Commit
•
015e9c0
1
Parent(s):
de5e36f
End of training
Browse files- README.md +3 -3
- all_results.json +12 -0
- eval_results.json +8 -0
- runs/Jun23_14-39-12_DESKTOP-SKBE9FB/events.out.tfevents.1719176029.DESKTOP-SKBE9FB.16296.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +981 -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: 0.
|
36 |
-
- Accuracy: 0.
|
37 |
|
38 |
## Model description
|
39 |
|
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
+
value: 0.8695652173913043
|
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: 0.6456
|
36 |
+
- Accuracy: 0.8696
|
37 |
|
38 |
## Model description
|
39 |
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 38.83,
|
3 |
+
"eval_accuracy": 0.8695652173913043,
|
4 |
+
"eval_loss": 0.6455863118171692,
|
5 |
+
"eval_runtime": 0.9822,
|
6 |
+
"eval_samples_per_second": 46.833,
|
7 |
+
"eval_steps_per_second": 6.109,
|
8 |
+
"train_loss": 0.2159135382026434,
|
9 |
+
"train_runtime": 637.3873,
|
10 |
+
"train_samples_per_second": 51.397,
|
11 |
+
"train_steps_per_second": 1.569
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 38.83,
|
3 |
+
"eval_accuracy": 0.8695652173913043,
|
4 |
+
"eval_loss": 0.6455863118171692,
|
5 |
+
"eval_runtime": 0.9822,
|
6 |
+
"eval_samples_per_second": 46.833,
|
7 |
+
"eval_steps_per_second": 6.109
|
8 |
+
}
|
runs/Jun23_14-39-12_DESKTOP-SKBE9FB/events.out.tfevents.1719176029.DESKTOP-SKBE9FB.16296.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8e816d8ee1d2541ec4717250e5448c2d816a002efa7103e16a26b553c9dadc15
|
3 |
+
size 411
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 38.83,
|
3 |
+
"train_loss": 0.2159135382026434,
|
4 |
+
"train_runtime": 637.3873,
|
5 |
+
"train_samples_per_second": 51.397,
|
6 |
+
"train_steps_per_second": 1.569
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,981 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.8695652173913043,
|
3 |
+
"best_model_checkpoint": "vit-base-patch16-224-ve-U13b-80RX\\checkpoint-540",
|
4 |
+
"epoch": 38.83495145631068,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.39,
|
13 |
+
"learning_rate": 1.1000000000000001e-05,
|
14 |
+
"loss": 1.386,
|
15 |
+
"step": 10
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.78,
|
19 |
+
"learning_rate": 2.2000000000000003e-05,
|
20 |
+
"loss": 1.3805,
|
21 |
+
"step": 20
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.97,
|
25 |
+
"eval_accuracy": 0.5,
|
26 |
+
"eval_loss": 1.3566619157791138,
|
27 |
+
"eval_runtime": 0.9332,
|
28 |
+
"eval_samples_per_second": 49.292,
|
29 |
+
"eval_steps_per_second": 6.429,
|
30 |
+
"step": 25
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 1.17,
|
34 |
+
"learning_rate": 3.3e-05,
|
35 |
+
"loss": 1.3523,
|
36 |
+
"step": 30
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 1.55,
|
40 |
+
"learning_rate": 4.4000000000000006e-05,
|
41 |
+
"loss": 1.283,
|
42 |
+
"step": 40
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 1.94,
|
46 |
+
"learning_rate": 5.5e-05,
|
47 |
+
"loss": 1.1576,
|
48 |
+
"step": 50
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 1.98,
|
52 |
+
"eval_accuracy": 0.43478260869565216,
|
53 |
+
"eval_loss": 1.1359825134277344,
|
54 |
+
"eval_runtime": 0.7817,
|
55 |
+
"eval_samples_per_second": 58.846,
|
56 |
+
"eval_steps_per_second": 7.676,
|
57 |
+
"step": 51
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 2.33,
|
61 |
+
"learning_rate": 5.4421052631578945e-05,
|
62 |
+
"loss": 1.0631,
|
63 |
+
"step": 60
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 2.72,
|
67 |
+
"learning_rate": 5.38421052631579e-05,
|
68 |
+
"loss": 0.9331,
|
69 |
+
"step": 70
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 2.99,
|
73 |
+
"eval_accuracy": 0.8043478260869565,
|
74 |
+
"eval_loss": 0.8530930876731873,
|
75 |
+
"eval_runtime": 0.8277,
|
76 |
+
"eval_samples_per_second": 55.576,
|
77 |
+
"eval_steps_per_second": 7.249,
|
78 |
+
"step": 77
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 3.11,
|
82 |
+
"learning_rate": 5.3263157894736844e-05,
|
83 |
+
"loss": 0.8206,
|
84 |
+
"step": 80
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 3.5,
|
88 |
+
"learning_rate": 5.268421052631579e-05,
|
89 |
+
"loss": 0.7139,
|
90 |
+
"step": 90
|
91 |
+
},
|
92 |
+
{
|
93 |
+
"epoch": 3.88,
|
94 |
+
"learning_rate": 5.210526315789474e-05,
|
95 |
+
"loss": 0.6657,
|
96 |
+
"step": 100
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 4.0,
|
100 |
+
"eval_accuracy": 0.782608695652174,
|
101 |
+
"eval_loss": 0.6856333613395691,
|
102 |
+
"eval_runtime": 0.7852,
|
103 |
+
"eval_samples_per_second": 58.586,
|
104 |
+
"eval_steps_per_second": 7.642,
|
105 |
+
"step": 103
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 4.27,
|
109 |
+
"learning_rate": 5.152631578947369e-05,
|
110 |
+
"loss": 0.6225,
|
111 |
+
"step": 110
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 4.66,
|
115 |
+
"learning_rate": 5.0947368421052636e-05,
|
116 |
+
"loss": 0.5642,
|
117 |
+
"step": 120
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 4.97,
|
121 |
+
"eval_accuracy": 0.782608695652174,
|
122 |
+
"eval_loss": 0.6162217855453491,
|
123 |
+
"eval_runtime": 0.8007,
|
124 |
+
"eval_samples_per_second": 57.45,
|
125 |
+
"eval_steps_per_second": 7.494,
|
126 |
+
"step": 128
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 5.05,
|
130 |
+
"learning_rate": 5.036842105263158e-05,
|
131 |
+
"loss": 0.4835,
|
132 |
+
"step": 130
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"epoch": 5.44,
|
136 |
+
"learning_rate": 4.978947368421053e-05,
|
137 |
+
"loss": 0.4062,
|
138 |
+
"step": 140
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 5.83,
|
142 |
+
"learning_rate": 4.921052631578948e-05,
|
143 |
+
"loss": 0.3632,
|
144 |
+
"step": 150
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 5.98,
|
148 |
+
"eval_accuracy": 0.8043478260869565,
|
149 |
+
"eval_loss": 0.590190589427948,
|
150 |
+
"eval_runtime": 0.7922,
|
151 |
+
"eval_samples_per_second": 58.068,
|
152 |
+
"eval_steps_per_second": 7.574,
|
153 |
+
"step": 154
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 6.21,
|
157 |
+
"learning_rate": 4.863157894736842e-05,
|
158 |
+
"loss": 0.3261,
|
159 |
+
"step": 160
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 6.6,
|
163 |
+
"learning_rate": 4.805263157894737e-05,
|
164 |
+
"loss": 0.2892,
|
165 |
+
"step": 170
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 6.99,
|
169 |
+
"learning_rate": 4.747368421052632e-05,
|
170 |
+
"loss": 0.3384,
|
171 |
+
"step": 180
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 6.99,
|
175 |
+
"eval_accuracy": 0.8043478260869565,
|
176 |
+
"eval_loss": 0.4995403587818146,
|
177 |
+
"eval_runtime": 0.7967,
|
178 |
+
"eval_samples_per_second": 57.739,
|
179 |
+
"eval_steps_per_second": 7.531,
|
180 |
+
"step": 180
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 7.38,
|
184 |
+
"learning_rate": 4.6894736842105264e-05,
|
185 |
+
"loss": 0.2762,
|
186 |
+
"step": 190
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 7.77,
|
190 |
+
"learning_rate": 4.631578947368421e-05,
|
191 |
+
"loss": 0.2261,
|
192 |
+
"step": 200
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 8.0,
|
196 |
+
"eval_accuracy": 0.7608695652173914,
|
197 |
+
"eval_loss": 0.6854467988014221,
|
198 |
+
"eval_runtime": 0.7832,
|
199 |
+
"eval_samples_per_second": 58.734,
|
200 |
+
"eval_steps_per_second": 7.661,
|
201 |
+
"step": 206
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 8.16,
|
205 |
+
"learning_rate": 4.573684210526316e-05,
|
206 |
+
"loss": 0.259,
|
207 |
+
"step": 210
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 8.54,
|
211 |
+
"learning_rate": 4.5157894736842106e-05,
|
212 |
+
"loss": 0.1924,
|
213 |
+
"step": 220
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 8.93,
|
217 |
+
"learning_rate": 4.457894736842105e-05,
|
218 |
+
"loss": 0.2066,
|
219 |
+
"step": 230
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 8.97,
|
223 |
+
"eval_accuracy": 0.782608695652174,
|
224 |
+
"eval_loss": 0.5605289936065674,
|
225 |
+
"eval_runtime": 1.1543,
|
226 |
+
"eval_samples_per_second": 39.852,
|
227 |
+
"eval_steps_per_second": 5.198,
|
228 |
+
"step": 231
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 9.32,
|
232 |
+
"learning_rate": 4.4000000000000006e-05,
|
233 |
+
"loss": 0.1525,
|
234 |
+
"step": 240
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 9.71,
|
238 |
+
"learning_rate": 4.342105263157895e-05,
|
239 |
+
"loss": 0.1635,
|
240 |
+
"step": 250
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 9.98,
|
244 |
+
"eval_accuracy": 0.7391304347826086,
|
245 |
+
"eval_loss": 0.720944881439209,
|
246 |
+
"eval_runtime": 0.7752,
|
247 |
+
"eval_samples_per_second": 59.34,
|
248 |
+
"eval_steps_per_second": 7.74,
|
249 |
+
"step": 257
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 10.1,
|
253 |
+
"learning_rate": 4.28421052631579e-05,
|
254 |
+
"loss": 0.1457,
|
255 |
+
"step": 260
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 10.49,
|
259 |
+
"learning_rate": 4.226315789473684e-05,
|
260 |
+
"loss": 0.1363,
|
261 |
+
"step": 270
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 10.87,
|
265 |
+
"learning_rate": 4.168421052631579e-05,
|
266 |
+
"loss": 0.1829,
|
267 |
+
"step": 280
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 10.99,
|
271 |
+
"eval_accuracy": 0.6956521739130435,
|
272 |
+
"eval_loss": 0.9292640089988708,
|
273 |
+
"eval_runtime": 0.8392,
|
274 |
+
"eval_samples_per_second": 54.814,
|
275 |
+
"eval_steps_per_second": 7.15,
|
276 |
+
"step": 283
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 11.26,
|
280 |
+
"learning_rate": 4.110526315789474e-05,
|
281 |
+
"loss": 0.1439,
|
282 |
+
"step": 290
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 11.65,
|
286 |
+
"learning_rate": 4.0526315789473684e-05,
|
287 |
+
"loss": 0.1455,
|
288 |
+
"step": 300
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 12.0,
|
292 |
+
"eval_accuracy": 0.782608695652174,
|
293 |
+
"eval_loss": 0.5999487042427063,
|
294 |
+
"eval_runtime": 0.7912,
|
295 |
+
"eval_samples_per_second": 58.141,
|
296 |
+
"eval_steps_per_second": 7.584,
|
297 |
+
"step": 309
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 12.04,
|
301 |
+
"learning_rate": 3.994736842105263e-05,
|
302 |
+
"loss": 0.1154,
|
303 |
+
"step": 310
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 12.43,
|
307 |
+
"learning_rate": 3.936842105263158e-05,
|
308 |
+
"loss": 0.1461,
|
309 |
+
"step": 320
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 12.82,
|
313 |
+
"learning_rate": 3.8789473684210526e-05,
|
314 |
+
"loss": 0.1072,
|
315 |
+
"step": 330
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 12.97,
|
319 |
+
"eval_accuracy": 0.782608695652174,
|
320 |
+
"eval_loss": 0.7919036149978638,
|
321 |
+
"eval_runtime": 0.8292,
|
322 |
+
"eval_samples_per_second": 55.476,
|
323 |
+
"eval_steps_per_second": 7.236,
|
324 |
+
"step": 334
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 13.2,
|
328 |
+
"learning_rate": 3.8210526315789476e-05,
|
329 |
+
"loss": 0.1154,
|
330 |
+
"step": 340
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 13.59,
|
334 |
+
"learning_rate": 3.7631578947368425e-05,
|
335 |
+
"loss": 0.1484,
|
336 |
+
"step": 350
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 13.98,
|
340 |
+
"learning_rate": 3.705263157894737e-05,
|
341 |
+
"loss": 0.1059,
|
342 |
+
"step": 360
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 13.98,
|
346 |
+
"eval_accuracy": 0.8043478260869565,
|
347 |
+
"eval_loss": 0.7782349586486816,
|
348 |
+
"eval_runtime": 0.7972,
|
349 |
+
"eval_samples_per_second": 57.703,
|
350 |
+
"eval_steps_per_second": 7.526,
|
351 |
+
"step": 360
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 14.37,
|
355 |
+
"learning_rate": 3.647368421052631e-05,
|
356 |
+
"loss": 0.0885,
|
357 |
+
"step": 370
|
358 |
+
},
|
359 |
+
{
|
360 |
+
"epoch": 14.76,
|
361 |
+
"learning_rate": 3.589473684210527e-05,
|
362 |
+
"loss": 0.0971,
|
363 |
+
"step": 380
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 14.99,
|
367 |
+
"eval_accuracy": 0.8043478260869565,
|
368 |
+
"eval_loss": 0.8248986005783081,
|
369 |
+
"eval_runtime": 0.7922,
|
370 |
+
"eval_samples_per_second": 58.067,
|
371 |
+
"eval_steps_per_second": 7.574,
|
372 |
+
"step": 386
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 15.15,
|
376 |
+
"learning_rate": 3.531578947368421e-05,
|
377 |
+
"loss": 0.1269,
|
378 |
+
"step": 390
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 15.53,
|
382 |
+
"learning_rate": 3.473684210526316e-05,
|
383 |
+
"loss": 0.1137,
|
384 |
+
"step": 400
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 15.92,
|
388 |
+
"learning_rate": 3.415789473684211e-05,
|
389 |
+
"loss": 0.0456,
|
390 |
+
"step": 410
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 16.0,
|
394 |
+
"eval_accuracy": 0.782608695652174,
|
395 |
+
"eval_loss": 0.7964589595794678,
|
396 |
+
"eval_runtime": 0.7832,
|
397 |
+
"eval_samples_per_second": 58.735,
|
398 |
+
"eval_steps_per_second": 7.661,
|
399 |
+
"step": 412
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 16.31,
|
403 |
+
"learning_rate": 3.357894736842105e-05,
|
404 |
+
"loss": 0.127,
|
405 |
+
"step": 420
|
406 |
+
},
|
407 |
+
{
|
408 |
+
"epoch": 16.7,
|
409 |
+
"learning_rate": 3.3e-05,
|
410 |
+
"loss": 0.0483,
|
411 |
+
"step": 430
|
412 |
+
},
|
413 |
+
{
|
414 |
+
"epoch": 16.97,
|
415 |
+
"eval_accuracy": 0.8260869565217391,
|
416 |
+
"eval_loss": 0.7163321375846863,
|
417 |
+
"eval_runtime": 0.8357,
|
418 |
+
"eval_samples_per_second": 55.044,
|
419 |
+
"eval_steps_per_second": 7.18,
|
420 |
+
"step": 437
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 17.09,
|
424 |
+
"learning_rate": 3.2421052631578945e-05,
|
425 |
+
"loss": 0.1415,
|
426 |
+
"step": 440
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 17.48,
|
430 |
+
"learning_rate": 3.1842105263157895e-05,
|
431 |
+
"loss": 0.1081,
|
432 |
+
"step": 450
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 17.86,
|
436 |
+
"learning_rate": 3.1263157894736845e-05,
|
437 |
+
"loss": 0.0832,
|
438 |
+
"step": 460
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 17.98,
|
442 |
+
"eval_accuracy": 0.782608695652174,
|
443 |
+
"eval_loss": 0.8122026324272156,
|
444 |
+
"eval_runtime": 0.7852,
|
445 |
+
"eval_samples_per_second": 58.584,
|
446 |
+
"eval_steps_per_second": 7.641,
|
447 |
+
"step": 463
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 18.25,
|
451 |
+
"learning_rate": 3.068421052631579e-05,
|
452 |
+
"loss": 0.0629,
|
453 |
+
"step": 470
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 18.64,
|
457 |
+
"learning_rate": 3.010526315789474e-05,
|
458 |
+
"loss": 0.055,
|
459 |
+
"step": 480
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 18.99,
|
463 |
+
"eval_accuracy": 0.782608695652174,
|
464 |
+
"eval_loss": 0.8250147104263306,
|
465 |
+
"eval_runtime": 0.7762,
|
466 |
+
"eval_samples_per_second": 59.264,
|
467 |
+
"eval_steps_per_second": 7.73,
|
468 |
+
"step": 489
|
469 |
+
},
|
470 |
+
{
|
471 |
+
"epoch": 19.03,
|
472 |
+
"learning_rate": 2.9526315789473684e-05,
|
473 |
+
"loss": 0.1095,
|
474 |
+
"step": 490
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 19.42,
|
478 |
+
"learning_rate": 2.894736842105263e-05,
|
479 |
+
"loss": 0.07,
|
480 |
+
"step": 500
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 19.81,
|
484 |
+
"learning_rate": 2.8368421052631583e-05,
|
485 |
+
"loss": 0.0753,
|
486 |
+
"step": 510
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 20.0,
|
490 |
+
"eval_accuracy": 0.8478260869565217,
|
491 |
+
"eval_loss": 0.6865554451942444,
|
492 |
+
"eval_runtime": 0.8217,
|
493 |
+
"eval_samples_per_second": 55.981,
|
494 |
+
"eval_steps_per_second": 7.302,
|
495 |
+
"step": 515
|
496 |
+
},
|
497 |
+
{
|
498 |
+
"epoch": 20.19,
|
499 |
+
"learning_rate": 2.778947368421053e-05,
|
500 |
+
"loss": 0.1031,
|
501 |
+
"step": 520
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 20.58,
|
505 |
+
"learning_rate": 2.7210526315789472e-05,
|
506 |
+
"loss": 0.0668,
|
507 |
+
"step": 530
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 20.97,
|
511 |
+
"learning_rate": 2.6631578947368422e-05,
|
512 |
+
"loss": 0.14,
|
513 |
+
"step": 540
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 20.97,
|
517 |
+
"eval_accuracy": 0.8695652173913043,
|
518 |
+
"eval_loss": 0.6455863118171692,
|
519 |
+
"eval_runtime": 0.8212,
|
520 |
+
"eval_samples_per_second": 56.015,
|
521 |
+
"eval_steps_per_second": 7.306,
|
522 |
+
"step": 540
|
523 |
+
},
|
524 |
+
{
|
525 |
+
"epoch": 21.36,
|
526 |
+
"learning_rate": 2.605263157894737e-05,
|
527 |
+
"loss": 0.0812,
|
528 |
+
"step": 550
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 21.75,
|
532 |
+
"learning_rate": 2.5473684210526318e-05,
|
533 |
+
"loss": 0.0506,
|
534 |
+
"step": 560
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 21.98,
|
538 |
+
"eval_accuracy": 0.782608695652174,
|
539 |
+
"eval_loss": 0.9126870632171631,
|
540 |
+
"eval_runtime": 0.8397,
|
541 |
+
"eval_samples_per_second": 54.781,
|
542 |
+
"eval_steps_per_second": 7.145,
|
543 |
+
"step": 566
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 22.14,
|
547 |
+
"learning_rate": 2.4894736842105264e-05,
|
548 |
+
"loss": 0.0413,
|
549 |
+
"step": 570
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 22.52,
|
553 |
+
"learning_rate": 2.431578947368421e-05,
|
554 |
+
"loss": 0.1031,
|
555 |
+
"step": 580
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 22.91,
|
559 |
+
"learning_rate": 2.373684210526316e-05,
|
560 |
+
"loss": 0.0963,
|
561 |
+
"step": 590
|
562 |
+
},
|
563 |
+
{
|
564 |
+
"epoch": 22.99,
|
565 |
+
"eval_accuracy": 0.8260869565217391,
|
566 |
+
"eval_loss": 0.6365217566490173,
|
567 |
+
"eval_runtime": 0.8112,
|
568 |
+
"eval_samples_per_second": 56.706,
|
569 |
+
"eval_steps_per_second": 7.396,
|
570 |
+
"step": 592
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 23.3,
|
574 |
+
"learning_rate": 2.3157894736842103e-05,
|
575 |
+
"loss": 0.0693,
|
576 |
+
"step": 600
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 23.69,
|
580 |
+
"learning_rate": 2.2578947368421053e-05,
|
581 |
+
"loss": 0.0612,
|
582 |
+
"step": 610
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 24.0,
|
586 |
+
"eval_accuracy": 0.8043478260869565,
|
587 |
+
"eval_loss": 0.8252330422401428,
|
588 |
+
"eval_runtime": 0.8217,
|
589 |
+
"eval_samples_per_second": 55.982,
|
590 |
+
"eval_steps_per_second": 7.302,
|
591 |
+
"step": 618
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 24.08,
|
595 |
+
"learning_rate": 2.2000000000000003e-05,
|
596 |
+
"loss": 0.0353,
|
597 |
+
"step": 620
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 24.47,
|
601 |
+
"learning_rate": 2.142105263157895e-05,
|
602 |
+
"loss": 0.1415,
|
603 |
+
"step": 630
|
604 |
+
},
|
605 |
+
{
|
606 |
+
"epoch": 24.85,
|
607 |
+
"learning_rate": 2.0842105263157895e-05,
|
608 |
+
"loss": 0.0875,
|
609 |
+
"step": 640
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 24.97,
|
613 |
+
"eval_accuracy": 0.7391304347826086,
|
614 |
+
"eval_loss": 0.8843982815742493,
|
615 |
+
"eval_runtime": 0.7792,
|
616 |
+
"eval_samples_per_second": 59.035,
|
617 |
+
"eval_steps_per_second": 7.7,
|
618 |
+
"step": 643
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 25.24,
|
622 |
+
"learning_rate": 2.0263157894736842e-05,
|
623 |
+
"loss": 0.1033,
|
624 |
+
"step": 650
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 25.63,
|
628 |
+
"learning_rate": 1.968421052631579e-05,
|
629 |
+
"loss": 0.1041,
|
630 |
+
"step": 660
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 25.98,
|
634 |
+
"eval_accuracy": 0.8260869565217391,
|
635 |
+
"eval_loss": 0.6594071984291077,
|
636 |
+
"eval_runtime": 0.7782,
|
637 |
+
"eval_samples_per_second": 59.112,
|
638 |
+
"eval_steps_per_second": 7.71,
|
639 |
+
"step": 669
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 26.02,
|
643 |
+
"learning_rate": 1.9105263157894738e-05,
|
644 |
+
"loss": 0.0429,
|
645 |
+
"step": 670
|
646 |
+
},
|
647 |
+
{
|
648 |
+
"epoch": 26.41,
|
649 |
+
"learning_rate": 1.8526315789473684e-05,
|
650 |
+
"loss": 0.1199,
|
651 |
+
"step": 680
|
652 |
+
},
|
653 |
+
{
|
654 |
+
"epoch": 26.8,
|
655 |
+
"learning_rate": 1.7947368421052634e-05,
|
656 |
+
"loss": 0.0512,
|
657 |
+
"step": 690
|
658 |
+
},
|
659 |
+
{
|
660 |
+
"epoch": 26.99,
|
661 |
+
"eval_accuracy": 0.782608695652174,
|
662 |
+
"eval_loss": 0.9883273839950562,
|
663 |
+
"eval_runtime": 0.7907,
|
664 |
+
"eval_samples_per_second": 58.177,
|
665 |
+
"eval_steps_per_second": 7.588,
|
666 |
+
"step": 695
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 27.18,
|
670 |
+
"learning_rate": 1.736842105263158e-05,
|
671 |
+
"loss": 0.0506,
|
672 |
+
"step": 700
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 27.57,
|
676 |
+
"learning_rate": 1.6789473684210526e-05,
|
677 |
+
"loss": 0.0676,
|
678 |
+
"step": 710
|
679 |
+
},
|
680 |
+
{
|
681 |
+
"epoch": 27.96,
|
682 |
+
"learning_rate": 1.6210526315789473e-05,
|
683 |
+
"loss": 0.0675,
|
684 |
+
"step": 720
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 28.0,
|
688 |
+
"eval_accuracy": 0.8043478260869565,
|
689 |
+
"eval_loss": 0.9216018915176392,
|
690 |
+
"eval_runtime": 0.8032,
|
691 |
+
"eval_samples_per_second": 57.272,
|
692 |
+
"eval_steps_per_second": 7.47,
|
693 |
+
"step": 721
|
694 |
+
},
|
695 |
+
{
|
696 |
+
"epoch": 28.35,
|
697 |
+
"learning_rate": 1.5631578947368422e-05,
|
698 |
+
"loss": 0.0958,
|
699 |
+
"step": 730
|
700 |
+
},
|
701 |
+
{
|
702 |
+
"epoch": 28.74,
|
703 |
+
"learning_rate": 1.505263157894737e-05,
|
704 |
+
"loss": 0.0492,
|
705 |
+
"step": 740
|
706 |
+
},
|
707 |
+
{
|
708 |
+
"epoch": 28.97,
|
709 |
+
"eval_accuracy": 0.8043478260869565,
|
710 |
+
"eval_loss": 0.9284108877182007,
|
711 |
+
"eval_runtime": 0.7852,
|
712 |
+
"eval_samples_per_second": 58.586,
|
713 |
+
"eval_steps_per_second": 7.642,
|
714 |
+
"step": 746
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 29.13,
|
718 |
+
"learning_rate": 1.4473684210526315e-05,
|
719 |
+
"loss": 0.0574,
|
720 |
+
"step": 750
|
721 |
+
},
|
722 |
+
{
|
723 |
+
"epoch": 29.51,
|
724 |
+
"learning_rate": 1.3894736842105265e-05,
|
725 |
+
"loss": 0.0527,
|
726 |
+
"step": 760
|
727 |
+
},
|
728 |
+
{
|
729 |
+
"epoch": 29.9,
|
730 |
+
"learning_rate": 1.3315789473684211e-05,
|
731 |
+
"loss": 0.0679,
|
732 |
+
"step": 770
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 29.98,
|
736 |
+
"eval_accuracy": 0.782608695652174,
|
737 |
+
"eval_loss": 0.9341010451316833,
|
738 |
+
"eval_runtime": 0.8282,
|
739 |
+
"eval_samples_per_second": 55.542,
|
740 |
+
"eval_steps_per_second": 7.245,
|
741 |
+
"step": 772
|
742 |
+
},
|
743 |
+
{
|
744 |
+
"epoch": 30.29,
|
745 |
+
"learning_rate": 1.2736842105263159e-05,
|
746 |
+
"loss": 0.0248,
|
747 |
+
"step": 780
|
748 |
+
},
|
749 |
+
{
|
750 |
+
"epoch": 30.68,
|
751 |
+
"learning_rate": 1.2157894736842105e-05,
|
752 |
+
"loss": 0.0996,
|
753 |
+
"step": 790
|
754 |
+
},
|
755 |
+
{
|
756 |
+
"epoch": 30.99,
|
757 |
+
"eval_accuracy": 0.8043478260869565,
|
758 |
+
"eval_loss": 0.9608001112937927,
|
759 |
+
"eval_runtime": 0.8052,
|
760 |
+
"eval_samples_per_second": 57.13,
|
761 |
+
"eval_steps_per_second": 7.452,
|
762 |
+
"step": 798
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 31.07,
|
766 |
+
"learning_rate": 1.1578947368421052e-05,
|
767 |
+
"loss": 0.0596,
|
768 |
+
"step": 800
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 31.46,
|
772 |
+
"learning_rate": 1.1000000000000001e-05,
|
773 |
+
"loss": 0.047,
|
774 |
+
"step": 810
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 31.84,
|
778 |
+
"learning_rate": 1.0421052631578948e-05,
|
779 |
+
"loss": 0.0729,
|
780 |
+
"step": 820
|
781 |
+
},
|
782 |
+
{
|
783 |
+
"epoch": 32.0,
|
784 |
+
"eval_accuracy": 0.782608695652174,
|
785 |
+
"eval_loss": 1.015515685081482,
|
786 |
+
"eval_runtime": 0.7852,
|
787 |
+
"eval_samples_per_second": 58.586,
|
788 |
+
"eval_steps_per_second": 7.642,
|
789 |
+
"step": 824
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 32.23,
|
793 |
+
"learning_rate": 9.842105263157896e-06,
|
794 |
+
"loss": 0.0446,
|
795 |
+
"step": 830
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 32.62,
|
799 |
+
"learning_rate": 9.263157894736842e-06,
|
800 |
+
"loss": 0.0296,
|
801 |
+
"step": 840
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 32.97,
|
805 |
+
"eval_accuracy": 0.782608695652174,
|
806 |
+
"eval_loss": 1.0314323902130127,
|
807 |
+
"eval_runtime": 0.7802,
|
808 |
+
"eval_samples_per_second": 58.96,
|
809 |
+
"eval_steps_per_second": 7.69,
|
810 |
+
"step": 849
|
811 |
+
},
|
812 |
+
{
|
813 |
+
"epoch": 33.01,
|
814 |
+
"learning_rate": 8.68421052631579e-06,
|
815 |
+
"loss": 0.0918,
|
816 |
+
"step": 850
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 33.4,
|
820 |
+
"learning_rate": 8.105263157894736e-06,
|
821 |
+
"loss": 0.0642,
|
822 |
+
"step": 860
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 33.79,
|
826 |
+
"learning_rate": 7.526315789473685e-06,
|
827 |
+
"loss": 0.0414,
|
828 |
+
"step": 870
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 33.98,
|
832 |
+
"eval_accuracy": 0.8043478260869565,
|
833 |
+
"eval_loss": 0.8358449339866638,
|
834 |
+
"eval_runtime": 0.7882,
|
835 |
+
"eval_samples_per_second": 58.362,
|
836 |
+
"eval_steps_per_second": 7.612,
|
837 |
+
"step": 875
|
838 |
+
},
|
839 |
+
{
|
840 |
+
"epoch": 34.17,
|
841 |
+
"learning_rate": 6.947368421052632e-06,
|
842 |
+
"loss": 0.0482,
|
843 |
+
"step": 880
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 34.56,
|
847 |
+
"learning_rate": 6.3684210526315795e-06,
|
848 |
+
"loss": 0.0462,
|
849 |
+
"step": 890
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 34.95,
|
853 |
+
"learning_rate": 5.789473684210526e-06,
|
854 |
+
"loss": 0.04,
|
855 |
+
"step": 900
|
856 |
+
},
|
857 |
+
{
|
858 |
+
"epoch": 34.99,
|
859 |
+
"eval_accuracy": 0.8043478260869565,
|
860 |
+
"eval_loss": 0.891183614730835,
|
861 |
+
"eval_runtime": 0.8252,
|
862 |
+
"eval_samples_per_second": 55.745,
|
863 |
+
"eval_steps_per_second": 7.271,
|
864 |
+
"step": 901
|
865 |
+
},
|
866 |
+
{
|
867 |
+
"epoch": 35.34,
|
868 |
+
"learning_rate": 5.210526315789474e-06,
|
869 |
+
"loss": 0.0262,
|
870 |
+
"step": 910
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 35.73,
|
874 |
+
"learning_rate": 4.631578947368421e-06,
|
875 |
+
"loss": 0.0179,
|
876 |
+
"step": 920
|
877 |
+
},
|
878 |
+
{
|
879 |
+
"epoch": 36.0,
|
880 |
+
"eval_accuracy": 0.8043478260869565,
|
881 |
+
"eval_loss": 0.8544241786003113,
|
882 |
+
"eval_runtime": 0.7722,
|
883 |
+
"eval_samples_per_second": 59.571,
|
884 |
+
"eval_steps_per_second": 7.77,
|
885 |
+
"step": 927
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 36.12,
|
889 |
+
"learning_rate": 4.052631578947368e-06,
|
890 |
+
"loss": 0.0205,
|
891 |
+
"step": 930
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 36.5,
|
895 |
+
"learning_rate": 3.473684210526316e-06,
|
896 |
+
"loss": 0.0586,
|
897 |
+
"step": 940
|
898 |
+
},
|
899 |
+
{
|
900 |
+
"epoch": 36.89,
|
901 |
+
"learning_rate": 2.894736842105263e-06,
|
902 |
+
"loss": 0.0665,
|
903 |
+
"step": 950
|
904 |
+
},
|
905 |
+
{
|
906 |
+
"epoch": 36.97,
|
907 |
+
"eval_accuracy": 0.8043478260869565,
|
908 |
+
"eval_loss": 0.9153819680213928,
|
909 |
+
"eval_runtime": 0.7792,
|
910 |
+
"eval_samples_per_second": 59.037,
|
911 |
+
"eval_steps_per_second": 7.7,
|
912 |
+
"step": 952
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 37.28,
|
916 |
+
"learning_rate": 2.3157894736842105e-06,
|
917 |
+
"loss": 0.0597,
|
918 |
+
"step": 960
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 37.67,
|
922 |
+
"learning_rate": 1.736842105263158e-06,
|
923 |
+
"loss": 0.0413,
|
924 |
+
"step": 970
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 37.98,
|
928 |
+
"eval_accuracy": 0.8043478260869565,
|
929 |
+
"eval_loss": 0.8834414482116699,
|
930 |
+
"eval_runtime": 0.8242,
|
931 |
+
"eval_samples_per_second": 55.81,
|
932 |
+
"eval_steps_per_second": 7.28,
|
933 |
+
"step": 978
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 38.06,
|
937 |
+
"learning_rate": 1.1578947368421053e-06,
|
938 |
+
"loss": 0.0218,
|
939 |
+
"step": 980
|
940 |
+
},
|
941 |
+
{
|
942 |
+
"epoch": 38.45,
|
943 |
+
"learning_rate": 5.789473684210526e-07,
|
944 |
+
"loss": 0.0512,
|
945 |
+
"step": 990
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 38.83,
|
949 |
+
"learning_rate": 0.0,
|
950 |
+
"loss": 0.04,
|
951 |
+
"step": 1000
|
952 |
+
},
|
953 |
+
{
|
954 |
+
"epoch": 38.83,
|
955 |
+
"eval_accuracy": 0.8043478260869565,
|
956 |
+
"eval_loss": 0.8807706236839294,
|
957 |
+
"eval_runtime": 0.8612,
|
958 |
+
"eval_samples_per_second": 53.414,
|
959 |
+
"eval_steps_per_second": 6.967,
|
960 |
+
"step": 1000
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 38.83,
|
964 |
+
"step": 1000,
|
965 |
+
"total_flos": 2.4650643760327066e+18,
|
966 |
+
"train_loss": 0.2159135382026434,
|
967 |
+
"train_runtime": 637.3873,
|
968 |
+
"train_samples_per_second": 51.397,
|
969 |
+
"train_steps_per_second": 1.569
|
970 |
+
}
|
971 |
+
],
|
972 |
+
"logging_steps": 10,
|
973 |
+
"max_steps": 1000,
|
974 |
+
"num_input_tokens_seen": 0,
|
975 |
+
"num_train_epochs": 40,
|
976 |
+
"save_steps": 500,
|
977 |
+
"total_flos": 2.4650643760327066e+18,
|
978 |
+
"train_batch_size": 8,
|
979 |
+
"trial_name": null,
|
980 |
+
"trial_params": null
|
981 |
+
}
|