Commit
•
0aab787
1
Parent(s):
0a7f88f
End of training
Browse files- README.md +3 -3
- all_results.json +12 -0
- eval_results.json +8 -0
- runs/May23_17-20-58_DESKTOP-SKBE9FB/events.out.tfevents.1716509541.DESKTOP-SKBE9FB.4988.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +1302 -0
README.md
CHANGED
@@ -22,7 +22,7 @@ model-index:
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
-
value: 0.
|
26 |
---
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
32 |
|
33 |
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the imagefolder dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
-
- Loss: 1.
|
36 |
-
- Accuracy: 0.
|
37 |
|
38 |
## Model description
|
39 |
|
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
+
value: 0.8333333333333334
|
26 |
---
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
32 |
|
33 |
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the imagefolder dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 1.2003
|
36 |
+
- Accuracy: 0.8333
|
37 |
|
38 |
## Model description
|
39 |
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 39.48,
|
3 |
+
"eval_accuracy": 0.8333333333333334,
|
4 |
+
"eval_loss": 1.2002999782562256,
|
5 |
+
"eval_runtime": 1.1818,
|
6 |
+
"eval_samples_per_second": 50.772,
|
7 |
+
"eval_steps_per_second": 1.692,
|
8 |
+
"train_loss": 0.11935894162835259,
|
9 |
+
"train_runtime": 3007.0998,
|
10 |
+
"train_samples_per_second": 65.179,
|
11 |
+
"train_steps_per_second": 0.505
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 39.48,
|
3 |
+
"eval_accuracy": 0.8333333333333334,
|
4 |
+
"eval_loss": 1.2002999782562256,
|
5 |
+
"eval_runtime": 1.1818,
|
6 |
+
"eval_samples_per_second": 50.772,
|
7 |
+
"eval_steps_per_second": 1.692
|
8 |
+
}
|
runs/May23_17-20-58_DESKTOP-SKBE9FB/events.out.tfevents.1716509541.DESKTOP-SKBE9FB.4988.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b251cfa9399fa672fc3a87db1e0580b872b85ed76e4c323e11fa5a7c153fcb45
|
3 |
+
size 411
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 39.48,
|
3 |
+
"train_loss": 0.11935894162835259,
|
4 |
+
"train_runtime": 3007.0998,
|
5 |
+
"train_samples_per_second": 65.179,
|
6 |
+
"train_steps_per_second": 0.505
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1302 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.8333333333333334,
|
3 |
+
"best_model_checkpoint": "vit-base-patch16-224-RU2-40\\checkpoint-1155",
|
4 |
+
"epoch": 39.48051948051948,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1520,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.26,
|
13 |
+
"learning_rate": 7.2368421052631575e-06,
|
14 |
+
"loss": 1.384,
|
15 |
+
"step": 10
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.52,
|
19 |
+
"learning_rate": 1.4473684210526315e-05,
|
20 |
+
"loss": 1.3667,
|
21 |
+
"step": 20
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.78,
|
25 |
+
"learning_rate": 2.1710526315789474e-05,
|
26 |
+
"loss": 1.3226,
|
27 |
+
"step": 30
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.99,
|
31 |
+
"eval_accuracy": 0.6,
|
32 |
+
"eval_loss": 1.2293227910995483,
|
33 |
+
"eval_runtime": 1.0844,
|
34 |
+
"eval_samples_per_second": 55.33,
|
35 |
+
"eval_steps_per_second": 1.844,
|
36 |
+
"step": 38
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 1.04,
|
40 |
+
"learning_rate": 2.894736842105263e-05,
|
41 |
+
"loss": 1.2601,
|
42 |
+
"step": 40
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 1.3,
|
46 |
+
"learning_rate": 3.618421052631579e-05,
|
47 |
+
"loss": 1.1706,
|
48 |
+
"step": 50
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 1.56,
|
52 |
+
"learning_rate": 4.342105263157895e-05,
|
53 |
+
"loss": 1.0292,
|
54 |
+
"step": 60
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 1.82,
|
58 |
+
"learning_rate": 5.0657894736842104e-05,
|
59 |
+
"loss": 0.9048,
|
60 |
+
"step": 70
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 2.0,
|
64 |
+
"eval_accuracy": 0.7,
|
65 |
+
"eval_loss": 0.7968690991401672,
|
66 |
+
"eval_runtime": 1.0427,
|
67 |
+
"eval_samples_per_second": 57.542,
|
68 |
+
"eval_steps_per_second": 1.918,
|
69 |
+
"step": 77
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 2.08,
|
73 |
+
"learning_rate": 5.484764542936288e-05,
|
74 |
+
"loss": 0.7837,
|
75 |
+
"step": 80
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 2.34,
|
79 |
+
"learning_rate": 5.446675900277008e-05,
|
80 |
+
"loss": 0.6251,
|
81 |
+
"step": 90
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 2.6,
|
85 |
+
"learning_rate": 5.408587257617729e-05,
|
86 |
+
"loss": 0.4961,
|
87 |
+
"step": 100
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 2.86,
|
91 |
+
"learning_rate": 5.370498614958449e-05,
|
92 |
+
"loss": 0.4039,
|
93 |
+
"step": 110
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 2.99,
|
97 |
+
"eval_accuracy": 0.7166666666666667,
|
98 |
+
"eval_loss": 0.6800220012664795,
|
99 |
+
"eval_runtime": 1.1199,
|
100 |
+
"eval_samples_per_second": 53.578,
|
101 |
+
"eval_steps_per_second": 1.786,
|
102 |
+
"step": 115
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 3.12,
|
106 |
+
"learning_rate": 5.332409972299169e-05,
|
107 |
+
"loss": 0.3074,
|
108 |
+
"step": 120
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 3.38,
|
112 |
+
"learning_rate": 5.294321329639889e-05,
|
113 |
+
"loss": 0.2772,
|
114 |
+
"step": 130
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 3.64,
|
118 |
+
"learning_rate": 5.256232686980609e-05,
|
119 |
+
"loss": 0.2567,
|
120 |
+
"step": 140
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 3.9,
|
124 |
+
"learning_rate": 5.21814404432133e-05,
|
125 |
+
"loss": 0.281,
|
126 |
+
"step": 150
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 4.0,
|
130 |
+
"eval_accuracy": 0.7666666666666667,
|
131 |
+
"eval_loss": 0.8892202973365784,
|
132 |
+
"eval_runtime": 1.134,
|
133 |
+
"eval_samples_per_second": 52.909,
|
134 |
+
"eval_steps_per_second": 1.764,
|
135 |
+
"step": 154
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 4.16,
|
139 |
+
"learning_rate": 5.18005540166205e-05,
|
140 |
+
"loss": 0.2168,
|
141 |
+
"step": 160
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 4.42,
|
145 |
+
"learning_rate": 5.14196675900277e-05,
|
146 |
+
"loss": 0.1975,
|
147 |
+
"step": 170
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 4.68,
|
151 |
+
"learning_rate": 5.1038781163434903e-05,
|
152 |
+
"loss": 0.1988,
|
153 |
+
"step": 180
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 4.94,
|
157 |
+
"learning_rate": 5.0657894736842104e-05,
|
158 |
+
"loss": 0.1755,
|
159 |
+
"step": 190
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 4.99,
|
163 |
+
"eval_accuracy": 0.7333333333333333,
|
164 |
+
"eval_loss": 0.9071826934814453,
|
165 |
+
"eval_runtime": 1.08,
|
166 |
+
"eval_samples_per_second": 55.555,
|
167 |
+
"eval_steps_per_second": 1.852,
|
168 |
+
"step": 192
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 5.19,
|
172 |
+
"learning_rate": 5.027700831024931e-05,
|
173 |
+
"loss": 0.1255,
|
174 |
+
"step": 200
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 5.45,
|
178 |
+
"learning_rate": 4.989612188365651e-05,
|
179 |
+
"loss": 0.1565,
|
180 |
+
"step": 210
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 5.71,
|
184 |
+
"learning_rate": 4.9515235457063714e-05,
|
185 |
+
"loss": 0.1548,
|
186 |
+
"step": 220
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 5.97,
|
190 |
+
"learning_rate": 4.9134349030470915e-05,
|
191 |
+
"loss": 0.1035,
|
192 |
+
"step": 230
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 6.0,
|
196 |
+
"eval_accuracy": 0.8166666666666667,
|
197 |
+
"eval_loss": 0.8035634756088257,
|
198 |
+
"eval_runtime": 1.0846,
|
199 |
+
"eval_samples_per_second": 55.318,
|
200 |
+
"eval_steps_per_second": 1.844,
|
201 |
+
"step": 231
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 6.23,
|
205 |
+
"learning_rate": 4.8753462603878116e-05,
|
206 |
+
"loss": 0.1059,
|
207 |
+
"step": 240
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 6.49,
|
211 |
+
"learning_rate": 4.837257617728532e-05,
|
212 |
+
"loss": 0.1615,
|
213 |
+
"step": 250
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 6.75,
|
217 |
+
"learning_rate": 4.7991689750692524e-05,
|
218 |
+
"loss": 0.1275,
|
219 |
+
"step": 260
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 6.99,
|
223 |
+
"eval_accuracy": 0.8,
|
224 |
+
"eval_loss": 0.8626561760902405,
|
225 |
+
"eval_runtime": 1.1485,
|
226 |
+
"eval_samples_per_second": 52.243,
|
227 |
+
"eval_steps_per_second": 1.741,
|
228 |
+
"step": 269
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 7.01,
|
232 |
+
"learning_rate": 4.7610803324099725e-05,
|
233 |
+
"loss": 0.1117,
|
234 |
+
"step": 270
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 7.27,
|
238 |
+
"learning_rate": 4.7229916897506926e-05,
|
239 |
+
"loss": 0.0851,
|
240 |
+
"step": 280
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 7.53,
|
244 |
+
"learning_rate": 4.684903047091413e-05,
|
245 |
+
"loss": 0.1013,
|
246 |
+
"step": 290
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 7.79,
|
250 |
+
"learning_rate": 4.6468144044321335e-05,
|
251 |
+
"loss": 0.107,
|
252 |
+
"step": 300
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 8.0,
|
256 |
+
"eval_accuracy": 0.8,
|
257 |
+
"eval_loss": 0.8712760806083679,
|
258 |
+
"eval_runtime": 1.0545,
|
259 |
+
"eval_samples_per_second": 56.898,
|
260 |
+
"eval_steps_per_second": 1.897,
|
261 |
+
"step": 308
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 8.05,
|
265 |
+
"learning_rate": 4.6087257617728535e-05,
|
266 |
+
"loss": 0.0832,
|
267 |
+
"step": 310
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 8.31,
|
271 |
+
"learning_rate": 4.570637119113573e-05,
|
272 |
+
"loss": 0.087,
|
273 |
+
"step": 320
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 8.57,
|
277 |
+
"learning_rate": 4.532548476454294e-05,
|
278 |
+
"loss": 0.0847,
|
279 |
+
"step": 330
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 8.83,
|
283 |
+
"learning_rate": 4.494459833795014e-05,
|
284 |
+
"loss": 0.0984,
|
285 |
+
"step": 340
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 8.99,
|
289 |
+
"eval_accuracy": 0.8,
|
290 |
+
"eval_loss": 0.965960681438446,
|
291 |
+
"eval_runtime": 1.0848,
|
292 |
+
"eval_samples_per_second": 55.31,
|
293 |
+
"eval_steps_per_second": 1.844,
|
294 |
+
"step": 346
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 9.09,
|
298 |
+
"learning_rate": 4.4563711911357346e-05,
|
299 |
+
"loss": 0.082,
|
300 |
+
"step": 350
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 9.35,
|
304 |
+
"learning_rate": 4.418282548476455e-05,
|
305 |
+
"loss": 0.0803,
|
306 |
+
"step": 360
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 9.61,
|
310 |
+
"learning_rate": 4.380193905817174e-05,
|
311 |
+
"loss": 0.0774,
|
312 |
+
"step": 370
|
313 |
+
},
|
314 |
+
{
|
315 |
+
"epoch": 9.87,
|
316 |
+
"learning_rate": 4.342105263157895e-05,
|
317 |
+
"loss": 0.0823,
|
318 |
+
"step": 380
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 10.0,
|
322 |
+
"eval_accuracy": 0.7833333333333333,
|
323 |
+
"eval_loss": 1.070397973060608,
|
324 |
+
"eval_runtime": 1.1053,
|
325 |
+
"eval_samples_per_second": 54.286,
|
326 |
+
"eval_steps_per_second": 1.81,
|
327 |
+
"step": 385
|
328 |
+
},
|
329 |
+
{
|
330 |
+
"epoch": 10.13,
|
331 |
+
"learning_rate": 4.304016620498615e-05,
|
332 |
+
"loss": 0.0741,
|
333 |
+
"step": 390
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 10.39,
|
337 |
+
"learning_rate": 4.265927977839336e-05,
|
338 |
+
"loss": 0.0542,
|
339 |
+
"step": 400
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 10.65,
|
343 |
+
"learning_rate": 4.227839335180056e-05,
|
344 |
+
"loss": 0.0679,
|
345 |
+
"step": 410
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 10.91,
|
349 |
+
"learning_rate": 4.189750692520776e-05,
|
350 |
+
"loss": 0.0771,
|
351 |
+
"step": 420
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 10.99,
|
355 |
+
"eval_accuracy": 0.7666666666666667,
|
356 |
+
"eval_loss": 0.9409402012825012,
|
357 |
+
"eval_runtime": 1.2296,
|
358 |
+
"eval_samples_per_second": 48.796,
|
359 |
+
"eval_steps_per_second": 1.627,
|
360 |
+
"step": 423
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 11.17,
|
364 |
+
"learning_rate": 4.151662049861496e-05,
|
365 |
+
"loss": 0.0665,
|
366 |
+
"step": 430
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 11.43,
|
370 |
+
"learning_rate": 4.113573407202216e-05,
|
371 |
+
"loss": 0.0413,
|
372 |
+
"step": 440
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 11.69,
|
376 |
+
"learning_rate": 4.075484764542937e-05,
|
377 |
+
"loss": 0.0644,
|
378 |
+
"step": 450
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 11.95,
|
382 |
+
"learning_rate": 4.037396121883656e-05,
|
383 |
+
"loss": 0.0527,
|
384 |
+
"step": 460
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 12.0,
|
388 |
+
"eval_accuracy": 0.7833333333333333,
|
389 |
+
"eval_loss": 1.0052143335342407,
|
390 |
+
"eval_runtime": 1.0721,
|
391 |
+
"eval_samples_per_second": 55.965,
|
392 |
+
"eval_steps_per_second": 1.865,
|
393 |
+
"step": 462
|
394 |
+
},
|
395 |
+
{
|
396 |
+
"epoch": 12.21,
|
397 |
+
"learning_rate": 3.999307479224377e-05,
|
398 |
+
"loss": 0.064,
|
399 |
+
"step": 470
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 12.47,
|
403 |
+
"learning_rate": 3.961218836565097e-05,
|
404 |
+
"loss": 0.0562,
|
405 |
+
"step": 480
|
406 |
+
},
|
407 |
+
{
|
408 |
+
"epoch": 12.73,
|
409 |
+
"learning_rate": 3.923130193905817e-05,
|
410 |
+
"loss": 0.0678,
|
411 |
+
"step": 490
|
412 |
+
},
|
413 |
+
{
|
414 |
+
"epoch": 12.99,
|
415 |
+
"learning_rate": 3.885041551246538e-05,
|
416 |
+
"loss": 0.0708,
|
417 |
+
"step": 500
|
418 |
+
},
|
419 |
+
{
|
420 |
+
"epoch": 12.99,
|
421 |
+
"eval_accuracy": 0.8,
|
422 |
+
"eval_loss": 0.9578073024749756,
|
423 |
+
"eval_runtime": 1.0768,
|
424 |
+
"eval_samples_per_second": 55.72,
|
425 |
+
"eval_steps_per_second": 1.857,
|
426 |
+
"step": 500
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 13.25,
|
430 |
+
"learning_rate": 3.8469529085872574e-05,
|
431 |
+
"loss": 0.0429,
|
432 |
+
"step": 510
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 13.51,
|
436 |
+
"learning_rate": 3.808864265927978e-05,
|
437 |
+
"loss": 0.0553,
|
438 |
+
"step": 520
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 13.77,
|
442 |
+
"learning_rate": 3.770775623268698e-05,
|
443 |
+
"loss": 0.0562,
|
444 |
+
"step": 530
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 14.0,
|
448 |
+
"eval_accuracy": 0.8166666666666667,
|
449 |
+
"eval_loss": 1.0712392330169678,
|
450 |
+
"eval_runtime": 1.0829,
|
451 |
+
"eval_samples_per_second": 55.404,
|
452 |
+
"eval_steps_per_second": 1.847,
|
453 |
+
"step": 539
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 14.03,
|
457 |
+
"learning_rate": 3.732686980609418e-05,
|
458 |
+
"loss": 0.0448,
|
459 |
+
"step": 540
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 14.29,
|
463 |
+
"learning_rate": 3.694598337950139e-05,
|
464 |
+
"loss": 0.041,
|
465 |
+
"step": 550
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 14.55,
|
469 |
+
"learning_rate": 3.6565096952908585e-05,
|
470 |
+
"loss": 0.0577,
|
471 |
+
"step": 560
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 14.81,
|
475 |
+
"learning_rate": 3.618421052631579e-05,
|
476 |
+
"loss": 0.0467,
|
477 |
+
"step": 570
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 14.99,
|
481 |
+
"eval_accuracy": 0.8166666666666667,
|
482 |
+
"eval_loss": 1.058632731437683,
|
483 |
+
"eval_runtime": 1.2825,
|
484 |
+
"eval_samples_per_second": 46.783,
|
485 |
+
"eval_steps_per_second": 1.559,
|
486 |
+
"step": 577
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 15.06,
|
490 |
+
"learning_rate": 3.5803324099722994e-05,
|
491 |
+
"loss": 0.035,
|
492 |
+
"step": 580
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 15.32,
|
496 |
+
"learning_rate": 3.5422437673130195e-05,
|
497 |
+
"loss": 0.0572,
|
498 |
+
"step": 590
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 15.58,
|
502 |
+
"learning_rate": 3.5041551246537395e-05,
|
503 |
+
"loss": 0.0451,
|
504 |
+
"step": 600
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 15.84,
|
508 |
+
"learning_rate": 3.4660664819944596e-05,
|
509 |
+
"loss": 0.0445,
|
510 |
+
"step": 610
|
511 |
+
},
|
512 |
+
{
|
513 |
+
"epoch": 16.0,
|
514 |
+
"eval_accuracy": 0.7666666666666667,
|
515 |
+
"eval_loss": 1.20657479763031,
|
516 |
+
"eval_runtime": 1.2453,
|
517 |
+
"eval_samples_per_second": 48.181,
|
518 |
+
"eval_steps_per_second": 1.606,
|
519 |
+
"step": 616
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 16.1,
|
523 |
+
"learning_rate": 3.4279778393351804e-05,
|
524 |
+
"loss": 0.0338,
|
525 |
+
"step": 620
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 16.36,
|
529 |
+
"learning_rate": 3.3898891966759005e-05,
|
530 |
+
"loss": 0.0448,
|
531 |
+
"step": 630
|
532 |
+
},
|
533 |
+
{
|
534 |
+
"epoch": 16.62,
|
535 |
+
"learning_rate": 3.3518005540166206e-05,
|
536 |
+
"loss": 0.0488,
|
537 |
+
"step": 640
|
538 |
+
},
|
539 |
+
{
|
540 |
+
"epoch": 16.88,
|
541 |
+
"learning_rate": 3.313711911357341e-05,
|
542 |
+
"loss": 0.0474,
|
543 |
+
"step": 650
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 16.99,
|
547 |
+
"eval_accuracy": 0.75,
|
548 |
+
"eval_loss": 1.1863175630569458,
|
549 |
+
"eval_runtime": 1.0935,
|
550 |
+
"eval_samples_per_second": 54.87,
|
551 |
+
"eval_steps_per_second": 1.829,
|
552 |
+
"step": 654
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 17.14,
|
556 |
+
"learning_rate": 3.275623268698061e-05,
|
557 |
+
"loss": 0.0389,
|
558 |
+
"step": 660
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 17.4,
|
562 |
+
"learning_rate": 3.2375346260387815e-05,
|
563 |
+
"loss": 0.0442,
|
564 |
+
"step": 670
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 17.66,
|
568 |
+
"learning_rate": 3.1994459833795016e-05,
|
569 |
+
"loss": 0.0388,
|
570 |
+
"step": 680
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 17.92,
|
574 |
+
"learning_rate": 3.161357340720222e-05,
|
575 |
+
"loss": 0.0263,
|
576 |
+
"step": 690
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 18.0,
|
580 |
+
"eval_accuracy": 0.8166666666666667,
|
581 |
+
"eval_loss": 1.1207224130630493,
|
582 |
+
"eval_runtime": 1.0882,
|
583 |
+
"eval_samples_per_second": 55.137,
|
584 |
+
"eval_steps_per_second": 1.838,
|
585 |
+
"step": 693
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 18.18,
|
589 |
+
"learning_rate": 3.123268698060942e-05,
|
590 |
+
"loss": 0.0379,
|
591 |
+
"step": 700
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 18.44,
|
595 |
+
"learning_rate": 3.085180055401662e-05,
|
596 |
+
"loss": 0.0614,
|
597 |
+
"step": 710
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 18.7,
|
601 |
+
"learning_rate": 3.0470914127423827e-05,
|
602 |
+
"loss": 0.0367,
|
603 |
+
"step": 720
|
604 |
+
},
|
605 |
+
{
|
606 |
+
"epoch": 18.96,
|
607 |
+
"learning_rate": 3.0090027700831024e-05,
|
608 |
+
"loss": 0.0307,
|
609 |
+
"step": 730
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 18.99,
|
613 |
+
"eval_accuracy": 0.8166666666666667,
|
614 |
+
"eval_loss": 1.1812899112701416,
|
615 |
+
"eval_runtime": 1.0999,
|
616 |
+
"eval_samples_per_second": 54.549,
|
617 |
+
"eval_steps_per_second": 1.818,
|
618 |
+
"step": 731
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 19.22,
|
622 |
+
"learning_rate": 2.9709141274238225e-05,
|
623 |
+
"loss": 0.041,
|
624 |
+
"step": 740
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 19.48,
|
628 |
+
"learning_rate": 2.9328254847645433e-05,
|
629 |
+
"loss": 0.0367,
|
630 |
+
"step": 750
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 19.74,
|
634 |
+
"learning_rate": 2.894736842105263e-05,
|
635 |
+
"loss": 0.0311,
|
636 |
+
"step": 760
|
637 |
+
},
|
638 |
+
{
|
639 |
+
"epoch": 20.0,
|
640 |
+
"learning_rate": 2.8566481994459838e-05,
|
641 |
+
"loss": 0.0393,
|
642 |
+
"step": 770
|
643 |
+
},
|
644 |
+
{
|
645 |
+
"epoch": 20.0,
|
646 |
+
"eval_accuracy": 0.75,
|
647 |
+
"eval_loss": 1.3761276006698608,
|
648 |
+
"eval_runtime": 1.1025,
|
649 |
+
"eval_samples_per_second": 54.423,
|
650 |
+
"eval_steps_per_second": 1.814,
|
651 |
+
"step": 770
|
652 |
+
},
|
653 |
+
{
|
654 |
+
"epoch": 20.26,
|
655 |
+
"learning_rate": 2.8185595567867035e-05,
|
656 |
+
"loss": 0.0213,
|
657 |
+
"step": 780
|
658 |
+
},
|
659 |
+
{
|
660 |
+
"epoch": 20.52,
|
661 |
+
"learning_rate": 2.7804709141274236e-05,
|
662 |
+
"loss": 0.0206,
|
663 |
+
"step": 790
|
664 |
+
},
|
665 |
+
{
|
666 |
+
"epoch": 20.78,
|
667 |
+
"learning_rate": 2.742382271468144e-05,
|
668 |
+
"loss": 0.0475,
|
669 |
+
"step": 800
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 20.99,
|
673 |
+
"eval_accuracy": 0.7666666666666667,
|
674 |
+
"eval_loss": 1.3007997274398804,
|
675 |
+
"eval_runtime": 1.252,
|
676 |
+
"eval_samples_per_second": 47.924,
|
677 |
+
"eval_steps_per_second": 1.597,
|
678 |
+
"step": 808
|
679 |
+
},
|
680 |
+
{
|
681 |
+
"epoch": 21.04,
|
682 |
+
"learning_rate": 2.7042936288088645e-05,
|
683 |
+
"loss": 0.0447,
|
684 |
+
"step": 810
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 21.3,
|
688 |
+
"learning_rate": 2.6662049861495846e-05,
|
689 |
+
"loss": 0.0274,
|
690 |
+
"step": 820
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 21.56,
|
694 |
+
"learning_rate": 2.6281163434903047e-05,
|
695 |
+
"loss": 0.0369,
|
696 |
+
"step": 830
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 21.82,
|
700 |
+
"learning_rate": 2.590027700831025e-05,
|
701 |
+
"loss": 0.0215,
|
702 |
+
"step": 840
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 22.0,
|
706 |
+
"eval_accuracy": 0.7333333333333333,
|
707 |
+
"eval_loss": 1.2625234127044678,
|
708 |
+
"eval_runtime": 1.2982,
|
709 |
+
"eval_samples_per_second": 46.219,
|
710 |
+
"eval_steps_per_second": 1.541,
|
711 |
+
"step": 847
|
712 |
+
},
|
713 |
+
{
|
714 |
+
"epoch": 22.08,
|
715 |
+
"learning_rate": 2.5519390581717452e-05,
|
716 |
+
"loss": 0.0434,
|
717 |
+
"step": 850
|
718 |
+
},
|
719 |
+
{
|
720 |
+
"epoch": 22.34,
|
721 |
+
"learning_rate": 2.5138504155124656e-05,
|
722 |
+
"loss": 0.0338,
|
723 |
+
"step": 860
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 22.6,
|
727 |
+
"learning_rate": 2.4757617728531857e-05,
|
728 |
+
"loss": 0.0275,
|
729 |
+
"step": 870
|
730 |
+
},
|
731 |
+
{
|
732 |
+
"epoch": 22.86,
|
733 |
+
"learning_rate": 2.4376731301939058e-05,
|
734 |
+
"loss": 0.0311,
|
735 |
+
"step": 880
|
736 |
+
},
|
737 |
+
{
|
738 |
+
"epoch": 22.99,
|
739 |
+
"eval_accuracy": 0.8,
|
740 |
+
"eval_loss": 1.1508437395095825,
|
741 |
+
"eval_runtime": 0.9933,
|
742 |
+
"eval_samples_per_second": 60.407,
|
743 |
+
"eval_steps_per_second": 2.014,
|
744 |
+
"step": 885
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 23.12,
|
748 |
+
"learning_rate": 2.3995844875346262e-05,
|
749 |
+
"loss": 0.0332,
|
750 |
+
"step": 890
|
751 |
+
},
|
752 |
+
{
|
753 |
+
"epoch": 23.38,
|
754 |
+
"learning_rate": 2.3614958448753463e-05,
|
755 |
+
"loss": 0.0503,
|
756 |
+
"step": 900
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 23.64,
|
760 |
+
"learning_rate": 2.3234072022160667e-05,
|
761 |
+
"loss": 0.0267,
|
762 |
+
"step": 910
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 23.9,
|
766 |
+
"learning_rate": 2.2853185595567865e-05,
|
767 |
+
"loss": 0.027,
|
768 |
+
"step": 920
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 24.0,
|
772 |
+
"eval_accuracy": 0.7666666666666667,
|
773 |
+
"eval_loss": 1.3034749031066895,
|
774 |
+
"eval_runtime": 1.0568,
|
775 |
+
"eval_samples_per_second": 56.773,
|
776 |
+
"eval_steps_per_second": 1.892,
|
777 |
+
"step": 924
|
778 |
+
},
|
779 |
+
{
|
780 |
+
"epoch": 24.16,
|
781 |
+
"learning_rate": 2.247229916897507e-05,
|
782 |
+
"loss": 0.0309,
|
783 |
+
"step": 930
|
784 |
+
},
|
785 |
+
{
|
786 |
+
"epoch": 24.42,
|
787 |
+
"learning_rate": 2.2091412742382273e-05,
|
788 |
+
"loss": 0.0194,
|
789 |
+
"step": 940
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 24.68,
|
793 |
+
"learning_rate": 2.1710526315789474e-05,
|
794 |
+
"loss": 0.0296,
|
795 |
+
"step": 950
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 24.94,
|
799 |
+
"learning_rate": 2.132963988919668e-05,
|
800 |
+
"loss": 0.0251,
|
801 |
+
"step": 960
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 24.99,
|
805 |
+
"eval_accuracy": 0.7666666666666667,
|
806 |
+
"eval_loss": 1.2270464897155762,
|
807 |
+
"eval_runtime": 1.0482,
|
808 |
+
"eval_samples_per_second": 57.24,
|
809 |
+
"eval_steps_per_second": 1.908,
|
810 |
+
"step": 962
|
811 |
+
},
|
812 |
+
{
|
813 |
+
"epoch": 25.19,
|
814 |
+
"learning_rate": 2.094875346260388e-05,
|
815 |
+
"loss": 0.0262,
|
816 |
+
"step": 970
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 25.45,
|
820 |
+
"learning_rate": 2.056786703601108e-05,
|
821 |
+
"loss": 0.0253,
|
822 |
+
"step": 980
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 25.71,
|
826 |
+
"learning_rate": 2.018698060941828e-05,
|
827 |
+
"loss": 0.0196,
|
828 |
+
"step": 990
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 25.97,
|
832 |
+
"learning_rate": 1.9806094182825486e-05,
|
833 |
+
"loss": 0.0161,
|
834 |
+
"step": 1000
|
835 |
+
},
|
836 |
+
{
|
837 |
+
"epoch": 26.0,
|
838 |
+
"eval_accuracy": 0.8166666666666667,
|
839 |
+
"eval_loss": 1.1470165252685547,
|
840 |
+
"eval_runtime": 1.06,
|
841 |
+
"eval_samples_per_second": 56.602,
|
842 |
+
"eval_steps_per_second": 1.887,
|
843 |
+
"step": 1001
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 26.23,
|
847 |
+
"learning_rate": 1.942520775623269e-05,
|
848 |
+
"loss": 0.027,
|
849 |
+
"step": 1010
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 26.49,
|
853 |
+
"learning_rate": 1.904432132963989e-05,
|
854 |
+
"loss": 0.0262,
|
855 |
+
"step": 1020
|
856 |
+
},
|
857 |
+
{
|
858 |
+
"epoch": 26.75,
|
859 |
+
"learning_rate": 1.866343490304709e-05,
|
860 |
+
"loss": 0.0258,
|
861 |
+
"step": 1030
|
862 |
+
},
|
863 |
+
{
|
864 |
+
"epoch": 26.99,
|
865 |
+
"eval_accuracy": 0.8166666666666667,
|
866 |
+
"eval_loss": 1.1472843885421753,
|
867 |
+
"eval_runtime": 1.07,
|
868 |
+
"eval_samples_per_second": 56.076,
|
869 |
+
"eval_steps_per_second": 1.869,
|
870 |
+
"step": 1039
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 27.01,
|
874 |
+
"learning_rate": 1.8282548476454293e-05,
|
875 |
+
"loss": 0.0192,
|
876 |
+
"step": 1040
|
877 |
+
},
|
878 |
+
{
|
879 |
+
"epoch": 27.27,
|
880 |
+
"learning_rate": 1.7901662049861497e-05,
|
881 |
+
"loss": 0.0165,
|
882 |
+
"step": 1050
|
883 |
+
},
|
884 |
+
{
|
885 |
+
"epoch": 27.53,
|
886 |
+
"learning_rate": 1.7520775623268698e-05,
|
887 |
+
"loss": 0.0209,
|
888 |
+
"step": 1060
|
889 |
+
},
|
890 |
+
{
|
891 |
+
"epoch": 27.79,
|
892 |
+
"learning_rate": 1.7139889196675902e-05,
|
893 |
+
"loss": 0.0142,
|
894 |
+
"step": 1070
|
895 |
+
},
|
896 |
+
{
|
897 |
+
"epoch": 28.0,
|
898 |
+
"eval_accuracy": 0.7666666666666667,
|
899 |
+
"eval_loss": 1.2325979471206665,
|
900 |
+
"eval_runtime": 1.1661,
|
901 |
+
"eval_samples_per_second": 51.455,
|
902 |
+
"eval_steps_per_second": 1.715,
|
903 |
+
"step": 1078
|
904 |
+
},
|
905 |
+
{
|
906 |
+
"epoch": 28.05,
|
907 |
+
"learning_rate": 1.6759002770083103e-05,
|
908 |
+
"loss": 0.0178,
|
909 |
+
"step": 1080
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 28.31,
|
913 |
+
"learning_rate": 1.6378116343490304e-05,
|
914 |
+
"loss": 0.0214,
|
915 |
+
"step": 1090
|
916 |
+
},
|
917 |
+
{
|
918 |
+
"epoch": 28.57,
|
919 |
+
"learning_rate": 1.5997229916897508e-05,
|
920 |
+
"loss": 0.0236,
|
921 |
+
"step": 1100
|
922 |
+
},
|
923 |
+
{
|
924 |
+
"epoch": 28.83,
|
925 |
+
"learning_rate": 1.561634349030471e-05,
|
926 |
+
"loss": 0.0151,
|
927 |
+
"step": 1110
|
928 |
+
},
|
929 |
+
{
|
930 |
+
"epoch": 28.99,
|
931 |
+
"eval_accuracy": 0.7666666666666667,
|
932 |
+
"eval_loss": 1.3977586030960083,
|
933 |
+
"eval_runtime": 1.0743,
|
934 |
+
"eval_samples_per_second": 55.848,
|
935 |
+
"eval_steps_per_second": 1.862,
|
936 |
+
"step": 1116
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 29.09,
|
940 |
+
"learning_rate": 1.5235457063711913e-05,
|
941 |
+
"loss": 0.0259,
|
942 |
+
"step": 1120
|
943 |
+
},
|
944 |
+
{
|
945 |
+
"epoch": 29.35,
|
946 |
+
"learning_rate": 1.4854570637119112e-05,
|
947 |
+
"loss": 0.0181,
|
948 |
+
"step": 1130
|
949 |
+
},
|
950 |
+
{
|
951 |
+
"epoch": 29.61,
|
952 |
+
"learning_rate": 1.4473684210526315e-05,
|
953 |
+
"loss": 0.0157,
|
954 |
+
"step": 1140
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 29.87,
|
958 |
+
"learning_rate": 1.4092797783933518e-05,
|
959 |
+
"loss": 0.021,
|
960 |
+
"step": 1150
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 30.0,
|
964 |
+
"eval_accuracy": 0.8333333333333334,
|
965 |
+
"eval_loss": 1.2002999782562256,
|
966 |
+
"eval_runtime": 1.0917,
|
967 |
+
"eval_samples_per_second": 54.958,
|
968 |
+
"eval_steps_per_second": 1.832,
|
969 |
+
"step": 1155
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 30.13,
|
973 |
+
"learning_rate": 1.371191135734072e-05,
|
974 |
+
"loss": 0.0184,
|
975 |
+
"step": 1160
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 30.39,
|
979 |
+
"learning_rate": 1.3331024930747923e-05,
|
980 |
+
"loss": 0.0163,
|
981 |
+
"step": 1170
|
982 |
+
},
|
983 |
+
{
|
984 |
+
"epoch": 30.65,
|
985 |
+
"learning_rate": 1.2950138504155125e-05,
|
986 |
+
"loss": 0.0254,
|
987 |
+
"step": 1180
|
988 |
+
},
|
989 |
+
{
|
990 |
+
"epoch": 30.91,
|
991 |
+
"learning_rate": 1.2569252077562328e-05,
|
992 |
+
"loss": 0.0158,
|
993 |
+
"step": 1190
|
994 |
+
},
|
995 |
+
{
|
996 |
+
"epoch": 30.99,
|
997 |
+
"eval_accuracy": 0.7666666666666667,
|
998 |
+
"eval_loss": 1.2488106489181519,
|
999 |
+
"eval_runtime": 1.044,
|
1000 |
+
"eval_samples_per_second": 57.472,
|
1001 |
+
"eval_steps_per_second": 1.916,
|
1002 |
+
"step": 1193
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 31.17,
|
1006 |
+
"learning_rate": 1.2188365650969529e-05,
|
1007 |
+
"loss": 0.0232,
|
1008 |
+
"step": 1200
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 31.43,
|
1012 |
+
"learning_rate": 1.1807479224376732e-05,
|
1013 |
+
"loss": 0.0099,
|
1014 |
+
"step": 1210
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 31.69,
|
1018 |
+
"learning_rate": 1.1426592797783932e-05,
|
1019 |
+
"loss": 0.0211,
|
1020 |
+
"step": 1220
|
1021 |
+
},
|
1022 |
+
{
|
1023 |
+
"epoch": 31.95,
|
1024 |
+
"learning_rate": 1.1045706371191137e-05,
|
1025 |
+
"loss": 0.0163,
|
1026 |
+
"step": 1230
|
1027 |
+
},
|
1028 |
+
{
|
1029 |
+
"epoch": 32.0,
|
1030 |
+
"eval_accuracy": 0.75,
|
1031 |
+
"eval_loss": 1.3231933116912842,
|
1032 |
+
"eval_runtime": 1.0878,
|
1033 |
+
"eval_samples_per_second": 55.156,
|
1034 |
+
"eval_steps_per_second": 1.839,
|
1035 |
+
"step": 1232
|
1036 |
+
},
|
1037 |
+
{
|
1038 |
+
"epoch": 32.21,
|
1039 |
+
"learning_rate": 1.066481994459834e-05,
|
1040 |
+
"loss": 0.0127,
|
1041 |
+
"step": 1240
|
1042 |
+
},
|
1043 |
+
{
|
1044 |
+
"epoch": 32.47,
|
1045 |
+
"learning_rate": 1.028393351800554e-05,
|
1046 |
+
"loss": 0.0206,
|
1047 |
+
"step": 1250
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 32.73,
|
1051 |
+
"learning_rate": 9.903047091412743e-06,
|
1052 |
+
"loss": 0.0232,
|
1053 |
+
"step": 1260
|
1054 |
+
},
|
1055 |
+
{
|
1056 |
+
"epoch": 32.99,
|
1057 |
+
"learning_rate": 9.522160664819945e-06,
|
1058 |
+
"loss": 0.0143,
|
1059 |
+
"step": 1270
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 32.99,
|
1063 |
+
"eval_accuracy": 0.8,
|
1064 |
+
"eval_loss": 1.246673583984375,
|
1065 |
+
"eval_runtime": 1.0784,
|
1066 |
+
"eval_samples_per_second": 55.637,
|
1067 |
+
"eval_steps_per_second": 1.855,
|
1068 |
+
"step": 1270
|
1069 |
+
},
|
1070 |
+
{
|
1071 |
+
"epoch": 33.25,
|
1072 |
+
"learning_rate": 9.141274238227146e-06,
|
1073 |
+
"loss": 0.009,
|
1074 |
+
"step": 1280
|
1075 |
+
},
|
1076 |
+
{
|
1077 |
+
"epoch": 33.51,
|
1078 |
+
"learning_rate": 8.760387811634349e-06,
|
1079 |
+
"loss": 0.0208,
|
1080 |
+
"step": 1290
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 33.77,
|
1084 |
+
"learning_rate": 8.379501385041551e-06,
|
1085 |
+
"loss": 0.02,
|
1086 |
+
"step": 1300
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 34.0,
|
1090 |
+
"eval_accuracy": 0.7833333333333333,
|
1091 |
+
"eval_loss": 1.3175911903381348,
|
1092 |
+
"eval_runtime": 1.0636,
|
1093 |
+
"eval_samples_per_second": 56.415,
|
1094 |
+
"eval_steps_per_second": 1.88,
|
1095 |
+
"step": 1309
|
1096 |
+
},
|
1097 |
+
{
|
1098 |
+
"epoch": 34.03,
|
1099 |
+
"learning_rate": 7.998614958448754e-06,
|
1100 |
+
"loss": 0.0179,
|
1101 |
+
"step": 1310
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 34.29,
|
1105 |
+
"learning_rate": 7.617728531855957e-06,
|
1106 |
+
"loss": 0.0148,
|
1107 |
+
"step": 1320
|
1108 |
+
},
|
1109 |
+
{
|
1110 |
+
"epoch": 34.55,
|
1111 |
+
"learning_rate": 7.2368421052631575e-06,
|
1112 |
+
"loss": 0.0168,
|
1113 |
+
"step": 1330
|
1114 |
+
},
|
1115 |
+
{
|
1116 |
+
"epoch": 34.81,
|
1117 |
+
"learning_rate": 6.85595567867036e-06,
|
1118 |
+
"loss": 0.0128,
|
1119 |
+
"step": 1340
|
1120 |
+
},
|
1121 |
+
{
|
1122 |
+
"epoch": 34.99,
|
1123 |
+
"eval_accuracy": 0.7666666666666667,
|
1124 |
+
"eval_loss": 1.3082655668258667,
|
1125 |
+
"eval_runtime": 1.0769,
|
1126 |
+
"eval_samples_per_second": 55.716,
|
1127 |
+
"eval_steps_per_second": 1.857,
|
1128 |
+
"step": 1347
|
1129 |
+
},
|
1130 |
+
{
|
1131 |
+
"epoch": 35.06,
|
1132 |
+
"learning_rate": 6.475069252077563e-06,
|
1133 |
+
"loss": 0.018,
|
1134 |
+
"step": 1350
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"epoch": 35.32,
|
1138 |
+
"learning_rate": 6.0941828254847645e-06,
|
1139 |
+
"loss": 0.0123,
|
1140 |
+
"step": 1360
|
1141 |
+
},
|
1142 |
+
{
|
1143 |
+
"epoch": 35.58,
|
1144 |
+
"learning_rate": 5.713296398891966e-06,
|
1145 |
+
"loss": 0.0124,
|
1146 |
+
"step": 1370
|
1147 |
+
},
|
1148 |
+
{
|
1149 |
+
"epoch": 35.84,
|
1150 |
+
"learning_rate": 5.33240997229917e-06,
|
1151 |
+
"loss": 0.0144,
|
1152 |
+
"step": 1380
|
1153 |
+
},
|
1154 |
+
{
|
1155 |
+
"epoch": 36.0,
|
1156 |
+
"eval_accuracy": 0.7666666666666667,
|
1157 |
+
"eval_loss": 1.3079514503479004,
|
1158 |
+
"eval_runtime": 1.0808,
|
1159 |
+
"eval_samples_per_second": 55.512,
|
1160 |
+
"eval_steps_per_second": 1.85,
|
1161 |
+
"step": 1386
|
1162 |
+
},
|
1163 |
+
{
|
1164 |
+
"epoch": 36.1,
|
1165 |
+
"learning_rate": 4.951523545706371e-06,
|
1166 |
+
"loss": 0.0156,
|
1167 |
+
"step": 1390
|
1168 |
+
},
|
1169 |
+
{
|
1170 |
+
"epoch": 36.36,
|
1171 |
+
"learning_rate": 4.570637119113573e-06,
|
1172 |
+
"loss": 0.0124,
|
1173 |
+
"step": 1400
|
1174 |
+
},
|
1175 |
+
{
|
1176 |
+
"epoch": 36.62,
|
1177 |
+
"learning_rate": 4.189750692520776e-06,
|
1178 |
+
"loss": 0.0134,
|
1179 |
+
"step": 1410
|
1180 |
+
},
|
1181 |
+
{
|
1182 |
+
"epoch": 36.88,
|
1183 |
+
"learning_rate": 3.8088642659279783e-06,
|
1184 |
+
"loss": 0.0109,
|
1185 |
+
"step": 1420
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 36.99,
|
1189 |
+
"eval_accuracy": 0.8,
|
1190 |
+
"eval_loss": 1.2998783588409424,
|
1191 |
+
"eval_runtime": 1.0406,
|
1192 |
+
"eval_samples_per_second": 57.661,
|
1193 |
+
"eval_steps_per_second": 1.922,
|
1194 |
+
"step": 1424
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 37.14,
|
1198 |
+
"learning_rate": 3.42797783933518e-06,
|
1199 |
+
"loss": 0.0147,
|
1200 |
+
"step": 1430
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 37.4,
|
1204 |
+
"learning_rate": 3.0470914127423822e-06,
|
1205 |
+
"loss": 0.0116,
|
1206 |
+
"step": 1440
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 37.66,
|
1210 |
+
"learning_rate": 2.666204986149585e-06,
|
1211 |
+
"loss": 0.0102,
|
1212 |
+
"step": 1450
|
1213 |
+
},
|
1214 |
+
{
|
1215 |
+
"epoch": 37.92,
|
1216 |
+
"learning_rate": 2.2853185595567866e-06,
|
1217 |
+
"loss": 0.0082,
|
1218 |
+
"step": 1460
|
1219 |
+
},
|
1220 |
+
{
|
1221 |
+
"epoch": 38.0,
|
1222 |
+
"eval_accuracy": 0.8,
|
1223 |
+
"eval_loss": 1.2717970609664917,
|
1224 |
+
"eval_runtime": 1.0837,
|
1225 |
+
"eval_samples_per_second": 55.367,
|
1226 |
+
"eval_steps_per_second": 1.846,
|
1227 |
+
"step": 1463
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 38.18,
|
1231 |
+
"learning_rate": 1.9044321329639892e-06,
|
1232 |
+
"loss": 0.0103,
|
1233 |
+
"step": 1470
|
1234 |
+
},
|
1235 |
+
{
|
1236 |
+
"epoch": 38.44,
|
1237 |
+
"learning_rate": 1.5235457063711911e-06,
|
1238 |
+
"loss": 0.0207,
|
1239 |
+
"step": 1480
|
1240 |
+
},
|
1241 |
+
{
|
1242 |
+
"epoch": 38.7,
|
1243 |
+
"learning_rate": 1.1426592797783933e-06,
|
1244 |
+
"loss": 0.0098,
|
1245 |
+
"step": 1490
|
1246 |
+
},
|
1247 |
+
{
|
1248 |
+
"epoch": 38.96,
|
1249 |
+
"learning_rate": 7.617728531855956e-07,
|
1250 |
+
"loss": 0.0064,
|
1251 |
+
"step": 1500
|
1252 |
+
},
|
1253 |
+
{
|
1254 |
+
"epoch": 38.99,
|
1255 |
+
"eval_accuracy": 0.7666666666666667,
|
1256 |
+
"eval_loss": 1.2587716579437256,
|
1257 |
+
"eval_runtime": 1.1594,
|
1258 |
+
"eval_samples_per_second": 51.749,
|
1259 |
+
"eval_steps_per_second": 1.725,
|
1260 |
+
"step": 1501
|
1261 |
+
},
|
1262 |
+
{
|
1263 |
+
"epoch": 39.22,
|
1264 |
+
"learning_rate": 3.808864265927978e-07,
|
1265 |
+
"loss": 0.0137,
|
1266 |
+
"step": 1510
|
1267 |
+
},
|
1268 |
+
{
|
1269 |
+
"epoch": 39.48,
|
1270 |
+
"learning_rate": 0.0,
|
1271 |
+
"loss": 0.0097,
|
1272 |
+
"step": 1520
|
1273 |
+
},
|
1274 |
+
{
|
1275 |
+
"epoch": 39.48,
|
1276 |
+
"eval_accuracy": 0.7666666666666667,
|
1277 |
+
"eval_loss": 1.2597320079803467,
|
1278 |
+
"eval_runtime": 1.105,
|
1279 |
+
"eval_samples_per_second": 54.301,
|
1280 |
+
"eval_steps_per_second": 1.81,
|
1281 |
+
"step": 1520
|
1282 |
+
},
|
1283 |
+
{
|
1284 |
+
"epoch": 39.48,
|
1285 |
+
"step": 1520,
|
1286 |
+
"total_flos": 1.4992488987811873e+19,
|
1287 |
+
"train_loss": 0.11935894162835259,
|
1288 |
+
"train_runtime": 3007.0998,
|
1289 |
+
"train_samples_per_second": 65.179,
|
1290 |
+
"train_steps_per_second": 0.505
|
1291 |
+
}
|
1292 |
+
],
|
1293 |
+
"logging_steps": 10,
|
1294 |
+
"max_steps": 1520,
|
1295 |
+
"num_input_tokens_seen": 0,
|
1296 |
+
"num_train_epochs": 40,
|
1297 |
+
"save_steps": 500,
|
1298 |
+
"total_flos": 1.4992488987811873e+19,
|
1299 |
+
"train_batch_size": 32,
|
1300 |
+
"trial_name": null,
|
1301 |
+
"trial_params": null
|
1302 |
+
}
|