258,
|
952 |
+
"loss": 0.164,
|
953 |
+
"step": 1380
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 0.89,
|
957 |
+
"learning_rate": 0.00015553422904670508,
|
958 |
+
"loss": 0.1632,
|
959 |
+
"step": 1390
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 0.9,
|
963 |
+
"learning_rate": 0.00015521433141394754,
|
964 |
+
"loss": 0.1124,
|
965 |
+
"step": 1400
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 0.9,
|
969 |
+
"eval_accuracy": 0.957,
|
970 |
+
"eval_loss": 0.14739342033863068,
|
971 |
+
"eval_runtime": 60.3393,
|
972 |
+
"eval_samples_per_second": 165.729,
|
973 |
+
"eval_steps_per_second": 10.358,
|
974 |
+
"step": 1400
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 0.9,
|
978 |
+
"learning_rate": 0.00015489443378119002,
|
979 |
+
"loss": 0.1396,
|
980 |
+
"step": 1410
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 0.91,
|
984 |
+
"learning_rate": 0.00015457453614843252,
|
985 |
+
"loss": 0.1766,
|
986 |
+
"step": 1420
|
987 |
+
},
|
988 |
+
{
|
989 |
+
"epoch": 0.91,
|
990 |
+
"learning_rate": 0.00015425463851567498,
|
991 |
+
"loss": 0.1866,
|
992 |
+
"step": 1430
|
993 |
+
},
|
994 |
+
{
|
995 |
+
"epoch": 0.92,
|
996 |
+
"learning_rate": 0.00015393474088291748,
|
997 |
+
"loss": 0.1443,
|
998 |
+
"step": 1440
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 0.93,
|
1002 |
+
"learning_rate": 0.00015361484325015996,
|
1003 |
+
"loss": 0.0976,
|
1004 |
+
"step": 1450
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 0.93,
|
1008 |
+
"learning_rate": 0.00015329494561740244,
|
1009 |
+
"loss": 0.1618,
|
1010 |
+
"step": 1460
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.94,
|
1014 |
+
"learning_rate": 0.00015297504798464492,
|
1015 |
+
"loss": 0.095,
|
1016 |
+
"step": 1470
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 0.95,
|
1020 |
+
"learning_rate": 0.0001526551503518874,
|
1021 |
+
"loss": 0.1946,
|
1022 |
+
"step": 1480
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 0.95,
|
1026 |
+
"learning_rate": 0.00015233525271912988,
|
1027 |
+
"loss": 0.1362,
|
1028 |
+
"step": 1490
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 0.96,
|
1032 |
+
"learning_rate": 0.00015201535508637236,
|
1033 |
+
"loss": 0.2286,
|
1034 |
+
"step": 1500
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 0.96,
|
1038 |
+
"eval_accuracy": 0.9426,
|
1039 |
+
"eval_loss": 0.18567727506160736,
|
1040 |
+
"eval_runtime": 60.124,
|
1041 |
+
"eval_samples_per_second": 166.323,
|
1042 |
+
"eval_steps_per_second": 10.395,
|
1043 |
+
"step": 1500
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 0.97,
|
1047 |
+
"learning_rate": 0.00015169545745361484,
|
1048 |
+
"loss": 0.168,
|
1049 |
+
"step": 1510
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 0.97,
|
1053 |
+
"learning_rate": 0.00015137555982085735,
|
1054 |
+
"loss": 0.085,
|
1055 |
+
"step": 1520
|
1056 |
+
},
|
1057 |
+
{
|
1058 |
+
"epoch": 0.98,
|
1059 |
+
"learning_rate": 0.0001510556621880998,
|
1060 |
+
"loss": 0.1409,
|
1061 |
+
"step": 1530
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 0.99,
|
1065 |
+
"learning_rate": 0.0001507357645553423,
|
1066 |
+
"loss": 0.2012,
|
1067 |
+
"step": 1540
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 0.99,
|
1071 |
+
"learning_rate": 0.0001504158669225848,
|
1072 |
+
"loss": 0.1522,
|
1073 |
+
"step": 1550
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 1.0,
|
1077 |
+
"learning_rate": 0.00015009596928982727,
|
1078 |
+
"loss": 0.0599,
|
1079 |
+
"step": 1560
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 1.0,
|
1083 |
+
"learning_rate": 0.00014977607165706975,
|
1084 |
+
"loss": 0.0955,
|
1085 |
+
"step": 1570
|
1086 |
+
},
|
1087 |
+
{
|
1088 |
+
"epoch": 1.01,
|
1089 |
+
"learning_rate": 0.00014945617402431223,
|
1090 |
+
"loss": 0.0773,
|
1091 |
+
"step": 1580
|
1092 |
+
},
|
1093 |
+
{
|
1094 |
+
"epoch": 1.02,
|
1095 |
+
"learning_rate": 0.0001491362763915547,
|
1096 |
+
"loss": 0.0747,
|
1097 |
+
"step": 1590
|
1098 |
+
},
|
1099 |
+
{
|
1100 |
+
"epoch": 1.02,
|
1101 |
+
"learning_rate": 0.0001488163787587972,
|
1102 |
+
"loss": 0.0489,
|
1103 |
+
"step": 1600
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 1.02,
|
1107 |
+
"eval_accuracy": 0.9683,
|
1108 |
+
"eval_loss": 0.10664957016706467,
|
1109 |
+
"eval_runtime": 53.6606,
|
1110 |
+
"eval_samples_per_second": 186.356,
|
1111 |
+
"eval_steps_per_second": 11.647,
|
1112 |
+
"step": 1600
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 1.03,
|
1116 |
+
"learning_rate": 0.00014849648112603967,
|
1117 |
+
"loss": 0.0523,
|
1118 |
+
"step": 1610
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 1.04,
|
1122 |
+
"learning_rate": 0.00014817658349328217,
|
1123 |
+
"loss": 0.0387,
|
1124 |
+
"step": 1620
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 1.04,
|
1128 |
+
"learning_rate": 0.00014785668586052463,
|
1129 |
+
"loss": 0.0738,
|
1130 |
+
"step": 1630
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 1.05,
|
1134 |
+
"learning_rate": 0.00014753678822776713,
|
1135 |
+
"loss": 0.0868,
|
1136 |
+
"step": 1640
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 1.06,
|
1140 |
+
"learning_rate": 0.0001472168905950096,
|
1141 |
+
"loss": 0.0614,
|
1142 |
+
"step": 1650
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 1.06,
|
1146 |
+
"learning_rate": 0.00014689699296225207,
|
1147 |
+
"loss": 0.0782,
|
1148 |
+
"step": 1660
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 1.07,
|
1152 |
+
"learning_rate": 0.00014657709532949457,
|
1153 |
+
"loss": 0.1417,
|
1154 |
+
"step": 1670
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 1.07,
|
1158 |
+
"learning_rate": 0.00014625719769673705,
|
1159 |
+
"loss": 0.07,
|
1160 |
+
"step": 1680
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 1.08,
|
1164 |
+
"learning_rate": 0.00014593730006397953,
|
1165 |
+
"loss": 0.0856,
|
1166 |
+
"step": 1690
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 1.09,
|
1170 |
+
"learning_rate": 0.000145617402431222,
|
1171 |
+
"loss": 0.0829,
|
1172 |
+
"step": 1700
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 1.09,
|
1176 |
+
"eval_accuracy": 0.9584,
|
1177 |
+
"eval_loss": 0.14192388951778412,
|
1178 |
+
"eval_runtime": 47.2202,
|
1179 |
+
"eval_samples_per_second": 211.774,
|
1180 |
+
"eval_steps_per_second": 13.236,
|
1181 |
+
"step": 1700
|
1182 |
+
},
|
1183 |
+
{
|
1184 |
+
"epoch": 1.09,
|
1185 |
+
"learning_rate": 0.0001452975047984645,
|
1186 |
+
"loss": 0.0773,
|
1187 |
+
"step": 1710
|
1188 |
+
},
|
1189 |
+
{
|
1190 |
+
"epoch": 1.1,
|
1191 |
+
"learning_rate": 0.000144977607165707,
|
1192 |
+
"loss": 0.1056,
|
1193 |
+
"step": 1720
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 1.11,
|
1197 |
+
"learning_rate": 0.00014465770953294945,
|
1198 |
+
"loss": 0.1197,
|
1199 |
+
"step": 1730
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 1.11,
|
1203 |
+
"learning_rate": 0.00014433781190019196,
|
1204 |
+
"loss": 0.1031,
|
1205 |
+
"step": 1740
|
1206 |
+
},
|
1207 |
+
{
|
1208 |
+
"epoch": 1.12,
|
1209 |
+
"learning_rate": 0.00014401791426743444,
|
1210 |
+
"loss": 0.0558,
|
1211 |
+
"step": 1750
|
1212 |
+
},
|
1213 |
+
{
|
1214 |
+
"epoch": 1.13,
|
1215 |
+
"learning_rate": 0.0001436980166346769,
|
1216 |
+
"loss": 0.0797,
|
1217 |
+
"step": 1760
|
1218 |
+
},
|
1219 |
+
{
|
1220 |
+
"epoch": 1.13,
|
1221 |
+
"learning_rate": 0.0001433781190019194,
|
1222 |
+
"loss": 0.12,
|
1223 |
+
"step": 1770
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 1.14,
|
1227 |
+
"learning_rate": 0.00014305822136916188,
|
1228 |
+
"loss": 0.1093,
|
1229 |
+
"step": 1780
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 1.15,
|
1233 |
+
"learning_rate": 0.00014273832373640436,
|
1234 |
+
"loss": 0.0932,
|
1235 |
+
"step": 1790
|
1236 |
+
},
|
1237 |
+
{
|
1238 |
+
"epoch": 1.15,
|
1239 |
+
"learning_rate": 0.00014241842610364684,
|
1240 |
+
"loss": 0.1269,
|
1241 |
+
"step": 1800
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 1.15,
|
1245 |
+
"eval_accuracy": 0.9597,
|
1246 |
+
"eval_loss": 0.13903552293777466,
|
1247 |
+
"eval_runtime": 47.4785,
|
1248 |
+
"eval_samples_per_second": 210.622,
|
1249 |
+
"eval_steps_per_second": 13.164,
|
1250 |
+
"step": 1800
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 1.16,
|
1254 |
+
"learning_rate": 0.00014209852847088932,
|
1255 |
+
"loss": 0.0779,
|
1256 |
+
"step": 1810
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 1.16,
|
1260 |
+
"learning_rate": 0.0001417786308381318,
|
1261 |
+
"loss": 0.1292,
|
1262 |
+
"step": 1820
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 1.17,
|
1266 |
+
"learning_rate": 0.00014145873320537428,
|
1267 |
+
"loss": 0.0975,
|
1268 |
+
"step": 1830
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 1.18,
|
1272 |
+
"learning_rate": 0.00014113883557261678,
|
1273 |
+
"loss": 0.126,
|
1274 |
+
"step": 1840
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 1.18,
|
1278 |
+
"learning_rate": 0.00014081893793985926,
|
1279 |
+
"loss": 0.0781,
|
1280 |
+
"step": 1850
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 1.19,
|
1284 |
+
"learning_rate": 0.00014049904030710172,
|
1285 |
+
"loss": 0.0943,
|
1286 |
+
"step": 1860
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 1.2,
|
1290 |
+
"learning_rate": 0.00014017914267434422,
|
1291 |
+
"loss": 0.0936,
|
1292 |
+
"step": 1870
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 1.2,
|
1296 |
+
"learning_rate": 0.0001398592450415867,
|
1297 |
+
"loss": 0.086,
|
1298 |
+
"step": 1880
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 1.21,
|
1302 |
+
"learning_rate": 0.00013953934740882918,
|
1303 |
+
"loss": 0.1549,
|
1304 |
+
"step": 1890
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 1.22,
|
1308 |
+
"learning_rate": 0.00013921944977607166,
|
1309 |
+
"loss": 0.0635,
|
1310 |
+
"step": 1900
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 1.22,
|
1314 |
+
"eval_accuracy": 0.961,
|
1315 |
+
"eval_loss": 0.1290891319513321,
|
1316 |
+
"eval_runtime": 47.9741,
|
1317 |
+
"eval_samples_per_second": 208.446,
|
1318 |
+
"eval_steps_per_second": 13.028,
|
1319 |
+
"step": 1900
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 1.22,
|
1323 |
+
"learning_rate": 0.00013889955214331414,
|
1324 |
+
"loss": 0.0465,
|
1325 |
+
"step": 1910
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 1.23,
|
1329 |
+
"learning_rate": 0.00013857965451055662,
|
1330 |
+
"loss": 0.0874,
|
1331 |
+
"step": 1920
|
1332 |
+
},
|
1333 |
+
{
|
1334 |
+
"epoch": 1.23,
|
1335 |
+
"learning_rate": 0.0001382597568777991,
|
1336 |
+
"loss": 0.0738,
|
1337 |
+
"step": 1930
|
1338 |
+
},
|
1339 |
+
{
|
1340 |
+
"epoch": 1.24,
|
1341 |
+
"learning_rate": 0.0001379398592450416,
|
1342 |
+
"loss": 0.0777,
|
1343 |
+
"step": 1940
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 1.25,
|
1347 |
+
"learning_rate": 0.0001376199616122841,
|
1348 |
+
"loss": 0.0866,
|
1349 |
+
"step": 1950
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 1.25,
|
1353 |
+
"learning_rate": 0.00013730006397952654,
|
1354 |
+
"loss": 0.0895,
|
1355 |
+
"step": 1960
|
1356 |
+
},
|
1357 |
+
{
|
1358 |
+
"epoch": 1.26,
|
1359 |
+
"learning_rate": 0.00013698016634676905,
|
1360 |
+
"loss": 0.0758,
|
1361 |
+
"step": 1970
|
1362 |
+
},
|
1363 |
+
{
|
1364 |
+
"epoch": 1.27,
|
1365 |
+
"learning_rate": 0.00013666026871401153,
|
1366 |
+
"loss": 0.0421,
|
1367 |
+
"step": 1980
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 1.27,
|
1371 |
+
"learning_rate": 0.000136340371081254,
|
1372 |
+
"loss": 0.0586,
|
1373 |
+
"step": 1990
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 1.28,
|
1377 |
+
"learning_rate": 0.0001360204734484965,
|
1378 |
+
"loss": 0.1098,
|
1379 |
+
"step": 2000
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 1.28,
|
1383 |
+
"eval_accuracy": 0.9582,
|
1384 |
+
"eval_loss": 0.1469649374485016,
|
1385 |
+
"eval_runtime": 48.3383,
|
1386 |
+
"eval_samples_per_second": 206.875,
|
1387 |
+
"eval_steps_per_second": 12.93,
|
1388 |
+
"step": 2000
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 1.29,
|
1392 |
+
"learning_rate": 0.00013570057581573897,
|
1393 |
+
"loss": 0.0549,
|
1394 |
+
"step": 2010
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 1.29,
|
1398 |
+
"learning_rate": 0.00013538067818298145,
|
1399 |
+
"loss": 0.0379,
|
1400 |
+
"step": 2020
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 1.3,
|
1404 |
+
"learning_rate": 0.00013506078055022393,
|
1405 |
+
"loss": 0.1064,
|
1406 |
+
"step": 2030
|
1407 |
+
},
|
1408 |
+
{
|
1409 |
+
"epoch": 1.31,
|
1410 |
+
"learning_rate": 0.00013474088291746643,
|
1411 |
+
"loss": 0.0964,
|
1412 |
+
"step": 2040
|
1413 |
+
},
|
1414 |
+
{
|
1415 |
+
"epoch": 1.31,
|
1416 |
+
"learning_rate": 0.0001344209852847089,
|
1417 |
+
"loss": 0.0705,
|
1418 |
+
"step": 2050
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 1.32,
|
1422 |
+
"learning_rate": 0.00013410108765195137,
|
1423 |
+
"loss": 0.0691,
|
1424 |
+
"step": 2060
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 1.32,
|
1428 |
+
"learning_rate": 0.00013378119001919387,
|
1429 |
+
"loss": 0.0611,
|
1430 |
+
"step": 2070
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 1.33,
|
1434 |
+
"learning_rate": 0.00013346129238643635,
|
1435 |
+
"loss": 0.0718,
|
1436 |
+
"step": 2080
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 1.34,
|
1440 |
+
"learning_rate": 0.00013314139475367883,
|
1441 |
+
"loss": 0.0535,
|
1442 |
+
"step": 2090
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 1.34,
|
1446 |
+
"learning_rate": 0.0001328214971209213,
|
1447 |
+
"loss": 0.1065,
|
1448 |
+
"step": 2100
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 1.34,
|
1452 |
+
"eval_accuracy": 0.9598,
|
1453 |
+
"eval_loss": 0.13903473317623138,
|
1454 |
+
"eval_runtime": 47.7783,
|
1455 |
+
"eval_samples_per_second": 209.3,
|
1456 |
+
"eval_steps_per_second": 13.081,
|
1457 |
+
"step": 2100
|
1458 |
+
},
|
1459 |
+
{
|
1460 |
+
"epoch": 1.35,
|
1461 |
+
"learning_rate": 0.0001325015994881638,
|
1462 |
+
"loss": 0.0715,
|
1463 |
+
"step": 2110
|
1464 |
+
},
|
1465 |
+
{
|
1466 |
+
"epoch": 1.36,
|
1467 |
+
"learning_rate": 0.00013218170185540627,
|
1468 |
+
"loss": 0.0849,
|
1469 |
+
"step": 2120
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 1.36,
|
1473 |
+
"learning_rate": 0.00013186180422264875,
|
1474 |
+
"loss": 0.0926,
|
1475 |
+
"step": 2130
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 1.37,
|
1479 |
+
"learning_rate": 0.00013154190658989126,
|
1480 |
+
"loss": 0.1035,
|
1481 |
+
"step": 2140
|
1482 |
+
},
|
1483 |
+
{
|
1484 |
+
"epoch": 1.38,
|
1485 |
+
"learning_rate": 0.0001312220089571337,
|
1486 |
+
"loss": 0.0495,
|
1487 |
+
"step": 2150
|
1488 |
+
},
|
1489 |
+
{
|
1490 |
+
"epoch": 1.38,
|
1491 |
+
"learning_rate": 0.0001309021113243762,
|
1492 |
+
"loss": 0.0785,
|
1493 |
+
"step": 2160
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 1.39,
|
1497 |
+
"learning_rate": 0.0001305822136916187,
|
1498 |
+
"loss": 0.053,
|
1499 |
+
"step": 2170
|
1500 |
+
},
|
1501 |
+
{
|
1502 |
+
"epoch": 1.39,
|
1503 |
+
"learning_rate": 0.00013026231605886118,
|
1504 |
+
"loss": 0.0576,
|
1505 |
+
"step": 2180
|
1506 |
+
},
|
1507 |
+
{
|
1508 |
+
"epoch": 1.4,
|
1509 |
+
"learning_rate": 0.00012994241842610366,
|
1510 |
+
"loss": 0.0728,
|
1511 |
+
"step": 2190
|
1512 |
+
},
|
1513 |
+
{
|
1514 |
+
"epoch": 1.41,
|
1515 |
+
"learning_rate": 0.00012962252079334614,
|
1516 |
+
"loss": 0.0753,
|
1517 |
+
"step": 2200
|
1518 |
+
},
|
1519 |
+
{
|
1520 |
+
"epoch": 1.41,
|
1521 |
+
"eval_accuracy": 0.9651,
|
1522 |
+
"eval_loss": 0.12330373376607895,
|
1523 |
+
"eval_runtime": 49.6152,
|
1524 |
+
"eval_samples_per_second": 201.551,
|
1525 |
+
"eval_steps_per_second": 12.597,
|
1526 |
+
"step": 2200
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 1.41,
|
1530 |
+
"learning_rate": 0.00012930262316058862,
|
1531 |
+
"loss": 0.0441,
|
1532 |
+
"step": 2210
|
1533 |
+
},
|
1534 |
+
{
|
1535 |
+
"epoch": 1.42,
|
1536 |
+
"learning_rate": 0.0001289827255278311,
|
1537 |
+
"loss": 0.0526,
|
1538 |
+
"step": 2220
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 1.43,
|
1542 |
+
"learning_rate": 0.00012866282789507358,
|
1543 |
+
"loss": 0.082,
|
1544 |
+
"step": 2230
|
1545 |
+
},
|
1546 |
+
{
|
1547 |
+
"epoch": 1.43,
|
1548 |
+
"learning_rate": 0.00012834293026231608,
|
1549 |
+
"loss": 0.0734,
|
1550 |
+
"step": 2240
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 1.44,
|
1554 |
+
"learning_rate": 0.00012802303262955854,
|
1555 |
+
"loss": 0.0465,
|
1556 |
+
"step": 2250
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 1.45,
|
1560 |
+
"learning_rate": 0.00012770313499680102,
|
1561 |
+
"loss": 0.0841,
|
1562 |
+
"step": 2260
|
1563 |
+
},
|
1564 |
+
{
|
1565 |
+
"epoch": 1.45,
|
1566 |
+
"learning_rate": 0.00012738323736404352,
|
1567 |
+
"loss": 0.1249,
|
1568 |
+
"step": 2270
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 1.46,
|
1572 |
+
"learning_rate": 0.000127063339731286,
|
1573 |
+
"loss": 0.0852,
|
1574 |
+
"step": 2280
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 1.47,
|
1578 |
+
"learning_rate": 0.00012674344209852848,
|
1579 |
+
"loss": 0.1211,
|
1580 |
+
"step": 2290
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 1.47,
|
1584 |
+
"learning_rate": 0.00012642354446577096,
|
1585 |
+
"loss": 0.0855,
|
1586 |
+
"step": 2300
|
1587 |
+
},
|
1588 |
+
{
|
1589 |
+
"epoch": 1.47,
|
1590 |
+
"eval_accuracy": 0.958,
|
1591 |
+
"eval_loss": 0.14163978397846222,
|
1592 |
+
"eval_runtime": 48.1054,
|
1593 |
+
"eval_samples_per_second": 207.877,
|
1594 |
+
"eval_steps_per_second": 12.992,
|
1595 |
+
"step": 2300
|
1596 |
+
},
|
1597 |
+
{
|
1598 |
+
"epoch": 1.48,
|
1599 |
+
"learning_rate": 0.00012610364683301344,
|
1600 |
+
"loss": 0.0525,
|
1601 |
+
"step": 2310
|
1602 |
+
},
|
1603 |
+
{
|
1604 |
+
"epoch": 1.48,
|
1605 |
+
"learning_rate": 0.00012578374920025592,
|
1606 |
+
"loss": 0.1068,
|
1607 |
+
"step": 2320
|
1608 |
+
},
|
1609 |
+
{
|
1610 |
+
"epoch": 1.49,
|
1611 |
+
"learning_rate": 0.0001254638515674984,
|
1612 |
+
"loss": 0.0847,
|
1613 |
+
"step": 2330
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 1.5,
|
1617 |
+
"learning_rate": 0.0001251439539347409,
|
1618 |
+
"loss": 0.0945,
|
1619 |
+
"step": 2340
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 1.5,
|
1623 |
+
"learning_rate": 0.00012482405630198336,
|
1624 |
+
"loss": 0.108,
|
1625 |
+
"step": 2350
|
1626 |
+
},
|
1627 |
+
{
|
1628 |
+
"epoch": 1.51,
|
1629 |
+
"learning_rate": 0.00012450415866922584,
|
1630 |
+
"loss": 0.0529,
|
1631 |
+
"step": 2360
|
1632 |
+
},
|
1633 |
+
{
|
1634 |
+
"epoch": 1.52,
|
1635 |
+
"learning_rate": 0.00012418426103646835,
|
1636 |
+
"loss": 0.0679,
|
1637 |
+
"step": 2370
|
1638 |
+
},
|
1639 |
+
{
|
1640 |
+
"epoch": 1.52,
|
1641 |
+
"learning_rate": 0.0001238643634037108,
|
1642 |
+
"loss": 0.1067,
|
1643 |
+
"step": 2380
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 1.53,
|
1647 |
+
"learning_rate": 0.0001235444657709533,
|
1648 |
+
"loss": 0.0752,
|
1649 |
+
"step": 2390
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 1.54,
|
1653 |
+
"learning_rate": 0.0001232245681381958,
|
1654 |
+
"loss": 0.0723,
|
1655 |
+
"step": 2400
|
1656 |
+
},
|
1657 |
+
{
|
1658 |
+
"epoch": 1.54,
|
1659 |
+
"eval_accuracy": 0.9606,
|
1660 |
+
"eval_loss": 0.13063974678516388,
|
1661 |
+
"eval_runtime": 47.4226,
|
1662 |
+
"eval_samples_per_second": 210.87,
|
1663 |
+
"eval_steps_per_second": 13.179,
|
1664 |
+
"step": 2400
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 1.54,
|
1668 |
+
"learning_rate": 0.00012290467050543827,
|
1669 |
+
"loss": 0.0722,
|
1670 |
+
"step": 2410
|
1671 |
+
},
|
1672 |
+
{
|
1673 |
+
"epoch": 1.55,
|
1674 |
+
"learning_rate": 0.00012258477287268075,
|
1675 |
+
"loss": 0.1103,
|
1676 |
+
"step": 2420
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 1.55,
|
1680 |
+
"learning_rate": 0.00012226487523992323,
|
1681 |
+
"loss": 0.0846,
|
1682 |
+
"step": 2430
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 1.56,
|
1686 |
+
"learning_rate": 0.00012194497760716572,
|
1687 |
+
"loss": 0.1057,
|
1688 |
+
"step": 2440
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 1.57,
|
1692 |
+
"learning_rate": 0.0001216250799744082,
|
1693 |
+
"loss": 0.0528,
|
1694 |
+
"step": 2450
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 1.57,
|
1698 |
+
"learning_rate": 0.00012130518234165067,
|
1699 |
+
"loss": 0.0776,
|
1700 |
+
"step": 2460
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 1.58,
|
1704 |
+
"learning_rate": 0.00012098528470889316,
|
1705 |
+
"loss": 0.0833,
|
1706 |
+
"step": 2470
|
1707 |
+
},
|
1708 |
+
{
|
1709 |
+
"epoch": 1.59,
|
1710 |
+
"learning_rate": 0.00012066538707613564,
|
1711 |
+
"loss": 0.0716,
|
1712 |
+
"step": 2480
|
1713 |
+
},
|
1714 |
+
{
|
1715 |
+
"epoch": 1.59,
|
1716 |
+
"learning_rate": 0.00012034548944337813,
|
1717 |
+
"loss": 0.0524,
|
1718 |
+
"step": 2490
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 1.6,
|
1722 |
+
"learning_rate": 0.00012002559181062061,
|
1723 |
+
"loss": 0.0515,
|
1724 |
+
"step": 2500
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 1.6,
|
1728 |
+
"eval_accuracy": 0.9652,
|
1729 |
+
"eval_loss": 0.11919395625591278,
|
1730 |
+
"eval_runtime": 47.5032,
|
1731 |
+
"eval_samples_per_second": 210.512,
|
1732 |
+
"eval_steps_per_second": 13.157,
|
1733 |
+
"step": 2500
|
1734 |
+
},
|
1735 |
+
{
|
1736 |
+
"epoch": 1.61,
|
1737 |
+
"learning_rate": 0.00011970569417786308,
|
1738 |
+
"loss": 0.0542,
|
1739 |
+
"step": 2510
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 1.61,
|
1743 |
+
"learning_rate": 0.00011938579654510557,
|
1744 |
+
"loss": 0.0625,
|
1745 |
+
"step": 2520
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 1.62,
|
1749 |
+
"learning_rate": 0.00011906589891234805,
|
1750 |
+
"loss": 0.0598,
|
1751 |
+
"step": 2530
|
1752 |
+
},
|
1753 |
+
{
|
1754 |
+
"epoch": 1.63,
|
1755 |
+
"learning_rate": 0.00011874600127959055,
|
1756 |
+
"loss": 0.0668,
|
1757 |
+
"step": 2540
|
1758 |
+
},
|
1759 |
+
{
|
1760 |
+
"epoch": 1.63,
|
1761 |
+
"learning_rate": 0.00011842610364683303,
|
1762 |
+
"loss": 0.0539,
|
1763 |
+
"step": 2550
|
1764 |
+
},
|
1765 |
+
{
|
1766 |
+
"epoch": 1.64,
|
1767 |
+
"learning_rate": 0.00011810620601407549,
|
1768 |
+
"loss": 0.0635,
|
1769 |
+
"step": 2560
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 1.64,
|
1773 |
+
"learning_rate": 0.00011778630838131799,
|
1774 |
+
"loss": 0.0501,
|
1775 |
+
"step": 2570
|
1776 |
+
},
|
1777 |
+
{
|
1778 |
+
"epoch": 1.65,
|
1779 |
+
"learning_rate": 0.00011746641074856047,
|
1780 |
+
"loss": 0.0573,
|
1781 |
+
"step": 2580
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 1.66,
|
1785 |
+
"learning_rate": 0.00011714651311580296,
|
1786 |
+
"loss": 0.0435,
|
1787 |
+
"step": 2590
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 1.66,
|
1791 |
+
"learning_rate": 0.00011682661548304543,
|
1792 |
+
"loss": 0.0997,
|
1793 |
+
"step": 2600
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 1.66,
|
1797 |
+
"eval_accuracy": 0.9689,
|
1798 |
+
"eval_loss": 0.11294878274202347,
|
1799 |
+
"eval_runtime": 48.242,
|
1800 |
+
"eval_samples_per_second": 207.288,
|
1801 |
+
"eval_steps_per_second": 12.956,
|
1802 |
+
"step": 2600
|
1803 |
+
},
|
1804 |
+
{
|
1805 |
+
"epoch": 1.67,
|
1806 |
+
"learning_rate": 0.0001165067178502879,
|
1807 |
+
"loss": 0.1147,
|
1808 |
+
"step": 2610
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 1.68,
|
1812 |
+
"learning_rate": 0.0001161868202175304,
|
1813 |
+
"loss": 0.0868,
|
1814 |
+
"step": 2620
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 1.68,
|
1818 |
+
"learning_rate": 0.00011586692258477288,
|
1819 |
+
"loss": 0.1187,
|
1820 |
+
"step": 2630
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 1.69,
|
1824 |
+
"learning_rate": 0.00011554702495201537,
|
1825 |
+
"loss": 0.0814,
|
1826 |
+
"step": 2640
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 1.7,
|
1830 |
+
"learning_rate": 0.00011522712731925784,
|
1831 |
+
"loss": 0.0916,
|
1832 |
+
"step": 2650
|
1833 |
+
},
|
1834 |
+
{
|
1835 |
+
"epoch": 1.7,
|
1836 |
+
"learning_rate": 0.00011490722968650032,
|
1837 |
+
"loss": 0.0995,
|
1838 |
+
"step": 2660
|
1839 |
+
},
|
1840 |
+
{
|
1841 |
+
"epoch": 1.71,
|
1842 |
+
"learning_rate": 0.00011458733205374281,
|
1843 |
+
"loss": 0.0913,
|
1844 |
+
"step": 2670
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 1.71,
|
1848 |
+
"learning_rate": 0.00011426743442098529,
|
1849 |
+
"loss": 0.0493,
|
1850 |
+
"step": 2680
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 1.72,
|
1854 |
+
"learning_rate": 0.00011394753678822778,
|
1855 |
+
"loss": 0.0965,
|
1856 |
+
"step": 2690
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 1.73,
|
1860 |
+
"learning_rate": 0.00011362763915547025,
|
1861 |
+
"loss": 0.0958,
|
1862 |
+
"step": 2700
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 1.73,
|
1866 |
+
"eval_accuracy": 0.9622,
|
1867 |
+
"eval_loss": 0.12688662111759186,
|
1868 |
+
"eval_runtime": 47.1481,
|
1869 |
+
"eval_samples_per_second": 212.097,
|
1870 |
+
"eval_steps_per_second": 13.256,
|
1871 |
+
"step": 2700
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 1.73,
|
1875 |
+
"learning_rate": 0.00011330774152271273,
|
1876 |
+
"loss": 0.0561,
|
1877 |
+
"step": 2710
|
1878 |
+
},
|
1879 |
+
{
|
1880 |
+
"epoch": 1.74,
|
1881 |
+
"learning_rate": 0.00011298784388995522,
|
1882 |
+
"loss": 0.0533,
|
1883 |
+
"step": 2720
|
1884 |
+
},
|
1885 |
+
{
|
1886 |
+
"epoch": 1.75,
|
1887 |
+
"learning_rate": 0.0001126679462571977,
|
1888 |
+
"loss": 0.0738,
|
1889 |
+
"step": 2730
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 1.75,
|
1893 |
+
"learning_rate": 0.0001123480486244402,
|
1894 |
+
"loss": 0.0765,
|
1895 |
+
"step": 2740
|
1896 |
+
},
|
1897 |
+
{
|
1898 |
+
"epoch": 1.76,
|
1899 |
+
"learning_rate": 0.00011202815099168266,
|
1900 |
+
"loss": 0.087,
|
1901 |
+
"step": 2750
|
1902 |
+
},
|
1903 |
+
{
|
1904 |
+
"epoch": 1.77,
|
1905 |
+
"learning_rate": 0.00011170825335892514,
|
1906 |
+
"loss": 0.036,
|
1907 |
+
"step": 2760
|
1908 |
+
},
|
1909 |
+
{
|
1910 |
+
"epoch": 1.77,
|
1911 |
+
"learning_rate": 0.00011138835572616764,
|
1912 |
+
"loss": 0.0699,
|
1913 |
+
"step": 2770
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 1.78,
|
1917 |
+
"learning_rate": 0.00011106845809341012,
|
1918 |
+
"loss": 0.0565,
|
1919 |
+
"step": 2780
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 1.79,
|
1923 |
+
"learning_rate": 0.00011074856046065261,
|
1924 |
+
"loss": 0.0718,
|
1925 |
+
"step": 2790
|
1926 |
+
},
|
1927 |
+
{
|
1928 |
+
"epoch": 1.79,
|
1929 |
+
"learning_rate": 0.00011042866282789508,
|
1930 |
+
"loss": 0.0502,
|
1931 |
+
"step": 2800
|
1932 |
+
},
|
1933 |
+
{
|
1934 |
+
"epoch": 1.79,
|
1935 |
+
"eval_accuracy": 0.9641,
|
1936 |
+
"eval_loss": 0.12282514572143555,
|
1937 |
+
"eval_runtime": 48.2449,
|
1938 |
+
"eval_samples_per_second": 207.276,
|
1939 |
+
"eval_steps_per_second": 12.955,
|
1940 |
+
"step": 2800
|
1941 |
+
},
|
1942 |
+
{
|
1943 |
+
"epoch": 1.8,
|
1944 |
+
"learning_rate": 0.00011010876519513756,
|
1945 |
+
"loss": 0.0535,
|
1946 |
+
"step": 2810
|
1947 |
+
},
|
1948 |
+
{
|
1949 |
+
"epoch": 1.8,
|
1950 |
+
"learning_rate": 0.00010978886756238005,
|
1951 |
+
"loss": 0.0484,
|
1952 |
+
"step": 2820
|
1953 |
+
},
|
1954 |
+
{
|
1955 |
+
"epoch": 1.81,
|
1956 |
+
"learning_rate": 0.00010946896992962253,
|
1957 |
+
"loss": 0.0597,
|
1958 |
+
"step": 2830
|
1959 |
+
},
|
1960 |
+
{
|
1961 |
+
"epoch": 1.82,
|
1962 |
+
"learning_rate": 0.00010914907229686502,
|
1963 |
+
"loss": 0.0604,
|
1964 |
+
"step": 2840
|
1965 |
+
},
|
1966 |
+
{
|
1967 |
+
"epoch": 1.82,
|
1968 |
+
"learning_rate": 0.00010882917466410749,
|
1969 |
+
"loss": 0.0644,
|
1970 |
+
"step": 2850
|
1971 |
+
},
|
1972 |
+
{
|
1973 |
+
"epoch": 1.83,
|
1974 |
+
"learning_rate": 0.00010850927703134997,
|
1975 |
+
"loss": 0.0727,
|
1976 |
+
"step": 2860
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 1.84,
|
1980 |
+
"learning_rate": 0.00010818937939859246,
|
1981 |
+
"loss": 0.0565,
|
1982 |
+
"step": 2870
|
1983 |
+
},
|
1984 |
+
{
|
1985 |
+
"epoch": 1.84,
|
1986 |
+
"learning_rate": 0.00010786948176583493,
|
1987 |
+
"loss": 0.0717,
|
1988 |
+
"step": 2880
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 1.85,
|
1992 |
+
"learning_rate": 0.00010754958413307744,
|
1993 |
+
"loss": 0.0892,
|
1994 |
+
"step": 2890
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 1.86,
|
1998 |
+
"learning_rate": 0.0001072296865003199,
|
1999 |
+
"loss": 0.0379,
|
2000 |
+
"step": 2900
|
2001 |
+
},
|
2002 |
+
{
|
2003 |
+
"epoch": 1.86,
|
2004 |
+
"eval_accuracy": 0.9639,
|
2005 |
+
"eval_loss": 0.12743479013442993,
|
2006 |
+
"eval_runtime": 47.9702,
|
2007 |
+
"eval_samples_per_second": 208.463,
|
2008 |
+
"eval_steps_per_second": 13.029,
|
2009 |
+
"step": 2900
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 1.86,
|
2013 |
+
"learning_rate": 0.00010690978886756238,
|
2014 |
+
"loss": 0.065,
|
2015 |
+
"step": 2910
|
2016 |
+
},
|
2017 |
+
{
|
2018 |
+
"epoch": 1.87,
|
2019 |
+
"learning_rate": 0.00010658989123480487,
|
2020 |
+
"loss": 0.0668,
|
2021 |
+
"step": 2920
|
2022 |
+
},
|
2023 |
+
{
|
2024 |
+
"epoch": 1.87,
|
2025 |
+
"learning_rate": 0.00010626999360204734,
|
2026 |
+
"loss": 0.0243,
|
2027 |
+
"step": 2930
|
2028 |
+
},
|
2029 |
+
{
|
2030 |
+
"epoch": 1.88,
|
2031 |
+
"learning_rate": 0.00010595009596928985,
|
2032 |
+
"loss": 0.0314,
|
2033 |
+
"step": 2940
|
2034 |
+
},
|
2035 |
+
{
|
2036 |
+
"epoch": 1.89,
|
2037 |
+
"learning_rate": 0.00010563019833653231,
|
2038 |
+
"loss": 0.0629,
|
2039 |
+
"step": 2950
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 1.89,
|
2043 |
+
"learning_rate": 0.0001053103007037748,
|
2044 |
+
"loss": 0.0629,
|
2045 |
+
"step": 2960
|
2046 |
+
},
|
2047 |
+
{
|
2048 |
+
"epoch": 1.9,
|
2049 |
+
"learning_rate": 0.00010499040307101729,
|
2050 |
+
"loss": 0.0671,
|
2051 |
+
"step": 2970
|
2052 |
+
},
|
2053 |
+
{
|
2054 |
+
"epoch": 1.91,
|
2055 |
+
"learning_rate": 0.00010467050543825975,
|
2056 |
+
"loss": 0.0863,
|
2057 |
+
"step": 2980
|
2058 |
+
},
|
2059 |
+
{
|
2060 |
+
"epoch": 1.91,
|
2061 |
+
"learning_rate": 0.00010435060780550226,
|
2062 |
+
"loss": 0.1094,
|
2063 |
+
"step": 2990
|
2064 |
+
},
|
2065 |
+
{
|
2066 |
+
"epoch": 1.92,
|
2067 |
+
"learning_rate": 0.00010403071017274473,
|
2068 |
+
"loss": 0.0814,
|
2069 |
+
"step": 3000
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 1.92,
|
2073 |
+
"eval_accuracy": 0.963,
|
2074 |
+
"eval_loss": 0.12582720816135406,
|
2075 |
+
"eval_runtime": 47.6406,
|
2076 |
+
"eval_samples_per_second": 209.905,
|
2077 |
+
"eval_steps_per_second": 13.119,
|
2078 |
+
"step": 3000
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 1.93,
|
2082 |
+
"learning_rate": 0.0001037108125399872,
|
2083 |
+
"loss": 0.091,
|
2084 |
+
"step": 3010
|
2085 |
+
},
|
2086 |
+
{
|
2087 |
+
"epoch": 1.93,
|
2088 |
+
"learning_rate": 0.0001033909149072297,
|
2089 |
+
"loss": 0.0237,
|
2090 |
+
"step": 3020
|
2091 |
+
},
|
2092 |
+
{
|
2093 |
+
"epoch": 1.94,
|
2094 |
+
"learning_rate": 0.00010307101727447217,
|
2095 |
+
"loss": 0.0732,
|
2096 |
+
"step": 3030
|
2097 |
+
},
|
2098 |
+
{
|
2099 |
+
"epoch": 1.94,
|
2100 |
+
"learning_rate": 0.00010275111964171466,
|
2101 |
+
"loss": 0.0283,
|
2102 |
+
"step": 3040
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 1.95,
|
2106 |
+
"learning_rate": 0.00010243122200895714,
|
2107 |
+
"loss": 0.0972,
|
2108 |
+
"step": 3050
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 1.96,
|
2112 |
+
"learning_rate": 0.00010211132437619962,
|
2113 |
+
"loss": 0.0706,
|
2114 |
+
"step": 3060
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 1.96,
|
2118 |
+
"learning_rate": 0.00010179142674344211,
|
2119 |
+
"loss": 0.029,
|
2120 |
+
"step": 3070
|
2121 |
+
},
|
2122 |
+
{
|
2123 |
+
"epoch": 1.97,
|
2124 |
+
"learning_rate": 0.00010147152911068458,
|
2125 |
+
"loss": 0.0587,
|
2126 |
+
"step": 3080
|
2127 |
+
},
|
2128 |
+
{
|
2129 |
+
"epoch": 1.98,
|
2130 |
+
"learning_rate": 0.00010115163147792707,
|
2131 |
+
"loss": 0.1159,
|
2132 |
+
"step": 3090
|
2133 |
+
},
|
2134 |
+
{
|
2135 |
+
"epoch": 1.98,
|
2136 |
+
"learning_rate": 0.00010083173384516955,
|
2137 |
+
"loss": 0.0892,
|
2138 |
+
"step": 3100
|
2139 |
+
},
|
2140 |
+
{
|
2141 |
+
"epoch": 1.98,
|
2142 |
+
"eval_accuracy": 0.9652,
|
2143 |
+
"eval_loss": 0.11734022200107574,
|
2144 |
+
"eval_runtime": 47.9972,
|
2145 |
+
"eval_samples_per_second": 208.346,
|
2146 |
+
"eval_steps_per_second": 13.022,
|
2147 |
+
"step": 3100
|
2148 |
+
},
|
2149 |
+
{
|
2150 |
+
"epoch": 1.99,
|
2151 |
+
"learning_rate": 0.00010051183621241203,
|
2152 |
+
"loss": 0.0566,
|
2153 |
+
"step": 3110
|
2154 |
+
},
|
2155 |
+
{
|
2156 |
+
"epoch": 2.0,
|
2157 |
+
"learning_rate": 0.00010019193857965453,
|
2158 |
+
"loss": 0.0434,
|
2159 |
+
"step": 3120
|
2160 |
+
},
|
2161 |
+
{
|
2162 |
+
"epoch": 2.0,
|
2163 |
+
"learning_rate": 9.987204094689699e-05,
|
2164 |
+
"loss": 0.0197,
|
2165 |
+
"step": 3130
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 2.01,
|
2169 |
+
"learning_rate": 9.955214331413948e-05,
|
2170 |
+
"loss": 0.0411,
|
2171 |
+
"step": 3140
|
2172 |
+
},
|
2173 |
+
{
|
2174 |
+
"epoch": 2.02,
|
2175 |
+
"learning_rate": 9.923224568138196e-05,
|
2176 |
+
"loss": 0.0278,
|
2177 |
+
"step": 3150
|
2178 |
+
},
|
2179 |
+
{
|
2180 |
+
"epoch": 2.02,
|
2181 |
+
"learning_rate": 9.891234804862444e-05,
|
2182 |
+
"loss": 0.0242,
|
2183 |
+
"step": 3160
|
2184 |
+
},
|
2185 |
+
{
|
2186 |
+
"epoch": 2.03,
|
2187 |
+
"learning_rate": 9.859245041586694e-05,
|
2188 |
+
"loss": 0.0324,
|
2189 |
+
"step": 3170
|
2190 |
+
},
|
2191 |
+
{
|
2192 |
+
"epoch": 2.03,
|
2193 |
+
"learning_rate": 9.82725527831094e-05,
|
2194 |
+
"loss": 0.0368,
|
2195 |
+
"step": 3180
|
2196 |
+
},
|
2197 |
+
{
|
2198 |
+
"epoch": 2.04,
|
2199 |
+
"learning_rate": 9.79526551503519e-05,
|
2200 |
+
"loss": 0.0174,
|
2201 |
+
"step": 3190
|
2202 |
+
},
|
2203 |
+
{
|
2204 |
+
"epoch": 2.05,
|
2205 |
+
"learning_rate": 9.763275751759438e-05,
|
2206 |
+
"loss": 0.0126,
|
2207 |
+
"step": 3200
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 2.05,
|
2211 |
+
"eval_accuracy": 0.9662,
|
2212 |
+
"eval_loss": 0.11732649058103561,
|
2213 |
+
"eval_runtime": 48.648,
|
2214 |
+
"eval_samples_per_second": 205.558,
|
2215 |
+
"eval_steps_per_second": 12.847,
|
2216 |
+
"step": 3200
|
2217 |
+
},
|
2218 |
+
{
|
2219 |
+
"epoch": 2.05,
|
2220 |
+
"learning_rate": 9.731285988483686e-05,
|
2221 |
+
"loss": 0.0398,
|
2222 |
+
"step": 3210
|
2223 |
+
},
|
2224 |
+
{
|
2225 |
+
"epoch": 2.06,
|
2226 |
+
"learning_rate": 9.699296225207935e-05,
|
2227 |
+
"loss": 0.0233,
|
2228 |
+
"step": 3220
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 2.07,
|
2232 |
+
"learning_rate": 9.667306461932182e-05,
|
2233 |
+
"loss": 0.0243,
|
2234 |
+
"step": 3230
|
2235 |
+
},
|
2236 |
+
{
|
2237 |
+
"epoch": 2.07,
|
2238 |
+
"learning_rate": 9.63531669865643e-05,
|
2239 |
+
"loss": 0.0247,
|
2240 |
+
"step": 3240
|
2241 |
+
},
|
2242 |
+
{
|
2243 |
+
"epoch": 2.08,
|
2244 |
+
"learning_rate": 9.603326935380679e-05,
|
2245 |
+
"loss": 0.0153,
|
2246 |
+
"step": 3250
|
2247 |
+
},
|
2248 |
+
{
|
2249 |
+
"epoch": 2.09,
|
2250 |
+
"learning_rate": 9.571337172104927e-05,
|
2251 |
+
"loss": 0.0199,
|
2252 |
+
"step": 3260
|
2253 |
+
},
|
2254 |
+
{
|
2255 |
+
"epoch": 2.09,
|
2256 |
+
"learning_rate": 9.539347408829176e-05,
|
2257 |
+
"loss": 0.0257,
|
2258 |
+
"step": 3270
|
2259 |
+
},
|
2260 |
+
{
|
2261 |
+
"epoch": 2.1,
|
2262 |
+
"learning_rate": 9.507357645553423e-05,
|
2263 |
+
"loss": 0.0221,
|
2264 |
+
"step": 3280
|
2265 |
+
},
|
2266 |
+
{
|
2267 |
+
"epoch": 2.1,
|
2268 |
+
"learning_rate": 9.475367882277671e-05,
|
2269 |
+
"loss": 0.0118,
|
2270 |
+
"step": 3290
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 2.11,
|
2274 |
+
"learning_rate": 9.44337811900192e-05,
|
2275 |
+
"loss": 0.04,
|
2276 |
+
"step": 3300
|
2277 |
+
},
|
2278 |
+
{
|
2279 |
+
"epoch": 2.11,
|
2280 |
+
"eval_accuracy": 0.972,
|
2281 |
+
"eval_loss": 0.10201091319322586,
|
2282 |
+
"eval_runtime": 47.4962,
|
2283 |
+
"eval_samples_per_second": 210.543,
|
2284 |
+
"eval_steps_per_second": 13.159,
|
2285 |
+
"step": 3300
|
2286 |
+
},
|
2287 |
+
{
|
2288 |
+
"epoch": 2.12,
|
2289 |
+
"learning_rate": 9.411388355726168e-05,
|
2290 |
+
"loss": 0.0027,
|
2291 |
+
"step": 3310
|
2292 |
+
},
|
2293 |
+
{
|
2294 |
+
"epoch": 2.12,
|
2295 |
+
"learning_rate": 9.379398592450416e-05,
|
2296 |
+
"loss": 0.0182,
|
2297 |
+
"step": 3320
|
2298 |
+
},
|
2299 |
+
{
|
2300 |
+
"epoch": 2.13,
|
2301 |
+
"learning_rate": 9.347408829174664e-05,
|
2302 |
+
"loss": 0.011,
|
2303 |
+
"step": 3330
|
2304 |
+
},
|
2305 |
+
{
|
2306 |
+
"epoch": 2.14,
|
2307 |
+
"learning_rate": 9.315419065898912e-05,
|
2308 |
+
"loss": 0.0502,
|
2309 |
+
"step": 3340
|
2310 |
+
},
|
2311 |
+
{
|
2312 |
+
"epoch": 2.14,
|
2313 |
+
"learning_rate": 9.283429302623161e-05,
|
2314 |
+
"loss": 0.0529,
|
2315 |
+
"step": 3350
|
2316 |
+
},
|
2317 |
+
{
|
2318 |
+
"epoch": 2.15,
|
2319 |
+
"learning_rate": 9.25143953934741e-05,
|
2320 |
+
"loss": 0.0202,
|
2321 |
+
"step": 3360
|
2322 |
+
},
|
2323 |
+
{
|
2324 |
+
"epoch": 2.16,
|
2325 |
+
"learning_rate": 9.219449776071657e-05,
|
2326 |
+
"loss": 0.0138,
|
2327 |
+
"step": 3370
|
2328 |
+
},
|
2329 |
+
{
|
2330 |
+
"epoch": 2.16,
|
2331 |
+
"learning_rate": 9.187460012795905e-05,
|
2332 |
+
"loss": 0.0037,
|
2333 |
+
"step": 3380
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 2.17,
|
2337 |
+
"learning_rate": 9.155470249520153e-05,
|
2338 |
+
"loss": 0.0033,
|
2339 |
+
"step": 3390
|
2340 |
+
},
|
2341 |
+
{
|
2342 |
+
"epoch": 2.18,
|
2343 |
+
"learning_rate": 9.123480486244403e-05,
|
2344 |
+
"loss": 0.0098,
|
2345 |
+
"step": 3400
|
2346 |
+
},
|
2347 |
+
{
|
2348 |
+
"epoch": 2.18,
|
2349 |
+
"eval_accuracy": 0.9731,
|
2350 |
+
"eval_loss": 0.10650923103094101,
|
2351 |
+
"eval_runtime": 47.4681,
|
2352 |
+
"eval_samples_per_second": 210.668,
|
2353 |
+
"eval_steps_per_second": 13.167,
|
2354 |
+
"step": 3400
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 2.18,
|
2358 |
+
"learning_rate": 9.091490722968651e-05,
|
2359 |
+
"loss": 0.0056,
|
2360 |
+
"step": 3410
|
2361 |
+
},
|
2362 |
+
{
|
2363 |
+
"epoch": 2.19,
|
2364 |
+
"learning_rate": 9.059500959692899e-05,
|
2365 |
+
"loss": 0.0289,
|
2366 |
+
"step": 3420
|
2367 |
+
},
|
2368 |
+
{
|
2369 |
+
"epoch": 2.19,
|
2370 |
+
"learning_rate": 9.027511196417147e-05,
|
2371 |
+
"loss": 0.0659,
|
2372 |
+
"step": 3430
|
2373 |
+
},
|
2374 |
+
{
|
2375 |
+
"epoch": 2.2,
|
2376 |
+
"learning_rate": 8.995521433141395e-05,
|
2377 |
+
"loss": 0.0388,
|
2378 |
+
"step": 3440
|
2379 |
+
},
|
2380 |
+
{
|
2381 |
+
"epoch": 2.21,
|
2382 |
+
"learning_rate": 8.963531669865644e-05,
|
2383 |
+
"loss": 0.0206,
|
2384 |
+
"step": 3450
|
2385 |
+
},
|
2386 |
+
{
|
2387 |
+
"epoch": 2.21,
|
2388 |
+
"learning_rate": 8.931541906589892e-05,
|
2389 |
+
"loss": 0.0202,
|
2390 |
+
"step": 3460
|
2391 |
+
},
|
2392 |
+
{
|
2393 |
+
"epoch": 2.22,
|
2394 |
+
"learning_rate": 8.89955214331414e-05,
|
2395 |
+
"loss": 0.0128,
|
2396 |
+
"step": 3470
|
2397 |
+
},
|
2398 |
+
{
|
2399 |
+
"epoch": 2.23,
|
2400 |
+
"learning_rate": 8.867562380038388e-05,
|
2401 |
+
"loss": 0.0367,
|
2402 |
+
"step": 3480
|
2403 |
+
},
|
2404 |
+
{
|
2405 |
+
"epoch": 2.23,
|
2406 |
+
"learning_rate": 8.835572616762636e-05,
|
2407 |
+
"loss": 0.0132,
|
2408 |
+
"step": 3490
|
2409 |
+
},
|
2410 |
+
{
|
2411 |
+
"epoch": 2.24,
|
2412 |
+
"learning_rate": 8.803582853486885e-05,
|
2413 |
+
"loss": 0.0159,
|
2414 |
+
"step": 3500
|
2415 |
+
},
|
2416 |
+
{
|
2417 |
+
"epoch": 2.24,
|
2418 |
+
"eval_accuracy": 0.9741,
|
2419 |
+
"eval_loss": 0.1040922999382019,
|
2420 |
+
"eval_runtime": 50.3231,
|
2421 |
+
"eval_samples_per_second": 198.716,
|
2422 |
+
"eval_steps_per_second": 12.42,
|
2423 |
+
"step": 3500
|
2424 |
+
},
|
2425 |
+
{
|
2426 |
+
"epoch": 2.25,
|
2427 |
+
"learning_rate": 8.771593090211133e-05,
|
2428 |
+
"loss": 0.0254,
|
2429 |
+
"step": 3510
|
2430 |
+
},
|
2431 |
+
{
|
2432 |
+
"epoch": 2.25,
|
2433 |
+
"learning_rate": 8.739603326935381e-05,
|
2434 |
+
"loss": 0.0333,
|
2435 |
+
"step": 3520
|
2436 |
+
},
|
2437 |
+
{
|
2438 |
+
"epoch": 2.26,
|
2439 |
+
"learning_rate": 8.707613563659629e-05,
|
2440 |
+
"loss": 0.0099,
|
2441 |
+
"step": 3530
|
2442 |
+
},
|
2443 |
+
{
|
2444 |
+
"epoch": 2.26,
|
2445 |
+
"learning_rate": 8.675623800383877e-05,
|
2446 |
+
"loss": 0.0211,
|
2447 |
+
"step": 3540
|
2448 |
+
},
|
2449 |
+
{
|
2450 |
+
"epoch": 2.27,
|
2451 |
+
"learning_rate": 8.643634037108127e-05,
|
2452 |
+
"loss": 0.0201,
|
2453 |
+
"step": 3550
|
2454 |
+
},
|
2455 |
+
{
|
2456 |
+
"epoch": 2.28,
|
2457 |
+
"learning_rate": 8.611644273832375e-05,
|
2458 |
+
"loss": 0.0223,
|
2459 |
+
"step": 3560
|
2460 |
+
},
|
2461 |
+
{
|
2462 |
+
"epoch": 2.28,
|
2463 |
+
"learning_rate": 8.579654510556623e-05,
|
2464 |
+
"loss": 0.0263,
|
2465 |
+
"step": 3570
|
2466 |
+
},
|
2467 |
+
{
|
2468 |
+
"epoch": 2.29,
|
2469 |
+
"learning_rate": 8.54766474728087e-05,
|
2470 |
+
"loss": 0.0067,
|
2471 |
+
"step": 3580
|
2472 |
+
},
|
2473 |
+
{
|
2474 |
+
"epoch": 2.3,
|
2475 |
+
"learning_rate": 8.515674984005118e-05,
|
2476 |
+
"loss": 0.008,
|
2477 |
+
"step": 3590
|
2478 |
+
},
|
2479 |
+
{
|
2480 |
+
"epoch": 2.3,
|
2481 |
+
"learning_rate": 8.483685220729366e-05,
|
2482 |
+
"loss": 0.0247,
|
2483 |
+
"step": 3600
|
2484 |
+
},
|
2485 |
+
{
|
2486 |
+
"epoch": 2.3,
|
2487 |
+
"eval_accuracy": 0.9716,
|
2488 |
+
"eval_loss": 0.10976359993219376,
|
2489 |
+
"eval_runtime": 47.3179,
|
2490 |
+
"eval_samples_per_second": 211.337,
|
2491 |
+
"eval_steps_per_second": 13.209,
|
2492 |
+
"step": 3600
|
2493 |
+
},
|
2494 |
+
{
|
2495 |
+
"epoch": 2.31,
|
2496 |
+
"learning_rate": 8.451695457453616e-05,
|
2497 |
+
"loss": 0.0691,
|
2498 |
+
"step": 3610
|
2499 |
+
},
|
2500 |
+
{
|
2501 |
+
"epoch": 2.32,
|
2502 |
+
"learning_rate": 8.419705694177864e-05,
|
2503 |
+
"loss": 0.0153,
|
2504 |
+
"step": 3620
|
2505 |
+
},
|
2506 |
+
{
|
2507 |
+
"epoch": 2.32,
|
2508 |
+
"learning_rate": 8.387715930902112e-05,
|
2509 |
+
"loss": 0.0115,
|
2510 |
+
"step": 3630
|
2511 |
+
},
|
2512 |
+
{
|
2513 |
+
"epoch": 2.33,
|
2514 |
+
"learning_rate": 8.35572616762636e-05,
|
2515 |
+
"loss": 0.0267,
|
2516 |
+
"step": 3640
|
2517 |
+
},
|
2518 |
+
{
|
2519 |
+
"epoch": 2.34,
|
2520 |
+
"learning_rate": 8.323736404350608e-05,
|
2521 |
+
"loss": 0.0102,
|
2522 |
+
"step": 3650
|
2523 |
+
},
|
2524 |
+
{
|
2525 |
+
"epoch": 2.34,
|
2526 |
+
"learning_rate": 8.291746641074857e-05,
|
2527 |
+
"loss": 0.0121,
|
2528 |
+
"step": 3660
|
2529 |
+
},
|
2530 |
+
{
|
2531 |
+
"epoch": 2.35,
|
2532 |
+
"learning_rate": 8.259756877799105e-05,
|
2533 |
+
"loss": 0.0255,
|
2534 |
+
"step": 3670
|
2535 |
+
},
|
2536 |
+
{
|
2537 |
+
"epoch": 2.35,
|
2538 |
+
"learning_rate": 8.227767114523353e-05,
|
2539 |
+
"loss": 0.0246,
|
2540 |
+
"step": 3680
|
2541 |
+
},
|
2542 |
+
{
|
2543 |
+
"epoch": 2.36,
|
2544 |
+
"learning_rate": 8.195777351247601e-05,
|
2545 |
+
"loss": 0.0574,
|
2546 |
+
"step": 3690
|
2547 |
+
},
|
2548 |
+
{
|
2549 |
+
"epoch": 2.37,
|
2550 |
+
"learning_rate": 8.163787587971849e-05,
|
2551 |
+
"loss": 0.0457,
|
2552 |
+
"step": 3700
|
2553 |
+
},
|
2554 |
+
{
|
2555 |
+
"epoch": 2.37,
|
2556 |
+
"eval_accuracy": 0.972,
|
2557 |
+
"eval_loss": 0.10533589124679565,
|
2558 |
+
"eval_runtime": 48.4465,
|
2559 |
+
"eval_samples_per_second": 206.413,
|
2560 |
+
"eval_steps_per_second": 12.901,
|
2561 |
+
"step": 3700
|
2562 |
+
},
|
2563 |
+
{
|
2564 |
+
"epoch": 2.37,
|
2565 |
+
"learning_rate": 8.131797824696098e-05,
|
2566 |
+
"loss": 0.0185,
|
2567 |
+
"step": 3710
|
2568 |
+
},
|
2569 |
+
{
|
2570 |
+
"epoch": 2.38,
|
2571 |
+
"learning_rate": 8.099808061420346e-05,
|
2572 |
+
"loss": 0.0066,
|
2573 |
+
"step": 3720
|
2574 |
+
},
|
2575 |
+
{
|
2576 |
+
"epoch": 2.39,
|
2577 |
+
"learning_rate": 8.067818298144594e-05,
|
2578 |
+
"loss": 0.0036,
|
2579 |
+
"step": 3730
|
2580 |
+
},
|
2581 |
+
{
|
2582 |
+
"epoch": 2.39,
|
2583 |
+
"learning_rate": 8.035828534868842e-05,
|
2584 |
+
"loss": 0.012,
|
2585 |
+
"step": 3740
|
2586 |
+
},
|
2587 |
+
{
|
2588 |
+
"epoch": 2.4,
|
2589 |
+
"learning_rate": 8.00383877159309e-05,
|
2590 |
+
"loss": 0.009,
|
2591 |
+
"step": 3750
|
2592 |
+
},
|
2593 |
+
{
|
2594 |
+
"epoch": 2.41,
|
2595 |
+
"learning_rate": 7.97184900831734e-05,
|
2596 |
+
"loss": 0.0113,
|
2597 |
+
"step": 3760
|
2598 |
+
},
|
2599 |
+
{
|
2600 |
+
"epoch": 2.41,
|
2601 |
+
"learning_rate": 7.939859245041588e-05,
|
2602 |
+
"loss": 0.0248,
|
2603 |
+
"step": 3770
|
2604 |
+
},
|
2605 |
+
{
|
2606 |
+
"epoch": 2.42,
|
2607 |
+
"learning_rate": 7.907869481765836e-05,
|
2608 |
+
"loss": 0.0051,
|
2609 |
+
"step": 3780
|
2610 |
+
},
|
2611 |
+
{
|
2612 |
+
"epoch": 2.42,
|
2613 |
+
"learning_rate": 7.875879718490084e-05,
|
2614 |
+
"loss": 0.0392,
|
2615 |
+
"step": 3790
|
2616 |
+
},
|
2617 |
+
{
|
2618 |
+
"epoch": 2.43,
|
2619 |
+
"learning_rate": 7.843889955214332e-05,
|
2620 |
+
"loss": 0.0621,
|
2621 |
+
"step": 3800
|
2622 |
+
},
|
2623 |
+
{
|
2624 |
+
"epoch": 2.43,
|
2625 |
+
"eval_accuracy": 0.9761,
|
2626 |
+
"eval_loss": 0.09739825129508972,
|
2627 |
+
"eval_runtime": 47.7784,
|
2628 |
+
"eval_samples_per_second": 209.3,
|
2629 |
+
"eval_steps_per_second": 13.081,
|
2630 |
+
"step": 3800
|
2631 |
+
},
|
2632 |
+
{
|
2633 |
+
"epoch": 2.44,
|
2634 |
+
"learning_rate": 7.811900191938581e-05,
|
2635 |
+
"loss": 0.0091,
|
2636 |
+
"step": 3810
|
2637 |
+
},
|
2638 |
+
{
|
2639 |
+
"epoch": 2.44,
|
2640 |
+
"learning_rate": 7.779910428662829e-05,
|
2641 |
+
"loss": 0.0228,
|
2642 |
+
"step": 3820
|
2643 |
+
},
|
2644 |
+
{
|
2645 |
+
"epoch": 2.45,
|
2646 |
+
"learning_rate": 7.747920665387077e-05,
|
2647 |
+
"loss": 0.0359,
|
2648 |
+
"step": 3830
|
2649 |
+
},
|
2650 |
+
{
|
2651 |
+
"epoch": 2.46,
|
2652 |
+
"learning_rate": 7.715930902111325e-05,
|
2653 |
+
"loss": 0.0575,
|
2654 |
+
"step": 3840
|
2655 |
+
},
|
2656 |
+
{
|
2657 |
+
"epoch": 2.46,
|
2658 |
+
"learning_rate": 7.683941138835573e-05,
|
2659 |
+
"loss": 0.0705,
|
2660 |
+
"step": 3850
|
2661 |
+
},
|
2662 |
+
{
|
2663 |
+
"epoch": 2.47,
|
2664 |
+
"learning_rate": 7.651951375559822e-05,
|
2665 |
+
"loss": 0.0266,
|
2666 |
+
"step": 3860
|
2667 |
+
},
|
2668 |
+
{
|
2669 |
+
"epoch": 2.48,
|
2670 |
+
"learning_rate": 7.61996161228407e-05,
|
2671 |
+
"loss": 0.025,
|
2672 |
+
"step": 3870
|
2673 |
+
},
|
2674 |
+
{
|
2675 |
+
"epoch": 2.48,
|
2676 |
+
"learning_rate": 7.587971849008317e-05,
|
2677 |
+
"loss": 0.0691,
|
2678 |
+
"step": 3880
|
2679 |
+
},
|
2680 |
+
{
|
2681 |
+
"epoch": 2.49,
|
2682 |
+
"learning_rate": 7.555982085732566e-05,
|
2683 |
+
"loss": 0.0026,
|
2684 |
+
"step": 3890
|
2685 |
+
},
|
2686 |
+
{
|
2687 |
+
"epoch": 2.5,
|
2688 |
+
"learning_rate": 7.523992322456814e-05,
|
2689 |
+
"loss": 0.0591,
|
2690 |
+
"step": 3900
|
2691 |
+
},
|
2692 |
+
{
|
2693 |
+
"epoch": 2.5,
|
2694 |
+
"eval_accuracy": 0.9713,
|
2695 |
+
"eval_loss": 0.10392673313617706,
|
2696 |
+
"eval_runtime": 47.9322,
|
2697 |
+
"eval_samples_per_second": 208.628,
|
2698 |
+
"eval_steps_per_second": 13.039,
|
2699 |
+
"step": 3900
|
2700 |
+
},
|
2701 |
+
{
|
2702 |
+
"epoch": 2.5,
|
2703 |
+
"learning_rate": 7.492002559181063e-05,
|
2704 |
+
"loss": 0.0062,
|
2705 |
+
"step": 3910
|
2706 |
+
},
|
2707 |
+
{
|
2708 |
+
"epoch": 2.51,
|
2709 |
+
"learning_rate": 7.46001279590531e-05,
|
2710 |
+
"loss": 0.0496,
|
2711 |
+
"step": 3920
|
2712 |
+
},
|
2713 |
+
{
|
2714 |
+
"epoch": 2.51,
|
2715 |
+
"learning_rate": 7.428023032629558e-05,
|
2716 |
+
"loss": 0.0039,
|
2717 |
+
"step": 3930
|
2718 |
+
},
|
2719 |
+
{
|
2720 |
+
"epoch": 2.52,
|
2721 |
+
"learning_rate": 7.396033269353807e-05,
|
2722 |
+
"loss": 0.0136,
|
2723 |
+
"step": 3940
|
2724 |
+
},
|
2725 |
+
{
|
2726 |
+
"epoch": 2.53,
|
2727 |
+
"learning_rate": 7.364043506078055e-05,
|
2728 |
+
"loss": 0.0372,
|
2729 |
+
"step": 3950
|
2730 |
+
},
|
2731 |
+
{
|
2732 |
+
"epoch": 2.53,
|
2733 |
+
"learning_rate": 7.332053742802303e-05,
|
2734 |
+
"loss": 0.0134,
|
2735 |
+
"step": 3960
|
2736 |
+
},
|
2737 |
+
{
|
2738 |
+
"epoch": 2.54,
|
2739 |
+
"learning_rate": 7.300063979526551e-05,
|
2740 |
+
"loss": 0.0085,
|
2741 |
+
"step": 3970
|
2742 |
+
},
|
2743 |
+
{
|
2744 |
+
"epoch": 2.55,
|
2745 |
+
"learning_rate": 7.268074216250799e-05,
|
2746 |
+
"loss": 0.0161,
|
2747 |
+
"step": 3980
|
2748 |
+
},
|
2749 |
+
{
|
2750 |
+
"epoch": 2.55,
|
2751 |
+
"learning_rate": 7.236084452975049e-05,
|
2752 |
+
"loss": 0.0126,
|
2753 |
+
"step": 3990
|
2754 |
+
},
|
2755 |
+
{
|
2756 |
+
"epoch": 2.56,
|
2757 |
+
"learning_rate": 7.204094689699297e-05,
|
2758 |
+
"loss": 0.0624,
|
2759 |
+
"step": 4000
|
2760 |
+
},
|
2761 |
+
{
|
2762 |
+
"epoch": 2.56,
|
2763 |
+
"eval_accuracy": 0.9729,
|
2764 |
+
"eval_loss": 0.11228275299072266,
|
2765 |
+
"eval_runtime": 48.1823,
|
2766 |
+
"eval_samples_per_second": 207.545,
|
2767 |
+
"eval_steps_per_second": 12.972,
|
2768 |
+
"step": 4000
|
2769 |
+
},
|
2770 |
+
{
|
2771 |
+
"epoch": 2.57,
|
2772 |
+
"learning_rate": 7.172104926423545e-05,
|
2773 |
+
"loss": 0.0075,
|
2774 |
+
"step": 4010
|
2775 |
+
},
|
2776 |
+
{
|
2777 |
+
"epoch": 2.57,
|
2778 |
+
"learning_rate": 7.140115163147793e-05,
|
2779 |
+
"loss": 0.0191,
|
2780 |
+
"step": 4020
|
2781 |
+
},
|
2782 |
+
{
|
2783 |
+
"epoch": 2.58,
|
2784 |
+
"learning_rate": 7.10812539987204e-05,
|
2785 |
+
"loss": 0.0201,
|
2786 |
+
"step": 4030
|
2787 |
+
},
|
2788 |
+
{
|
2789 |
+
"epoch": 2.58,
|
2790 |
+
"learning_rate": 7.07613563659629e-05,
|
2791 |
+
"loss": 0.0092,
|
2792 |
+
"step": 4040
|
2793 |
+
},
|
2794 |
+
{
|
2795 |
+
"epoch": 2.59,
|
2796 |
+
"learning_rate": 7.044145873320538e-05,
|
2797 |
+
"loss": 0.0299,
|
2798 |
+
"step": 4050
|
2799 |
+
},
|
2800 |
+
{
|
2801 |
+
"epoch": 2.6,
|
2802 |
+
"learning_rate": 7.012156110044786e-05,
|
2803 |
+
"loss": 0.0233,
|
2804 |
+
"step": 4060
|
2805 |
+
},
|
2806 |
+
{
|
2807 |
+
"epoch": 2.6,
|
2808 |
+
"learning_rate": 6.980166346769034e-05,
|
2809 |
+
"loss": 0.0054,
|
2810 |
+
"step": 4070
|
2811 |
+
},
|
2812 |
+
{
|
2813 |
+
"epoch": 2.61,
|
2814 |
+
"learning_rate": 6.948176583493282e-05,
|
2815 |
+
"loss": 0.0243,
|
2816 |
+
"step": 4080
|
2817 |
+
},
|
2818 |
+
{
|
2819 |
+
"epoch": 2.62,
|
2820 |
+
"learning_rate": 6.916186820217531e-05,
|
2821 |
+
"loss": 0.0286,
|
2822 |
+
"step": 4090
|
2823 |
+
},
|
2824 |
+
{
|
2825 |
+
"epoch": 2.62,
|
2826 |
+
"learning_rate": 6.884197056941779e-05,
|
2827 |
+
"loss": 0.0309,
|
2828 |
+
"step": 4100
|
2829 |
+
},
|
2830 |
+
{
|
2831 |
+
"epoch": 2.62,
|
2832 |
+
"eval_accuracy": 0.9751,
|
2833 |
+
"eval_loss": 0.09711159765720367,
|
2834 |
+
"eval_runtime": 46.9031,
|
2835 |
+
"eval_samples_per_second": 213.206,
|
2836 |
+
"eval_steps_per_second": 13.325,
|
2837 |
+
"step": 4100
|
2838 |
+
},
|
2839 |
+
{
|
2840 |
+
"epoch": 2.63,
|
2841 |
+
"learning_rate": 6.852207293666027e-05,
|
2842 |
+
"loss": 0.0211,
|
2843 |
+
"step": 4110
|
2844 |
+
},
|
2845 |
+
{
|
2846 |
+
"epoch": 2.64,
|
2847 |
+
"learning_rate": 6.820217530390275e-05,
|
2848 |
+
"loss": 0.0168,
|
2849 |
+
"step": 4120
|
2850 |
+
},
|
2851 |
+
{
|
2852 |
+
"epoch": 2.64,
|
2853 |
+
"learning_rate": 6.788227767114523e-05,
|
2854 |
+
"loss": 0.0195,
|
2855 |
+
"step": 4130
|
2856 |
+
},
|
2857 |
+
{
|
2858 |
+
"epoch": 2.65,
|
2859 |
+
"learning_rate": 6.756238003838772e-05,
|
2860 |
+
"loss": 0.0072,
|
2861 |
+
"step": 4140
|
2862 |
+
},
|
2863 |
+
{
|
2864 |
+
"epoch": 2.66,
|
2865 |
+
"learning_rate": 6.72424824056302e-05,
|
2866 |
+
"loss": 0.0243,
|
2867 |
+
"step": 4150
|
2868 |
+
},
|
2869 |
+
{
|
2870 |
+
"epoch": 2.66,
|
2871 |
+
"learning_rate": 6.692258477287268e-05,
|
2872 |
+
"loss": 0.027,
|
2873 |
+
"step": 4160
|
2874 |
+
},
|
2875 |
+
{
|
2876 |
+
"epoch": 2.67,
|
2877 |
+
"learning_rate": 6.660268714011516e-05,
|
2878 |
+
"loss": 0.039,
|
2879 |
+
"step": 4170
|
2880 |
+
},
|
2881 |
+
{
|
2882 |
+
"epoch": 2.67,
|
2883 |
+
"learning_rate": 6.628278950735764e-05,
|
2884 |
+
"loss": 0.017,
|
2885 |
+
"step": 4180
|
2886 |
+
},
|
2887 |
+
{
|
2888 |
+
"epoch": 2.68,
|
2889 |
+
"learning_rate": 6.596289187460014e-05,
|
2890 |
+
"loss": 0.008,
|
2891 |
+
"step": 4190
|
2892 |
+
},
|
2893 |
+
{
|
2894 |
+
"epoch": 2.69,
|
2895 |
+
"learning_rate": 6.564299424184262e-05,
|
2896 |
+
"loss": 0.015,
|
2897 |
+
"step": 4200
|
2898 |
+
},
|
2899 |
+
{
|
2900 |
+
"epoch": 2.69,
|
2901 |
+
"eval_accuracy": 0.9738,
|
2902 |
+
"eval_loss": 0.104751355946064,
|
2903 |
+
"eval_runtime": 48.1137,
|
2904 |
+
"eval_samples_per_second": 207.841,
|
2905 |
+
"eval_steps_per_second": 12.99,
|
2906 |
+
"step": 4200
|
2907 |
+
},
|
2908 |
+
{
|
2909 |
+
"epoch": 2.69,
|
2910 |
+
"learning_rate": 6.53230966090851e-05,
|
2911 |
+
"loss": 0.0041,
|
2912 |
+
"step": 4210
|
2913 |
+
},
|
2914 |
+
{
|
2915 |
+
"epoch": 2.7,
|
2916 |
+
"learning_rate": 6.500319897632758e-05,
|
2917 |
+
"loss": 0.004,
|
2918 |
+
"step": 4220
|
2919 |
+
},
|
2920 |
+
{
|
2921 |
+
"epoch": 2.71,
|
2922 |
+
"learning_rate": 6.468330134357006e-05,
|
2923 |
+
"loss": 0.0206,
|
2924 |
+
"step": 4230
|
2925 |
+
},
|
2926 |
+
{
|
2927 |
+
"epoch": 2.71,
|
2928 |
+
"learning_rate": 6.436340371081254e-05,
|
2929 |
+
"loss": 0.012,
|
2930 |
+
"step": 4240
|
2931 |
+
},
|
2932 |
+
{
|
2933 |
+
"epoch": 2.72,
|
2934 |
+
"learning_rate": 6.404350607805503e-05,
|
2935 |
+
"loss": 0.0303,
|
2936 |
+
"step": 4250
|
2937 |
+
},
|
2938 |
+
{
|
2939 |
+
"epoch": 2.73,
|
2940 |
+
"learning_rate": 6.372360844529751e-05,
|
2941 |
+
"loss": 0.039,
|
2942 |
+
"step": 4260
|
2943 |
+
},
|
2944 |
+
{
|
2945 |
+
"epoch": 2.73,
|
2946 |
+
"learning_rate": 6.340371081253999e-05,
|
2947 |
+
"loss": 0.0225,
|
2948 |
+
"step": 4270
|
2949 |
+
},
|
2950 |
+
{
|
2951 |
+
"epoch": 2.74,
|
2952 |
+
"learning_rate": 6.308381317978247e-05,
|
2953 |
+
"loss": 0.019,
|
2954 |
+
"step": 4280
|
2955 |
+
},
|
2956 |
+
{
|
2957 |
+
"epoch": 2.74,
|
2958 |
+
"learning_rate": 6.276391554702495e-05,
|
2959 |
+
"loss": 0.0014,
|
2960 |
+
"step": 4290
|
2961 |
+
},
|
2962 |
+
{
|
2963 |
+
"epoch": 2.75,
|
2964 |
+
"learning_rate": 6.244401791426744e-05,
|
2965 |
+
"loss": 0.0155,
|
2966 |
+
"step": 4300
|
2967 |
+
},
|
2968 |
+
{
|
2969 |
+
"epoch": 2.75,
|
2970 |
+
"eval_accuracy": 0.9749,
|
2971 |
+
"eval_loss": 0.0932016596198082,
|
2972 |
+
"eval_runtime": 61.5553,
|
2973 |
+
"eval_samples_per_second": 162.456,
|
2974 |
+
"eval_steps_per_second": 10.153,
|
2975 |
+
"step": 4300
|
2976 |
+
},
|
2977 |
+
{
|
2978 |
+
"epoch": 2.76,
|
2979 |
+
"learning_rate": 6.212412028150992e-05,
|
2980 |
+
"loss": 0.0027,
|
2981 |
+
"step": 4310
|
2982 |
+
},
|
2983 |
+
{
|
2984 |
+
"epoch": 2.76,
|
2985 |
+
"learning_rate": 6.18042226487524e-05,
|
2986 |
+
"loss": 0.0181,
|
2987 |
+
"step": 4320
|
2988 |
+
},
|
2989 |
+
{
|
2990 |
+
"epoch": 2.77,
|
2991 |
+
"learning_rate": 6.148432501599488e-05,
|
2992 |
+
"loss": 0.0063,
|
2993 |
+
"step": 4330
|
2994 |
+
},
|
2995 |
+
{
|
2996 |
+
"epoch": 2.78,
|
2997 |
+
"learning_rate": 6.116442738323736e-05,
|
2998 |
+
"loss": 0.0188,
|
2999 |
+
"step": 4340
|
3000 |
+
},
|
3001 |
+
{
|
3002 |
+
"epoch": 2.78,
|
3003 |
+
"learning_rate": 6.084452975047985e-05,
|
3004 |
+
"loss": 0.0195,
|
3005 |
+
"step": 4350
|
3006 |
+
},
|
3007 |
+
{
|
3008 |
+
"epoch": 2.79,
|
3009 |
+
"learning_rate": 6.0524632117722334e-05,
|
3010 |
+
"loss": 0.0057,
|
3011 |
+
"step": 4360
|
3012 |
+
},
|
3013 |
+
{
|
3014 |
+
"epoch": 2.8,
|
3015 |
+
"learning_rate": 6.020473448496481e-05,
|
3016 |
+
"loss": 0.0059,
|
3017 |
+
"step": 4370
|
3018 |
+
},
|
3019 |
+
{
|
3020 |
+
"epoch": 2.8,
|
3021 |
+
"learning_rate": 5.9884836852207293e-05,
|
3022 |
+
"loss": 0.0339,
|
3023 |
+
"step": 4380
|
3024 |
+
},
|
3025 |
+
{
|
3026 |
+
"epoch": 2.81,
|
3027 |
+
"learning_rate": 5.956493921944978e-05,
|
3028 |
+
"loss": 0.0381,
|
3029 |
+
"step": 4390
|
3030 |
+
},
|
3031 |
+
{
|
3032 |
+
"epoch": 2.82,
|
3033 |
+
"learning_rate": 5.924504158669226e-05,
|
3034 |
+
"loss": 0.0269,
|
3035 |
+
"step": 4400
|
3036 |
+
},
|
3037 |
+
{
|
3038 |
+
"epoch": 2.82,
|
3039 |
+
"eval_accuracy": 0.9702,
|
3040 |
+
"eval_loss": 0.11733580380678177,
|
3041 |
+
"eval_runtime": 61.9528,
|
3042 |
+
"eval_samples_per_second": 161.413,
|
3043 |
+
"eval_steps_per_second": 10.088,
|
3044 |
+
"step": 4400
|
3045 |
+
},
|
3046 |
+
{
|
3047 |
+
"epoch": 2.82,
|
3048 |
+
"learning_rate": 5.8925143953934746e-05,
|
3049 |
+
"loss": 0.0158,
|
3050 |
+
"step": 4410
|
3051 |
+
},
|
3052 |
+
{
|
3053 |
+
"epoch": 2.83,
|
3054 |
+
"learning_rate": 5.860524632117722e-05,
|
3055 |
+
"loss": 0.0214,
|
3056 |
+
"step": 4420
|
3057 |
+
},
|
3058 |
+
{
|
3059 |
+
"epoch": 2.83,
|
3060 |
+
"learning_rate": 5.8285348688419706e-05,
|
3061 |
+
"loss": 0.0484,
|
3062 |
+
"step": 4430
|
3063 |
+
},
|
3064 |
+
{
|
3065 |
+
"epoch": 2.84,
|
3066 |
+
"learning_rate": 5.796545105566219e-05,
|
3067 |
+
"loss": 0.0152,
|
3068 |
+
"step": 4440
|
3069 |
+
},
|
3070 |
+
{
|
3071 |
+
"epoch": 2.85,
|
3072 |
+
"learning_rate": 5.764555342290467e-05,
|
3073 |
+
"loss": 0.1147,
|
3074 |
+
"step": 4450
|
3075 |
+
},
|
3076 |
+
{
|
3077 |
+
"epoch": 2.85,
|
3078 |
+
"learning_rate": 5.732565579014716e-05,
|
3079 |
+
"loss": 0.0292,
|
3080 |
+
"step": 4460
|
3081 |
+
},
|
3082 |
+
{
|
3083 |
+
"epoch": 2.86,
|
3084 |
+
"learning_rate": 5.700575815738963e-05,
|
3085 |
+
"loss": 0.052,
|
3086 |
+
"step": 4470
|
3087 |
+
},
|
3088 |
+
{
|
3089 |
+
"epoch": 2.87,
|
3090 |
+
"learning_rate": 5.668586052463212e-05,
|
3091 |
+
"loss": 0.0055,
|
3092 |
+
"step": 4480
|
3093 |
+
},
|
3094 |
+
{
|
3095 |
+
"epoch": 2.87,
|
3096 |
+
"learning_rate": 5.63659628918746e-05,
|
3097 |
+
"loss": 0.0053,
|
3098 |
+
"step": 4490
|
3099 |
+
},
|
3100 |
+
{
|
3101 |
+
"epoch": 2.88,
|
3102 |
+
"learning_rate": 5.6046065259117085e-05,
|
3103 |
+
"loss": 0.0021,
|
3104 |
+
"step": 4500
|
3105 |
+
},
|
3106 |
+
{
|
3107 |
+
"epoch": 2.88,
|
3108 |
+
"eval_accuracy": 0.974,
|
3109 |
+
"eval_loss": 0.0957212820649147,
|
3110 |
+
"eval_runtime": 62.9349,
|
3111 |
+
"eval_samples_per_second": 158.894,
|
3112 |
+
"eval_steps_per_second": 9.931,
|
3113 |
+
"step": 4500
|
3114 |
+
},
|
3115 |
+
{
|
3116 |
+
"epoch": 2.89,
|
3117 |
+
"learning_rate": 5.572616762635957e-05,
|
3118 |
+
"loss": 0.043,
|
3119 |
+
"step": 4510
|
3120 |
+
},
|
3121 |
+
{
|
3122 |
+
"epoch": 2.89,
|
3123 |
+
"learning_rate": 5.5406269993602045e-05,
|
3124 |
+
"loss": 0.0207,
|
3125 |
+
"step": 4520
|
3126 |
+
},
|
3127 |
+
{
|
3128 |
+
"epoch": 2.9,
|
3129 |
+
"learning_rate": 5.508637236084453e-05,
|
3130 |
+
"loss": 0.006,
|
3131 |
+
"step": 4530
|
3132 |
+
},
|
3133 |
+
{
|
3134 |
+
"epoch": 2.9,
|
3135 |
+
"learning_rate": 5.476647472808701e-05,
|
3136 |
+
"loss": 0.0043,
|
3137 |
+
"step": 4540
|
3138 |
+
},
|
3139 |
+
{
|
3140 |
+
"epoch": 2.91,
|
3141 |
+
"learning_rate": 5.44465770953295e-05,
|
3142 |
+
"loss": 0.0257,
|
3143 |
+
"step": 4550
|
3144 |
+
},
|
3145 |
+
{
|
3146 |
+
"epoch": 2.92,
|
3147 |
+
"learning_rate": 5.4126679462571984e-05,
|
3148 |
+
"loss": 0.0241,
|
3149 |
+
"step": 4560
|
3150 |
+
},
|
3151 |
+
{
|
3152 |
+
"epoch": 2.92,
|
3153 |
+
"learning_rate": 5.380678182981446e-05,
|
3154 |
+
"loss": 0.019,
|
3155 |
+
"step": 4570
|
3156 |
+
},
|
3157 |
+
{
|
3158 |
+
"epoch": 2.93,
|
3159 |
+
"learning_rate": 5.3486884197056944e-05,
|
3160 |
+
"loss": 0.0204,
|
3161 |
+
"step": 4580
|
3162 |
+
},
|
3163 |
+
{
|
3164 |
+
"epoch": 2.94,
|
3165 |
+
"learning_rate": 5.3166986564299424e-05,
|
3166 |
+
"loss": 0.0182,
|
3167 |
+
"step": 4590
|
3168 |
+
},
|
3169 |
+
{
|
3170 |
+
"epoch": 2.94,
|
3171 |
+
"learning_rate": 5.284708893154191e-05,
|
3172 |
+
"loss": 0.0238,
|
3173 |
+
"step": 4600
|
3174 |
+
},
|
3175 |
+
{
|
3176 |
+
"epoch": 2.94,
|
3177 |
+
"eval_accuracy": 0.9742,
|
3178 |
+
"eval_loss": 0.09226856380701065,
|
3179 |
+
"eval_runtime": 62.7299,
|
3180 |
+
"eval_samples_per_second": 159.414,
|
3181 |
+
"eval_steps_per_second": 9.963,
|
3182 |
+
"step": 4600
|
3183 |
+
},
|
3184 |
+
{
|
3185 |
+
"epoch": 2.95,
|
3186 |
+
"learning_rate": 5.25271912987844e-05,
|
3187 |
+
"loss": 0.0036,
|
3188 |
+
"step": 4610
|
3189 |
+
},
|
3190 |
+
{
|
3191 |
+
"epoch": 2.96,
|
3192 |
+
"learning_rate": 5.220729366602687e-05,
|
3193 |
+
"loss": 0.0672,
|
3194 |
+
"step": 4620
|
3195 |
+
},
|
3196 |
+
{
|
3197 |
+
"epoch": 2.96,
|
3198 |
+
"learning_rate": 5.188739603326935e-05,
|
3199 |
+
"loss": 0.0383,
|
3200 |
+
"step": 4630
|
3201 |
+
},
|
3202 |
+
{
|
3203 |
+
"epoch": 2.97,
|
3204 |
+
"learning_rate": 5.1567498400511836e-05,
|
3205 |
+
"loss": 0.0044,
|
3206 |
+
"step": 4640
|
3207 |
+
},
|
3208 |
+
{
|
3209 |
+
"epoch": 2.98,
|
3210 |
+
"learning_rate": 5.124760076775432e-05,
|
3211 |
+
"loss": 0.0257,
|
3212 |
+
"step": 4650
|
3213 |
+
},
|
3214 |
+
{
|
3215 |
+
"epoch": 2.98,
|
3216 |
+
"learning_rate": 5.092770313499681e-05,
|
3217 |
+
"loss": 0.0011,
|
3218 |
+
"step": 4660
|
3219 |
+
},
|
3220 |
+
{
|
3221 |
+
"epoch": 2.99,
|
3222 |
+
"learning_rate": 5.060780550223928e-05,
|
3223 |
+
"loss": 0.0034,
|
3224 |
+
"step": 4670
|
3225 |
+
},
|
3226 |
+
{
|
3227 |
+
"epoch": 2.99,
|
3228 |
+
"learning_rate": 5.028790786948176e-05,
|
3229 |
+
"loss": 0.0256,
|
3230 |
+
"step": 4680
|
3231 |
+
},
|
3232 |
+
{
|
3233 |
+
"epoch": 3.0,
|
3234 |
+
"learning_rate": 4.996801023672425e-05,
|
3235 |
+
"loss": 0.0129,
|
3236 |
+
"step": 4690
|
3237 |
+
},
|
3238 |
+
{
|
3239 |
+
"epoch": 3.01,
|
3240 |
+
"learning_rate": 4.9648112603966736e-05,
|
3241 |
+
"loss": 0.0015,
|
3242 |
+
"step": 4700
|
3243 |
+
},
|
3244 |
+
{
|
3245 |
+
"epoch": 3.01,
|
3246 |
+
"eval_accuracy": 0.9748,
|
3247 |
+
"eval_loss": 0.09875385463237762,
|
3248 |
+
"eval_runtime": 62.4215,
|
3249 |
+
"eval_samples_per_second": 160.201,
|
3250 |
+
"eval_steps_per_second": 10.013,
|
3251 |
+
"step": 4700
|
3252 |
+
},
|
3253 |
+
{
|
3254 |
+
"epoch": 3.01,
|
3255 |
+
"learning_rate": 4.9328214971209215e-05,
|
3256 |
+
"loss": 0.0154,
|
3257 |
+
"step": 4710
|
3258 |
+
},
|
3259 |
+
{
|
3260 |
+
"epoch": 3.02,
|
3261 |
+
"learning_rate": 4.9008317338451695e-05,
|
3262 |
+
"loss": 0.0012,
|
3263 |
+
"step": 4720
|
3264 |
+
},
|
3265 |
+
{
|
3266 |
+
"epoch": 3.03,
|
3267 |
+
"learning_rate": 4.868841970569418e-05,
|
3268 |
+
"loss": 0.001,
|
3269 |
+
"step": 4730
|
3270 |
+
},
|
3271 |
+
{
|
3272 |
+
"epoch": 3.03,
|
3273 |
+
"learning_rate": 4.836852207293666e-05,
|
3274 |
+
"loss": 0.0069,
|
3275 |
+
"step": 4740
|
3276 |
+
},
|
3277 |
+
{
|
3278 |
+
"epoch": 3.04,
|
3279 |
+
"learning_rate": 4.804862444017915e-05,
|
3280 |
+
"loss": 0.0035,
|
3281 |
+
"step": 4750
|
3282 |
+
},
|
3283 |
+
{
|
3284 |
+
"epoch": 3.05,
|
3285 |
+
"learning_rate": 4.772872680742163e-05,
|
3286 |
+
"loss": 0.0024,
|
3287 |
+
"step": 4760
|
3288 |
+
},
|
3289 |
+
{
|
3290 |
+
"epoch": 3.05,
|
3291 |
+
"learning_rate": 4.740882917466411e-05,
|
3292 |
+
"loss": 0.0026,
|
3293 |
+
"step": 4770
|
3294 |
+
},
|
3295 |
+
{
|
3296 |
+
"epoch": 3.06,
|
3297 |
+
"learning_rate": 4.7088931541906594e-05,
|
3298 |
+
"loss": 0.0014,
|
3299 |
+
"step": 4780
|
3300 |
+
},
|
3301 |
+
{
|
3302 |
+
"epoch": 3.06,
|
3303 |
+
"learning_rate": 4.6769033909149074e-05,
|
3304 |
+
"loss": 0.0008,
|
3305 |
+
"step": 4790
|
3306 |
+
},
|
3307 |
+
{
|
3308 |
+
"epoch": 3.07,
|
3309 |
+
"learning_rate": 4.644913627639156e-05,
|
3310 |
+
"loss": 0.0016,
|
3311 |
+
"step": 4800
|
3312 |
+
},
|
3313 |
+
{
|
3314 |
+
"epoch": 3.07,
|
3315 |
+
"eval_accuracy": 0.9791,
|
3316 |
+
"eval_loss": 0.08298031985759735,
|
3317 |
+
"eval_runtime": 61.1981,
|
3318 |
+
"eval_samples_per_second": 163.404,
|
3319 |
+
"eval_steps_per_second": 10.213,
|
3320 |
+
"step": 4800
|
3321 |
+
},
|
3322 |
+
{
|
3323 |
+
"epoch": 3.08,
|
3324 |
+
"learning_rate": 4.612923864363404e-05,
|
3325 |
+
"loss": 0.0024,
|
3326 |
+
"step": 4810
|
3327 |
+
},
|
3328 |
+
{
|
3329 |
+
"epoch": 3.08,
|
3330 |
+
"learning_rate": 4.580934101087652e-05,
|
3331 |
+
"loss": 0.0015,
|
3332 |
+
"step": 4820
|
3333 |
+
},
|
3334 |
+
{
|
3335 |
+
"epoch": 3.09,
|
3336 |
+
"learning_rate": 4.548944337811901e-05,
|
3337 |
+
"loss": 0.0007,
|
3338 |
+
"step": 4830
|
3339 |
+
},
|
3340 |
+
{
|
3341 |
+
"epoch": 3.1,
|
3342 |
+
"learning_rate": 4.516954574536149e-05,
|
3343 |
+
"loss": 0.0008,
|
3344 |
+
"step": 4840
|
3345 |
+
},
|
3346 |
+
{
|
3347 |
+
"epoch": 3.1,
|
3348 |
+
"learning_rate": 4.4849648112603973e-05,
|
3349 |
+
"loss": 0.0139,
|
3350 |
+
"step": 4850
|
3351 |
+
},
|
3352 |
+
{
|
3353 |
+
"epoch": 3.11,
|
3354 |
+
"learning_rate": 4.4529750479846447e-05,
|
3355 |
+
"loss": 0.0019,
|
3356 |
+
"step": 4860
|
3357 |
+
},
|
3358 |
+
{
|
3359 |
+
"epoch": 3.12,
|
3360 |
+
"learning_rate": 4.420985284708893e-05,
|
3361 |
+
"loss": 0.0011,
|
3362 |
+
"step": 4870
|
3363 |
+
},
|
3364 |
+
{
|
3365 |
+
"epoch": 3.12,
|
3366 |
+
"learning_rate": 4.388995521433142e-05,
|
3367 |
+
"loss": 0.001,
|
3368 |
+
"step": 4880
|
3369 |
+
},
|
3370 |
+
{
|
3371 |
+
"epoch": 3.13,
|
3372 |
+
"learning_rate": 4.35700575815739e-05,
|
3373 |
+
"loss": 0.0009,
|
3374 |
+
"step": 4890
|
3375 |
+
},
|
3376 |
+
{
|
3377 |
+
"epoch": 3.13,
|
3378 |
+
"learning_rate": 4.325015994881638e-05,
|
3379 |
+
"loss": 0.001,
|
3380 |
+
"step": 4900
|
3381 |
+
},
|
3382 |
+
{
|
3383 |
+
"epoch": 3.13,
|
3384 |
+
"eval_accuracy": 0.9795,
|
3385 |
+
"eval_loss": 0.08240580558776855,
|
3386 |
+
"eval_runtime": 61.0908,
|
3387 |
+
"eval_samples_per_second": 163.691,
|
3388 |
+
"eval_steps_per_second": 10.231,
|
3389 |
+
"step": 4900
|
3390 |
+
},
|
3391 |
+
{
|
3392 |
+
"epoch": 3.14,
|
3393 |
+
"learning_rate": 4.293026231605886e-05,
|
3394 |
+
"loss": 0.001,
|
3395 |
+
"step": 4910
|
3396 |
+
},
|
3397 |
+
{
|
3398 |
+
"epoch": 3.15,
|
3399 |
+
"learning_rate": 4.2610364683301346e-05,
|
3400 |
+
"loss": 0.0007,
|
3401 |
+
"step": 4920
|
3402 |
+
},
|
3403 |
+
{
|
3404 |
+
"epoch": 3.15,
|
3405 |
+
"learning_rate": 4.229046705054383e-05,
|
3406 |
+
"loss": 0.0011,
|
3407 |
+
"step": 4930
|
3408 |
+
},
|
3409 |
+
{
|
3410 |
+
"epoch": 3.16,
|
3411 |
+
"learning_rate": 4.197056941778631e-05,
|
3412 |
+
"loss": 0.0008,
|
3413 |
+
"step": 4940
|
3414 |
+
},
|
3415 |
+
{
|
3416 |
+
"epoch": 3.17,
|
3417 |
+
"learning_rate": 4.165067178502879e-05,
|
3418 |
+
"loss": 0.0013,
|
3419 |
+
"step": 4950
|
3420 |
+
},
|
3421 |
+
{
|
3422 |
+
"epoch": 3.17,
|
3423 |
+
"learning_rate": 4.133077415227127e-05,
|
3424 |
+
"loss": 0.0035,
|
3425 |
+
"step": 4960
|
3426 |
+
},
|
3427 |
+
{
|
3428 |
+
"epoch": 3.18,
|
3429 |
+
"learning_rate": 4.101087651951376e-05,
|
3430 |
+
"loss": 0.0008,
|
3431 |
+
"step": 4970
|
3432 |
+
},
|
3433 |
+
{
|
3434 |
+
"epoch": 3.19,
|
3435 |
+
"learning_rate": 4.0690978886756245e-05,
|
3436 |
+
"loss": 0.0035,
|
3437 |
+
"step": 4980
|
3438 |
+
},
|
3439 |
+
{
|
3440 |
+
"epoch": 3.19,
|
3441 |
+
"learning_rate": 4.0371081253998725e-05,
|
3442 |
+
"loss": 0.001,
|
3443 |
+
"step": 4990
|
3444 |
+
},
|
3445 |
+
{
|
3446 |
+
"epoch": 3.2,
|
3447 |
+
"learning_rate": 4.0051183621241205e-05,
|
3448 |
+
"loss": 0.0008,
|
3449 |
+
"step": 5000
|
3450 |
+
},
|
3451 |
+
{
|
3452 |
+
"epoch": 3.2,
|
3453 |
+
"eval_accuracy": 0.976,
|
3454 |
+
"eval_loss": 0.0969667136669159,
|
3455 |
+
"eval_runtime": 61.996,
|
3456 |
+
"eval_samples_per_second": 161.301,
|
3457 |
+
"eval_steps_per_second": 10.081,
|
3458 |
+
"step": 5000
|
3459 |
+
},
|
3460 |
+
{
|
3461 |
+
"epoch": 3.21,
|
3462 |
+
"learning_rate": 3.9731285988483684e-05,
|
3463 |
+
"loss": 0.0133,
|
3464 |
+
"step": 5010
|
3465 |
+
},
|
3466 |
+
{
|
3467 |
+
"epoch": 3.21,
|
3468 |
+
"learning_rate": 3.941138835572617e-05,
|
3469 |
+
"loss": 0.0008,
|
3470 |
+
"step": 5020
|
3471 |
+
},
|
3472 |
+
{
|
3473 |
+
"epoch": 3.22,
|
3474 |
+
"learning_rate": 3.909149072296866e-05,
|
3475 |
+
"loss": 0.0037,
|
3476 |
+
"step": 5030
|
3477 |
+
},
|
3478 |
+
{
|
3479 |
+
"epoch": 3.22,
|
3480 |
+
"learning_rate": 3.877159309021113e-05,
|
3481 |
+
"loss": 0.0007,
|
3482 |
+
"step": 5040
|
3483 |
+
},
|
3484 |
+
{
|
3485 |
+
"epoch": 3.23,
|
3486 |
+
"learning_rate": 3.845169545745362e-05,
|
3487 |
+
"loss": 0.0015,
|
3488 |
+
"step": 5050
|
3489 |
+
},
|
3490 |
+
{
|
3491 |
+
"epoch": 3.24,
|
3492 |
+
"learning_rate": 3.81317978246961e-05,
|
3493 |
+
"loss": 0.0008,
|
3494 |
+
"step": 5060
|
3495 |
+
},
|
3496 |
+
{
|
3497 |
+
"epoch": 3.24,
|
3498 |
+
"learning_rate": 3.7811900191938584e-05,
|
3499 |
+
"loss": 0.0014,
|
3500 |
+
"step": 5070
|
3501 |
+
},
|
3502 |
+
{
|
3503 |
+
"epoch": 3.25,
|
3504 |
+
"learning_rate": 3.7492002559181063e-05,
|
3505 |
+
"loss": 0.0007,
|
3506 |
+
"step": 5080
|
3507 |
+
},
|
3508 |
+
{
|
3509 |
+
"epoch": 3.26,
|
3510 |
+
"learning_rate": 3.717210492642354e-05,
|
3511 |
+
"loss": 0.0012,
|
3512 |
+
"step": 5090
|
3513 |
+
},
|
3514 |
+
{
|
3515 |
+
"epoch": 3.26,
|
3516 |
+
"learning_rate": 3.685220729366603e-05,
|
3517 |
+
"loss": 0.0149,
|
3518 |
+
"step": 5100
|
3519 |
+
},
|
3520 |
+
{
|
3521 |
+
"epoch": 3.26,
|
3522 |
+
"eval_accuracy": 0.9796,
|
3523 |
+
"eval_loss": 0.0864199697971344,
|
3524 |
+
"eval_runtime": 61.7022,
|
3525 |
+
"eval_samples_per_second": 162.069,
|
3526 |
+
"eval_steps_per_second": 10.129,
|
3527 |
+
"step": 5100
|
3528 |
+
},
|
3529 |
+
{
|
3530 |
+
"epoch": 3.27,
|
3531 |
+
"learning_rate": 3.653230966090851e-05,
|
3532 |
+
"loss": 0.001,
|
3533 |
+
"step": 5110
|
3534 |
+
},
|
3535 |
+
{
|
3536 |
+
"epoch": 3.28,
|
3537 |
+
"learning_rate": 3.6212412028150996e-05,
|
3538 |
+
"loss": 0.001,
|
3539 |
+
"step": 5120
|
3540 |
+
},
|
3541 |
+
{
|
3542 |
+
"epoch": 3.28,
|
3543 |
+
"learning_rate": 3.5892514395393476e-05,
|
3544 |
+
"loss": 0.0015,
|
3545 |
+
"step": 5130
|
3546 |
+
},
|
3547 |
+
{
|
3548 |
+
"epoch": 3.29,
|
3549 |
+
"learning_rate": 3.5572616762635956e-05,
|
3550 |
+
"loss": 0.0018,
|
3551 |
+
"step": 5140
|
3552 |
+
},
|
3553 |
+
{
|
3554 |
+
"epoch": 3.29,
|
3555 |
+
"learning_rate": 3.525271912987844e-05,
|
3556 |
+
"loss": 0.002,
|
3557 |
+
"step": 5150
|
3558 |
+
},
|
3559 |
+
{
|
3560 |
+
"epoch": 3.3,
|
3561 |
+
"learning_rate": 3.493282149712092e-05,
|
3562 |
+
"loss": 0.0019,
|
3563 |
+
"step": 5160
|
3564 |
+
},
|
3565 |
+
{
|
3566 |
+
"epoch": 3.31,
|
3567 |
+
"learning_rate": 3.461292386436341e-05,
|
3568 |
+
"loss": 0.0063,
|
3569 |
+
"step": 5170
|
3570 |
+
},
|
3571 |
+
{
|
3572 |
+
"epoch": 3.31,
|
3573 |
+
"learning_rate": 3.429302623160589e-05,
|
3574 |
+
"loss": 0.0008,
|
3575 |
+
"step": 5180
|
3576 |
+
},
|
3577 |
+
{
|
3578 |
+
"epoch": 3.32,
|
3579 |
+
"learning_rate": 3.397312859884837e-05,
|
3580 |
+
"loss": 0.0273,
|
3581 |
+
"step": 5190
|
3582 |
+
},
|
3583 |
+
{
|
3584 |
+
"epoch": 3.33,
|
3585 |
+
"learning_rate": 3.3653230966090855e-05,
|
3586 |
+
"loss": 0.008,
|
3587 |
+
"step": 5200
|
3588 |
+
},
|
3589 |
+
{
|
3590 |
+
"epoch": 3.33,
|
3591 |
+
"eval_accuracy": 0.9778,
|
3592 |
+
"eval_loss": 0.09051262587308884,
|
3593 |
+
"eval_runtime": 62.6123,
|
3594 |
+
"eval_samples_per_second": 159.713,
|
3595 |
+
"eval_steps_per_second": 9.982,
|
3596 |
+
"step": 5200
|
3597 |
+
},
|
3598 |
+
{
|
3599 |
+
"epoch": 3.33,
|
3600 |
+
"learning_rate": 3.3333333333333335e-05,
|
3601 |
+
"loss": 0.0007,
|
3602 |
+
"step": 5210
|
3603 |
+
},
|
3604 |
+
{
|
3605 |
+
"epoch": 3.34,
|
3606 |
+
"learning_rate": 3.3013435700575815e-05,
|
3607 |
+
"loss": 0.001,
|
3608 |
+
"step": 5220
|
3609 |
+
},
|
3610 |
+
{
|
3611 |
+
"epoch": 3.35,
|
3612 |
+
"learning_rate": 3.26935380678183e-05,
|
3613 |
+
"loss": 0.0065,
|
3614 |
+
"step": 5230
|
3615 |
+
},
|
3616 |
+
{
|
3617 |
+
"epoch": 3.35,
|
3618 |
+
"learning_rate": 3.237364043506078e-05,
|
3619 |
+
"loss": 0.0006,
|
3620 |
+
"step": 5240
|
3621 |
+
},
|
3622 |
+
{
|
3623 |
+
"epoch": 3.36,
|
3624 |
+
"learning_rate": 3.205374280230327e-05,
|
3625 |
+
"loss": 0.0007,
|
3626 |
+
"step": 5250
|
3627 |
+
},
|
3628 |
+
{
|
3629 |
+
"epoch": 3.37,
|
3630 |
+
"learning_rate": 3.173384516954575e-05,
|
3631 |
+
"loss": 0.0165,
|
3632 |
+
"step": 5260
|
3633 |
+
},
|
3634 |
+
{
|
3635 |
+
"epoch": 3.37,
|
3636 |
+
"learning_rate": 3.141394753678823e-05,
|
3637 |
+
"loss": 0.0119,
|
3638 |
+
"step": 5270
|
3639 |
+
},
|
3640 |
+
{
|
3641 |
+
"epoch": 3.38,
|
3642 |
+
"learning_rate": 3.1094049904030714e-05,
|
3643 |
+
"loss": 0.0006,
|
3644 |
+
"step": 5280
|
3645 |
+
},
|
3646 |
+
{
|
3647 |
+
"epoch": 3.38,
|
3648 |
+
"learning_rate": 3.0774152271273194e-05,
|
3649 |
+
"loss": 0.0074,
|
3650 |
+
"step": 5290
|
3651 |
+
},
|
3652 |
+
{
|
3653 |
+
"epoch": 3.39,
|
3654 |
+
"learning_rate": 3.0454254638515677e-05,
|
3655 |
+
"loss": 0.0064,
|
3656 |
+
"step": 5300
|
3657 |
+
},
|
3658 |
+
{
|
3659 |
+
"epoch": 3.39,
|
3660 |
+
"eval_accuracy": 0.9799,
|
3661 |
+
"eval_loss": 0.08600812405347824,
|
3662 |
+
"eval_runtime": 63.7405,
|
3663 |
+
"eval_samples_per_second": 156.886,
|
3664 |
+
"eval_steps_per_second": 9.805,
|
3665 |
+
"step": 5300
|
3666 |
+
},
|
3667 |
+
{
|
3668 |
+
"epoch": 3.4,
|
3669 |
+
"learning_rate": 3.0134357005758157e-05,
|
3670 |
+
"loss": 0.0027,
|
3671 |
+
"step": 5310
|
3672 |
+
},
|
3673 |
+
{
|
3674 |
+
"epoch": 3.4,
|
3675 |
+
"learning_rate": 2.981445937300064e-05,
|
3676 |
+
"loss": 0.0008,
|
3677 |
+
"step": 5320
|
3678 |
+
},
|
3679 |
+
{
|
3680 |
+
"epoch": 3.41,
|
3681 |
+
"learning_rate": 2.9494561740243127e-05,
|
3682 |
+
"loss": 0.0008,
|
3683 |
+
"step": 5330
|
3684 |
+
},
|
3685 |
+
{
|
3686 |
+
"epoch": 3.42,
|
3687 |
+
"learning_rate": 2.9174664107485606e-05,
|
3688 |
+
"loss": 0.0016,
|
3689 |
+
"step": 5340
|
3690 |
+
},
|
3691 |
+
{
|
3692 |
+
"epoch": 3.42,
|
3693 |
+
"learning_rate": 2.885476647472809e-05,
|
3694 |
+
"loss": 0.0174,
|
3695 |
+
"step": 5350
|
3696 |
+
},
|
3697 |
+
{
|
3698 |
+
"epoch": 3.43,
|
3699 |
+
"learning_rate": 2.853486884197057e-05,
|
3700 |
+
"loss": 0.0014,
|
3701 |
+
"step": 5360
|
3702 |
+
},
|
3703 |
+
{
|
3704 |
+
"epoch": 3.44,
|
3705 |
+
"learning_rate": 2.8214971209213053e-05,
|
3706 |
+
"loss": 0.0007,
|
3707 |
+
"step": 5370
|
3708 |
+
},
|
3709 |
+
{
|
3710 |
+
"epoch": 3.44,
|
3711 |
+
"learning_rate": 2.789507357645554e-05,
|
3712 |
+
"loss": 0.0007,
|
3713 |
+
"step": 5380
|
3714 |
+
},
|
3715 |
+
{
|
3716 |
+
"epoch": 3.45,
|
3717 |
+
"learning_rate": 2.7575175943698016e-05,
|
3718 |
+
"loss": 0.0035,
|
3719 |
+
"step": 5390
|
3720 |
+
},
|
3721 |
+
{
|
3722 |
+
"epoch": 3.45,
|
3723 |
+
"learning_rate": 2.7255278310940502e-05,
|
3724 |
+
"loss": 0.0009,
|
3725 |
+
"step": 5400
|
3726 |
+
},
|
3727 |
+
{
|
3728 |
+
"epoch": 3.45,
|
3729 |
+
"eval_accuracy": 0.9781,
|
3730 |
+
"eval_loss": 0.08675622195005417,
|
3731 |
+
"eval_runtime": 61.5807,
|
3732 |
+
"eval_samples_per_second": 162.388,
|
3733 |
+
"eval_steps_per_second": 10.149,
|
3734 |
+
"step": 5400
|
3735 |
+
},
|
3736 |
+
{
|
3737 |
+
"epoch": 3.46,
|
3738 |
+
"learning_rate": 2.6935380678182982e-05,
|
3739 |
+
"loss": 0.0007,
|
3740 |
+
"step": 5410
|
3741 |
+
},
|
3742 |
+
{
|
3743 |
+
"epoch": 3.47,
|
3744 |
+
"learning_rate": 2.6615483045425465e-05,
|
3745 |
+
"loss": 0.0009,
|
3746 |
+
"step": 5420
|
3747 |
+
},
|
3748 |
+
{
|
3749 |
+
"epoch": 3.47,
|
3750 |
+
"learning_rate": 2.6295585412667952e-05,
|
3751 |
+
"loss": 0.0011,
|
3752 |
+
"step": 5430
|
3753 |
+
},
|
3754 |
+
{
|
3755 |
+
"epoch": 3.48,
|
3756 |
+
"learning_rate": 2.5975687779910428e-05,
|
3757 |
+
"loss": 0.0006,
|
3758 |
+
"step": 5440
|
3759 |
+
},
|
3760 |
+
{
|
3761 |
+
"epoch": 3.49,
|
3762 |
+
"learning_rate": 2.5655790147152915e-05,
|
3763 |
+
"loss": 0.0007,
|
3764 |
+
"step": 5450
|
3765 |
+
},
|
3766 |
+
{
|
3767 |
+
"epoch": 3.49,
|
3768 |
+
"learning_rate": 2.533589251439539e-05,
|
3769 |
+
"loss": 0.0009,
|
3770 |
+
"step": 5460
|
3771 |
+
},
|
3772 |
+
{
|
3773 |
+
"epoch": 3.5,
|
3774 |
+
"learning_rate": 2.5015994881637878e-05,
|
3775 |
+
"loss": 0.0112,
|
3776 |
+
"step": 5470
|
3777 |
+
},
|
3778 |
+
{
|
3779 |
+
"epoch": 3.51,
|
3780 |
+
"learning_rate": 2.4696097248880358e-05,
|
3781 |
+
"loss": 0.0006,
|
3782 |
+
"step": 5480
|
3783 |
+
},
|
3784 |
+
{
|
3785 |
+
"epoch": 3.51,
|
3786 |
+
"learning_rate": 2.437619961612284e-05,
|
3787 |
+
"loss": 0.0007,
|
3788 |
+
"step": 5490
|
3789 |
+
},
|
3790 |
+
{
|
3791 |
+
"epoch": 3.52,
|
3792 |
+
"learning_rate": 2.4056301983365324e-05,
|
3793 |
+
"loss": 0.0007,
|
3794 |
+
"step": 5500
|
3795 |
+
},
|
3796 |
+
{
|
3797 |
+
"epoch": 3.52,
|
3798 |
+
"eval_accuracy": 0.9786,
|
3799 |
+
"eval_loss": 0.08598293364048004,
|
3800 |
+
"eval_runtime": 60.1457,
|
3801 |
+
"eval_samples_per_second": 166.263,
|
3802 |
+
"eval_steps_per_second": 10.391,
|
3803 |
+
"step": 5500
|
3804 |
+
},
|
3805 |
+
{
|
3806 |
+
"epoch": 3.53,
|
3807 |
+
"learning_rate": 2.3736404350607807e-05,
|
3808 |
+
"loss": 0.0008,
|
3809 |
+
"step": 5510
|
3810 |
+
},
|
3811 |
+
{
|
3812 |
+
"epoch": 3.53,
|
3813 |
+
"learning_rate": 2.341650671785029e-05,
|
3814 |
+
"loss": 0.0033,
|
3815 |
+
"step": 5520
|
3816 |
+
},
|
3817 |
+
{
|
3818 |
+
"epoch": 3.54,
|
3819 |
+
"learning_rate": 2.309660908509277e-05,
|
3820 |
+
"loss": 0.0009,
|
3821 |
+
"step": 5530
|
3822 |
+
},
|
3823 |
+
{
|
3824 |
+
"epoch": 3.54,
|
3825 |
+
"learning_rate": 2.2776711452335254e-05,
|
3826 |
+
"loss": 0.0006,
|
3827 |
+
"step": 5540
|
3828 |
+
},
|
3829 |
+
{
|
3830 |
+
"epoch": 3.55,
|
3831 |
+
"learning_rate": 2.2456813819577733e-05,
|
3832 |
+
"loss": 0.0007,
|
3833 |
+
"step": 5550
|
3834 |
+
},
|
3835 |
+
{
|
3836 |
+
"epoch": 3.56,
|
3837 |
+
"learning_rate": 2.213691618682022e-05,
|
3838 |
+
"loss": 0.0006,
|
3839 |
+
"step": 5560
|
3840 |
+
},
|
3841 |
+
{
|
3842 |
+
"epoch": 3.56,
|
3843 |
+
"learning_rate": 2.1817018554062703e-05,
|
3844 |
+
"loss": 0.0013,
|
3845 |
+
"step": 5570
|
3846 |
+
},
|
3847 |
+
{
|
3848 |
+
"epoch": 3.57,
|
3849 |
+
"learning_rate": 2.1497120921305183e-05,
|
3850 |
+
"loss": 0.0037,
|
3851 |
+
"step": 5580
|
3852 |
+
},
|
3853 |
+
{
|
3854 |
+
"epoch": 3.58,
|
3855 |
+
"learning_rate": 2.1177223288547666e-05,
|
3856 |
+
"loss": 0.0009,
|
3857 |
+
"step": 5590
|
3858 |
+
},
|
3859 |
+
{
|
3860 |
+
"epoch": 3.58,
|
3861 |
+
"learning_rate": 2.0857325655790146e-05,
|
3862 |
+
"loss": 0.0006,
|
3863 |
+
"step": 5600
|
3864 |
+
},
|
3865 |
+
{
|
3866 |
+
"epoch": 3.58,
|
3867 |
+
"eval_accuracy": 0.9796,
|
3868 |
+
"eval_loss": 0.0850202739238739,
|
3869 |
+
"eval_runtime": 62.0241,
|
3870 |
+
"eval_samples_per_second": 161.228,
|
3871 |
+
"eval_steps_per_second": 10.077,
|
3872 |
+
"step": 5600
|
3873 |
+
},
|
3874 |
+
{
|
3875 |
+
"epoch": 3.59,
|
3876 |
+
"learning_rate": 2.0537428023032633e-05,
|
3877 |
+
"loss": 0.0047,
|
3878 |
+
"step": 5610
|
3879 |
+
},
|
3880 |
+
{
|
3881 |
+
"epoch": 3.6,
|
3882 |
+
"learning_rate": 2.0217530390275112e-05,
|
3883 |
+
"loss": 0.0013,
|
3884 |
+
"step": 5620
|
3885 |
+
},
|
3886 |
+
{
|
3887 |
+
"epoch": 3.6,
|
3888 |
+
"learning_rate": 1.9897632757517596e-05,
|
3889 |
+
"loss": 0.0018,
|
3890 |
+
"step": 5630
|
3891 |
+
},
|
3892 |
+
{
|
3893 |
+
"epoch": 3.61,
|
3894 |
+
"learning_rate": 1.957773512476008e-05,
|
3895 |
+
"loss": 0.0006,
|
3896 |
+
"step": 5640
|
3897 |
+
},
|
3898 |
+
{
|
3899 |
+
"epoch": 3.61,
|
3900 |
+
"learning_rate": 1.925783749200256e-05,
|
3901 |
+
"loss": 0.0134,
|
3902 |
+
"step": 5650
|
3903 |
+
},
|
3904 |
+
{
|
3905 |
+
"epoch": 3.62,
|
3906 |
+
"learning_rate": 1.8937939859245045e-05,
|
3907 |
+
"loss": 0.0014,
|
3908 |
+
"step": 5660
|
3909 |
+
},
|
3910 |
+
{
|
3911 |
+
"epoch": 3.63,
|
3912 |
+
"learning_rate": 1.8618042226487525e-05,
|
3913 |
+
"loss": 0.0018,
|
3914 |
+
"step": 5670
|
3915 |
+
},
|
3916 |
+
{
|
3917 |
+
"epoch": 3.63,
|
3918 |
+
"learning_rate": 1.8298144593730008e-05,
|
3919 |
+
"loss": 0.0007,
|
3920 |
+
"step": 5680
|
3921 |
+
},
|
3922 |
+
{
|
3923 |
+
"epoch": 3.64,
|
3924 |
+
"learning_rate": 1.7978246960972488e-05,
|
3925 |
+
"loss": 0.0006,
|
3926 |
+
"step": 5690
|
3927 |
+
},
|
3928 |
+
{
|
3929 |
+
"epoch": 3.65,
|
3930 |
+
"learning_rate": 1.765834932821497e-05,
|
3931 |
+
"loss": 0.0005,
|
3932 |
+
"step": 5700
|
3933 |
+
},
|
3934 |
+
{
|
3935 |
+
"epoch": 3.65,
|
3936 |
+
"eval_accuracy": 0.9804,
|
3937 |
+
"eval_loss": 0.08065660297870636,
|
3938 |
+
"eval_runtime": 63.2588,
|
3939 |
+
"eval_samples_per_second": 158.081,
|
3940 |
+
"eval_steps_per_second": 9.88,
|
3941 |
+
"step": 5700
|
3942 |
+
},
|
3943 |
+
{
|
3944 |
+
"epoch": 3.65,
|
3945 |
+
"learning_rate": 1.7338451695457454e-05,
|
3946 |
+
"loss": 0.01,
|
3947 |
+
"step": 5710
|
3948 |
+
},
|
3949 |
+
{
|
3950 |
+
"epoch": 3.66,
|
3951 |
+
"learning_rate": 1.7018554062699938e-05,
|
3952 |
+
"loss": 0.0005,
|
3953 |
+
"step": 5720
|
3954 |
+
},
|
3955 |
+
{
|
3956 |
+
"epoch": 3.67,
|
3957 |
+
"learning_rate": 1.669865642994242e-05,
|
3958 |
+
"loss": 0.0009,
|
3959 |
+
"step": 5730
|
3960 |
+
},
|
3961 |
+
{
|
3962 |
+
"epoch": 3.67,
|
3963 |
+
"learning_rate": 1.63787587971849e-05,
|
3964 |
+
"loss": 0.0005,
|
3965 |
+
"step": 5740
|
3966 |
+
},
|
3967 |
+
{
|
3968 |
+
"epoch": 3.68,
|
3969 |
+
"learning_rate": 1.6058861164427384e-05,
|
3970 |
+
"loss": 0.0007,
|
3971 |
+
"step": 5750
|
3972 |
+
},
|
3973 |
+
{
|
3974 |
+
"epoch": 3.69,
|
3975 |
+
"learning_rate": 1.5738963531669867e-05,
|
3976 |
+
"loss": 0.002,
|
3977 |
+
"step": 5760
|
3978 |
+
},
|
3979 |
+
{
|
3980 |
+
"epoch": 3.69,
|
3981 |
+
"learning_rate": 1.541906589891235e-05,
|
3982 |
+
"loss": 0.0005,
|
3983 |
+
"step": 5770
|
3984 |
+
},
|
3985 |
+
{
|
3986 |
+
"epoch": 3.7,
|
3987 |
+
"learning_rate": 1.5099168266154832e-05,
|
3988 |
+
"loss": 0.0008,
|
3989 |
+
"step": 5780
|
3990 |
+
},
|
3991 |
+
{
|
3992 |
+
"epoch": 3.7,
|
3993 |
+
"learning_rate": 1.4779270633397313e-05,
|
3994 |
+
"loss": 0.0026,
|
3995 |
+
"step": 5790
|
3996 |
+
},
|
3997 |
+
{
|
3998 |
+
"epoch": 3.71,
|
3999 |
+
"learning_rate": 1.4459373000639795e-05,
|
4000 |
+
"loss": 0.0005,
|
4001 |
+
"step": 5800
|
4002 |
+
},
|
4003 |
+
{
|
4004 |
+
"epoch": 3.71,
|
4005 |
+
"eval_accuracy": 0.981,
|
4006 |
+
"eval_loss": 0.07981295883655548,
|
4007 |
+
"eval_runtime": 62.5683,
|
4008 |
+
"eval_samples_per_second": 159.825,
|
4009 |
+
"eval_steps_per_second": 9.989,
|
4010 |
+
"step": 5800
|
4011 |
+
},
|
4012 |
+
{
|
4013 |
+
"epoch": 3.72,
|
4014 |
+
"learning_rate": 1.413947536788228e-05,
|
4015 |
+
"loss": 0.0015,
|
4016 |
+
"step": 5810
|
4017 |
+
},
|
4018 |
+
{
|
4019 |
+
"epoch": 3.72,
|
4020 |
+
"learning_rate": 1.3819577735124761e-05,
|
4021 |
+
"loss": 0.0007,
|
4022 |
+
"step": 5820
|
4023 |
+
},
|
4024 |
+
{
|
4025 |
+
"epoch": 3.73,
|
4026 |
+
"learning_rate": 1.3499680102367243e-05,
|
4027 |
+
"loss": 0.0058,
|
4028 |
+
"step": 5830
|
4029 |
+
},
|
4030 |
+
{
|
4031 |
+
"epoch": 3.74,
|
4032 |
+
"learning_rate": 1.3179782469609724e-05,
|
4033 |
+
"loss": 0.0006,
|
4034 |
+
"step": 5840
|
4035 |
+
},
|
4036 |
+
{
|
4037 |
+
"epoch": 3.74,
|
4038 |
+
"learning_rate": 1.2859884836852207e-05,
|
4039 |
+
"loss": 0.0058,
|
4040 |
+
"step": 5850
|
4041 |
+
},
|
4042 |
+
{
|
4043 |
+
"epoch": 3.75,
|
4044 |
+
"learning_rate": 1.253998720409469e-05,
|
4045 |
+
"loss": 0.0007,
|
4046 |
+
"step": 5860
|
4047 |
+
},
|
4048 |
+
{
|
4049 |
+
"epoch": 3.76,
|
4050 |
+
"learning_rate": 1.2220089571337174e-05,
|
4051 |
+
"loss": 0.0005,
|
4052 |
+
"step": 5870
|
4053 |
+
},
|
4054 |
+
{
|
4055 |
+
"epoch": 3.76,
|
4056 |
+
"learning_rate": 1.1900191938579655e-05,
|
4057 |
+
"loss": 0.0008,
|
4058 |
+
"step": 5880
|
4059 |
+
},
|
4060 |
+
{
|
4061 |
+
"epoch": 3.77,
|
4062 |
+
"learning_rate": 1.1580294305822137e-05,
|
4063 |
+
"loss": 0.0004,
|
4064 |
+
"step": 5890
|
4065 |
+
},
|
4066 |
+
{
|
4067 |
+
"epoch": 3.77,
|
4068 |
+
"learning_rate": 1.126039667306462e-05,
|
4069 |
+
"loss": 0.0016,
|
4070 |
+
"step": 5900
|
4071 |
+
},
|
4072 |
+
{
|
4073 |
+
"epoch": 3.77,
|
4074 |
+
"eval_accuracy": 0.9808,
|
4075 |
+
"eval_loss": 0.07946650683879852,
|
4076 |
+
"eval_runtime": 60.9852,
|
4077 |
+
"eval_samples_per_second": 163.974,
|
4078 |
+
"eval_steps_per_second": 10.248,
|
4079 |
+
"step": 5900
|
4080 |
+
},
|
4081 |
+
{
|
4082 |
+
"epoch": 3.78,
|
4083 |
+
"learning_rate": 1.0940499040307102e-05,
|
4084 |
+
"loss": 0.0006,
|
4085 |
+
"step": 5910
|
4086 |
+
},
|
4087 |
+
{
|
4088 |
+
"epoch": 3.79,
|
4089 |
+
"learning_rate": 1.0620601407549585e-05,
|
4090 |
+
"loss": 0.0058,
|
4091 |
+
"step": 5920
|
4092 |
+
},
|
4093 |
+
{
|
4094 |
+
"epoch": 3.79,
|
4095 |
+
"learning_rate": 1.0300703774792066e-05,
|
4096 |
+
"loss": 0.0035,
|
4097 |
+
"step": 5930
|
4098 |
+
},
|
4099 |
+
{
|
4100 |
+
"epoch": 3.8,
|
4101 |
+
"learning_rate": 9.98080614203455e-06,
|
4102 |
+
"loss": 0.0024,
|
4103 |
+
"step": 5940
|
4104 |
+
},
|
4105 |
+
{
|
4106 |
+
"epoch": 3.81,
|
4107 |
+
"learning_rate": 9.660908509277033e-06,
|
4108 |
+
"loss": 0.0005,
|
4109 |
+
"step": 5950
|
4110 |
+
},
|
4111 |
+
{
|
4112 |
+
"epoch": 3.81,
|
4113 |
+
"learning_rate": 9.341010876519514e-06,
|
4114 |
+
"loss": 0.0048,
|
4115 |
+
"step": 5960
|
4116 |
+
},
|
4117 |
+
{
|
4118 |
+
"epoch": 3.82,
|
4119 |
+
"learning_rate": 9.021113243761997e-06,
|
4120 |
+
"loss": 0.001,
|
4121 |
+
"step": 5970
|
4122 |
+
},
|
4123 |
+
{
|
4124 |
+
"epoch": 3.83,
|
4125 |
+
"learning_rate": 8.701215611004479e-06,
|
4126 |
+
"loss": 0.0006,
|
4127 |
+
"step": 5980
|
4128 |
+
},
|
4129 |
+
{
|
4130 |
+
"epoch": 3.83,
|
4131 |
+
"learning_rate": 8.38131797824696e-06,
|
4132 |
+
"loss": 0.001,
|
4133 |
+
"step": 5990
|
4134 |
+
},
|
4135 |
+
{
|
4136 |
+
"epoch": 3.84,
|
4137 |
+
"learning_rate": 8.061420345489444e-06,
|
4138 |
+
"loss": 0.0068,
|
4139 |
+
"step": 6000
|
4140 |
+
},
|
4141 |
+
{
|
4142 |
+
"epoch": 3.84,
|
4143 |
+
"eval_accuracy": 0.9806,
|
4144 |
+
"eval_loss": 0.08210848271846771,
|
4145 |
+
"eval_runtime": 64.268,
|
4146 |
+
"eval_samples_per_second": 155.598,
|
4147 |
+
"eval_steps_per_second": 9.725,
|
4148 |
+
"step": 6000
|
4149 |
+
},
|
4150 |
+
{
|
4151 |
+
"epoch": 3.85,
|
4152 |
+
"learning_rate": 7.741522712731925e-06,
|
4153 |
+
"loss": 0.0004,
|
4154 |
+
"step": 6010
|
4155 |
+
},
|
4156 |
+
{
|
4157 |
+
"epoch": 3.85,
|
4158 |
+
"learning_rate": 7.421625079974409e-06,
|
4159 |
+
"loss": 0.0005,
|
4160 |
+
"step": 6020
|
4161 |
+
},
|
4162 |
+
{
|
4163 |
+
"epoch": 3.86,
|
4164 |
+
"learning_rate": 7.101727447216891e-06,
|
4165 |
+
"loss": 0.0007,
|
4166 |
+
"step": 6030
|
4167 |
+
},
|
4168 |
+
{
|
4169 |
+
"epoch": 3.86,
|
4170 |
+
"learning_rate": 6.781829814459373e-06,
|
4171 |
+
"loss": 0.0005,
|
4172 |
+
"step": 6040
|
4173 |
+
},
|
4174 |
+
{
|
4175 |
+
"epoch": 3.87,
|
4176 |
+
"learning_rate": 6.461932181701856e-06,
|
4177 |
+
"loss": 0.0164,
|
4178 |
+
"step": 6050
|
4179 |
+
},
|
4180 |
+
{
|
4181 |
+
"epoch": 3.88,
|
4182 |
+
"learning_rate": 6.142034548944338e-06,
|
4183 |
+
"loss": 0.0005,
|
4184 |
+
"step": 6060
|
4185 |
+
},
|
4186 |
+
{
|
4187 |
+
"epoch": 3.88,
|
4188 |
+
"learning_rate": 5.82213691618682e-06,
|
4189 |
+
"loss": 0.0007,
|
4190 |
+
"step": 6070
|
4191 |
+
},
|
4192 |
+
{
|
4193 |
+
"epoch": 3.89,
|
4194 |
+
"learning_rate": 5.502239283429303e-06,
|
4195 |
+
"loss": 0.0006,
|
4196 |
+
"step": 6080
|
4197 |
+
},
|
4198 |
+
{
|
4199 |
+
"epoch": 3.9,
|
4200 |
+
"learning_rate": 5.182341650671786e-06,
|
4201 |
+
"loss": 0.0007,
|
4202 |
+
"step": 6090
|
4203 |
+
},
|
4204 |
+
{
|
4205 |
+
"epoch": 3.9,
|
4206 |
+
"learning_rate": 4.862444017914268e-06,
|
4207 |
+
"loss": 0.001,
|
4208 |
+
"step": 6100
|
4209 |
+
},
|
4210 |
+
{
|
4211 |
+
"epoch": 3.9,
|
4212 |
+
"eval_accuracy": 0.981,
|
4213 |
+
"eval_loss": 0.08095718175172806,
|
4214 |
+
"eval_runtime": 62.2344,
|
4215 |
+
"eval_samples_per_second": 160.683,
|
4216 |
+
"eval_steps_per_second": 10.043,
|
4217 |
+
"step": 6100
|
4218 |
+
},
|
4219 |
+
{
|
4220 |
+
"epoch": 3.91,
|
4221 |
+
"learning_rate": 4.5425463851567495e-06,
|
4222 |
+
"loss": 0.0005,
|
4223 |
+
"step": 6110
|
4224 |
+
},
|
4225 |
+
{
|
4226 |
+
"epoch": 3.92,
|
4227 |
+
"learning_rate": 4.222648752399233e-06,
|
4228 |
+
"loss": 0.0006,
|
4229 |
+
"step": 6120
|
4230 |
+
},
|
4231 |
+
{
|
4232 |
+
"epoch": 3.92,
|
4233 |
+
"learning_rate": 3.902751119641715e-06,
|
4234 |
+
"loss": 0.0067,
|
4235 |
+
"step": 6130
|
4236 |
+
},
|
4237 |
+
{
|
4238 |
+
"epoch": 3.93,
|
4239 |
+
"learning_rate": 3.5828534868841974e-06,
|
4240 |
+
"loss": 0.001,
|
4241 |
+
"step": 6140
|
4242 |
+
},
|
4243 |
+
{
|
4244 |
+
"epoch": 3.93,
|
4245 |
+
"learning_rate": 3.2629558541266794e-06,
|
4246 |
+
"loss": 0.0009,
|
4247 |
+
"step": 6150
|
4248 |
+
},
|
4249 |
+
{
|
4250 |
+
"epoch": 3.94,
|
4251 |
+
"learning_rate": 2.943058221369162e-06,
|
4252 |
+
"loss": 0.0005,
|
4253 |
+
"step": 6160
|
4254 |
+
},
|
4255 |
+
{
|
4256 |
+
"epoch": 3.95,
|
4257 |
+
"learning_rate": 2.6231605886116445e-06,
|
4258 |
+
"loss": 0.0005,
|
4259 |
+
"step": 6170
|
4260 |
+
},
|
4261 |
+
{
|
4262 |
+
"epoch": 3.95,
|
4263 |
+
"learning_rate": 2.303262955854127e-06,
|
4264 |
+
"loss": 0.0006,
|
4265 |
+
"step": 6180
|
4266 |
+
},
|
4267 |
+
{
|
4268 |
+
"epoch": 3.96,
|
4269 |
+
"learning_rate": 1.9833653230966092e-06,
|
4270 |
+
"loss": 0.005,
|
4271 |
+
"step": 6190
|
4272 |
+
},
|
4273 |
+
{
|
4274 |
+
"epoch": 3.97,
|
4275 |
+
"learning_rate": 1.6634676903390916e-06,
|
4276 |
+
"loss": 0.0009,
|
4277 |
+
"step": 6200
|
4278 |
+
},
|
4279 |
+
{
|
4280 |
+
"epoch": 3.97,
|
4281 |
+
"eval_accuracy": 0.9812,
|
4282 |
+
"eval_loss": 0.07883724570274353,
|
4283 |
+
"eval_runtime": 62.3579,
|
4284 |
+
"eval_samples_per_second": 160.364,
|
4285 |
+
"eval_steps_per_second": 10.023,
|
4286 |
+
"step": 6200
|
4287 |
+
},
|
4288 |
+
{
|
4289 |
+
"epoch": 3.97,
|
4290 |
+
"learning_rate": 1.343570057581574e-06,
|
4291 |
+
"loss": 0.0006,
|
4292 |
+
"step": 6210
|
4293 |
+
},
|
4294 |
+
{
|
4295 |
+
"epoch": 3.98,
|
4296 |
+
"learning_rate": 1.0236724248240563e-06,
|
4297 |
+
"loss": 0.0005,
|
4298 |
+
"step": 6220
|
4299 |
+
},
|
4300 |
+
{
|
4301 |
+
"epoch": 3.99,
|
4302 |
+
"learning_rate": 7.037747920665388e-07,
|
4303 |
+
"loss": 0.0005,
|
4304 |
+
"step": 6230
|
4305 |
+
},
|
4306 |
+
{
|
4307 |
+
"epoch": 3.99,
|
4308 |
+
"learning_rate": 3.838771593090211e-07,
|
4309 |
+
"loss": 0.0005,
|
4310 |
+
"step": 6240
|
4311 |
+
},
|
4312 |
+
{
|
4313 |
+
"epoch": 4.0,
|
4314 |
+
"learning_rate": 6.397952655150352e-08,
|
4315 |
+
"loss": 0.0005,
|
4316 |
+
"step": 6250
|
4317 |
+
},
|
4318 |
+
{
|
4319 |
+
"epoch": 4.0,
|
4320 |
+
"step": 6252,
|
4321 |
+
"total_flos": 1.54995091808256e+19,
|
4322 |
+
"train_loss": 0.08309991182349008,
|
4323 |
+
"train_runtime": 5212.8854,
|
4324 |
+
"train_samples_per_second": 38.366,
|
4325 |
+
"train_steps_per_second": 1.199
|
4326 |
+
}
|
4327 |
+
],
|
4328 |
+
"max_steps": 6252,
|
4329 |
+
"num_train_epochs": 4,
|
4330 |
+
"total_flos": 1.54995091808256e+19,
|
4331 |
+
"trial_name": null,
|
4332 |
+
"trial_params": null
|
4333 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:db1d34eacb0aa2d1405bd06bae81df959b1b2e36a899c7e53a3788632452635a
|
3 |
+
size 3311
|