End of training
Browse files- README.md +1 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- runs/Apr17_11-56-18_1ed504b80988/events.out.tfevents.1713358490.1ed504b80988.34.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +1008 -0
README.md
CHANGED
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
17 |
|
18 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
- Loss: 0.
|
21 |
|
22 |
## Model description
|
23 |
|
|
|
17 |
|
18 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 0.0088
|
21 |
|
22 |
## Model description
|
23 |
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_loss": 0.008771373890340328,
|
4 |
+
"eval_runtime": 43.1028,
|
5 |
+
"eval_samples_per_second": 88.741,
|
6 |
+
"eval_steps_per_second": 2.784,
|
7 |
+
"total_flos": 4.2775914245208883e+18,
|
8 |
+
"train_loss": 0.028647772959333094,
|
9 |
+
"train_runtime": 3141.2683,
|
10 |
+
"train_samples_per_second": 54.785,
|
11 |
+
"train_steps_per_second": 0.428
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_loss": 0.008771373890340328,
|
4 |
+
"eval_runtime": 43.1028,
|
5 |
+
"eval_samples_per_second": 88.741,
|
6 |
+
"eval_steps_per_second": 2.784
|
7 |
+
}
|
runs/Apr17_11-56-18_1ed504b80988/events.out.tfevents.1713358490.1ed504b80988.34.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3341a3ebc8ff9ceda394c17fe3616683f71e620ec42aed827a2dbd8eabb52c98
|
3 |
+
size 359
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"total_flos": 4.2775914245208883e+18,
|
4 |
+
"train_loss": 0.028647772959333094,
|
5 |
+
"train_runtime": 3141.2683,
|
6 |
+
"train_samples_per_second": 54.785,
|
7 |
+
"train_steps_per_second": 0.428
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1008 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.008771373890340328,
|
3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-Kontur-competition/checkpoint-807",
|
4 |
+
"epoch": 5.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1345,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.04,
|
13 |
+
"grad_norm": 5.5162248611450195,
|
14 |
+
"learning_rate": 3.7037037037037037e-06,
|
15 |
+
"loss": 0.6819,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.07,
|
20 |
+
"grad_norm": 3.1403090953826904,
|
21 |
+
"learning_rate": 7.4074074074074075e-06,
|
22 |
+
"loss": 0.5394,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.11,
|
27 |
+
"grad_norm": 2.9430086612701416,
|
28 |
+
"learning_rate": 1.1111111111111112e-05,
|
29 |
+
"loss": 0.3795,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.15,
|
34 |
+
"grad_norm": 3.1442313194274902,
|
35 |
+
"learning_rate": 1.4814814814814815e-05,
|
36 |
+
"loss": 0.2726,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.19,
|
41 |
+
"grad_norm": 2.8489677906036377,
|
42 |
+
"learning_rate": 1.8518518518518518e-05,
|
43 |
+
"loss": 0.1909,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.22,
|
48 |
+
"grad_norm": 2.6733438968658447,
|
49 |
+
"learning_rate": 2.2222222222222223e-05,
|
50 |
+
"loss": 0.0903,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.26,
|
55 |
+
"grad_norm": 6.4875969886779785,
|
56 |
+
"learning_rate": 2.5925925925925925e-05,
|
57 |
+
"loss": 0.064,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.3,
|
62 |
+
"grad_norm": 10.464859008789062,
|
63 |
+
"learning_rate": 2.962962962962963e-05,
|
64 |
+
"loss": 0.0439,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.33,
|
69 |
+
"grad_norm": 3.629943609237671,
|
70 |
+
"learning_rate": 3.3333333333333335e-05,
|
71 |
+
"loss": 0.0384,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.37,
|
76 |
+
"grad_norm": 12.326812744140625,
|
77 |
+
"learning_rate": 3.7037037037037037e-05,
|
78 |
+
"loss": 0.0418,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.41,
|
83 |
+
"grad_norm": 4.8540754318237305,
|
84 |
+
"learning_rate": 4.074074074074074e-05,
|
85 |
+
"loss": 0.0378,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.45,
|
90 |
+
"grad_norm": 7.512868881225586,
|
91 |
+
"learning_rate": 4.4444444444444447e-05,
|
92 |
+
"loss": 0.0261,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.48,
|
97 |
+
"grad_norm": 4.206182479858398,
|
98 |
+
"learning_rate": 4.814814814814815e-05,
|
99 |
+
"loss": 0.0317,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.52,
|
104 |
+
"grad_norm": 4.706695556640625,
|
105 |
+
"learning_rate": 4.979338842975207e-05,
|
106 |
+
"loss": 0.0353,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.56,
|
111 |
+
"grad_norm": 1.3356839418411255,
|
112 |
+
"learning_rate": 4.9380165289256205e-05,
|
113 |
+
"loss": 0.0261,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.59,
|
118 |
+
"grad_norm": 0.30266404151916504,
|
119 |
+
"learning_rate": 4.896694214876033e-05,
|
120 |
+
"loss": 0.0174,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.63,
|
125 |
+
"grad_norm": 4.922573566436768,
|
126 |
+
"learning_rate": 4.855371900826447e-05,
|
127 |
+
"loss": 0.0533,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.67,
|
132 |
+
"grad_norm": 2.4459125995635986,
|
133 |
+
"learning_rate": 4.8140495867768596e-05,
|
134 |
+
"loss": 0.0275,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.71,
|
139 |
+
"grad_norm": 4.6485915184021,
|
140 |
+
"learning_rate": 4.772727272727273e-05,
|
141 |
+
"loss": 0.0382,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.74,
|
146 |
+
"grad_norm": 7.907899379730225,
|
147 |
+
"learning_rate": 4.731404958677686e-05,
|
148 |
+
"loss": 0.0312,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.78,
|
153 |
+
"grad_norm": 1.321392297744751,
|
154 |
+
"learning_rate": 4.6900826446280993e-05,
|
155 |
+
"loss": 0.0174,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.82,
|
160 |
+
"grad_norm": 0.5613566040992737,
|
161 |
+
"learning_rate": 4.648760330578513e-05,
|
162 |
+
"loss": 0.0128,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.86,
|
167 |
+
"grad_norm": 0.357917845249176,
|
168 |
+
"learning_rate": 4.607438016528926e-05,
|
169 |
+
"loss": 0.0173,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.89,
|
174 |
+
"grad_norm": 1.047433853149414,
|
175 |
+
"learning_rate": 4.566115702479339e-05,
|
176 |
+
"loss": 0.0231,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.93,
|
181 |
+
"grad_norm": 0.5947940349578857,
|
182 |
+
"learning_rate": 4.524793388429752e-05,
|
183 |
+
"loss": 0.0279,
|
184 |
+
"step": 250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.97,
|
188 |
+
"grad_norm": 1.0275371074676514,
|
189 |
+
"learning_rate": 4.4834710743801654e-05,
|
190 |
+
"loss": 0.0216,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 1.0,
|
195 |
+
"eval_loss": 0.06352325528860092,
|
196 |
+
"eval_runtime": 43.5548,
|
197 |
+
"eval_samples_per_second": 87.82,
|
198 |
+
"eval_steps_per_second": 2.755,
|
199 |
+
"step": 269
|
200 |
+
},
|
201 |
+
{
|
202 |
+
"epoch": 1.0,
|
203 |
+
"grad_norm": 8.136920928955078,
|
204 |
+
"learning_rate": 4.442148760330579e-05,
|
205 |
+
"loss": 0.0226,
|
206 |
+
"step": 270
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 1.04,
|
210 |
+
"grad_norm": 5.6582255363464355,
|
211 |
+
"learning_rate": 4.400826446280992e-05,
|
212 |
+
"loss": 0.0142,
|
213 |
+
"step": 280
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 1.08,
|
217 |
+
"grad_norm": 7.312340259552002,
|
218 |
+
"learning_rate": 4.359504132231405e-05,
|
219 |
+
"loss": 0.0298,
|
220 |
+
"step": 290
|
221 |
+
},
|
222 |
+
{
|
223 |
+
"epoch": 1.12,
|
224 |
+
"grad_norm": 0.31576940417289734,
|
225 |
+
"learning_rate": 4.318181818181819e-05,
|
226 |
+
"loss": 0.018,
|
227 |
+
"step": 300
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 1.15,
|
231 |
+
"grad_norm": 14.974953651428223,
|
232 |
+
"learning_rate": 4.2768595041322315e-05,
|
233 |
+
"loss": 0.0367,
|
234 |
+
"step": 310
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 1.19,
|
238 |
+
"grad_norm": 4.256837844848633,
|
239 |
+
"learning_rate": 4.235537190082644e-05,
|
240 |
+
"loss": 0.0064,
|
241 |
+
"step": 320
|
242 |
+
},
|
243 |
+
{
|
244 |
+
"epoch": 1.23,
|
245 |
+
"grad_norm": 0.972306489944458,
|
246 |
+
"learning_rate": 4.194214876033058e-05,
|
247 |
+
"loss": 0.014,
|
248 |
+
"step": 330
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 1.26,
|
252 |
+
"grad_norm": 0.03949430584907532,
|
253 |
+
"learning_rate": 4.152892561983471e-05,
|
254 |
+
"loss": 0.0121,
|
255 |
+
"step": 340
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 1.3,
|
259 |
+
"grad_norm": 0.8921181559562683,
|
260 |
+
"learning_rate": 4.111570247933885e-05,
|
261 |
+
"loss": 0.0135,
|
262 |
+
"step": 350
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 1.34,
|
266 |
+
"grad_norm": 4.770934581756592,
|
267 |
+
"learning_rate": 4.0702479338842975e-05,
|
268 |
+
"loss": 0.0279,
|
269 |
+
"step": 360
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 1.38,
|
273 |
+
"grad_norm": 12.435590744018555,
|
274 |
+
"learning_rate": 4.028925619834711e-05,
|
275 |
+
"loss": 0.0263,
|
276 |
+
"step": 370
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 1.41,
|
280 |
+
"grad_norm": 4.150902271270752,
|
281 |
+
"learning_rate": 3.9876033057851245e-05,
|
282 |
+
"loss": 0.0169,
|
283 |
+
"step": 380
|
284 |
+
},
|
285 |
+
{
|
286 |
+
"epoch": 1.45,
|
287 |
+
"grad_norm": 1.9341884851455688,
|
288 |
+
"learning_rate": 3.946280991735537e-05,
|
289 |
+
"loss": 0.0126,
|
290 |
+
"step": 390
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 1.49,
|
294 |
+
"grad_norm": 2.019587516784668,
|
295 |
+
"learning_rate": 3.90495867768595e-05,
|
296 |
+
"loss": 0.0075,
|
297 |
+
"step": 400
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 1.52,
|
301 |
+
"grad_norm": 0.34746670722961426,
|
302 |
+
"learning_rate": 3.8636363636363636e-05,
|
303 |
+
"loss": 0.0186,
|
304 |
+
"step": 410
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 1.56,
|
308 |
+
"grad_norm": 4.3460187911987305,
|
309 |
+
"learning_rate": 3.822314049586777e-05,
|
310 |
+
"loss": 0.015,
|
311 |
+
"step": 420
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 1.6,
|
315 |
+
"grad_norm": 0.06722448021173477,
|
316 |
+
"learning_rate": 3.7809917355371906e-05,
|
317 |
+
"loss": 0.0314,
|
318 |
+
"step": 430
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 1.64,
|
322 |
+
"grad_norm": 5.969712257385254,
|
323 |
+
"learning_rate": 3.7396694214876034e-05,
|
324 |
+
"loss": 0.0132,
|
325 |
+
"step": 440
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 1.67,
|
329 |
+
"grad_norm": 5.939075946807861,
|
330 |
+
"learning_rate": 3.698347107438017e-05,
|
331 |
+
"loss": 0.0191,
|
332 |
+
"step": 450
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 1.71,
|
336 |
+
"grad_norm": 2.4963324069976807,
|
337 |
+
"learning_rate": 3.65702479338843e-05,
|
338 |
+
"loss": 0.0107,
|
339 |
+
"step": 460
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 1.75,
|
343 |
+
"grad_norm": 4.111177921295166,
|
344 |
+
"learning_rate": 3.615702479338843e-05,
|
345 |
+
"loss": 0.0204,
|
346 |
+
"step": 470
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 1.78,
|
350 |
+
"grad_norm": 0.07025858759880066,
|
351 |
+
"learning_rate": 3.574380165289256e-05,
|
352 |
+
"loss": 0.0142,
|
353 |
+
"step": 480
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 1.82,
|
357 |
+
"grad_norm": 9.363666534423828,
|
358 |
+
"learning_rate": 3.5330578512396694e-05,
|
359 |
+
"loss": 0.0251,
|
360 |
+
"step": 490
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 1.86,
|
364 |
+
"grad_norm": 2.941878080368042,
|
365 |
+
"learning_rate": 3.491735537190083e-05,
|
366 |
+
"loss": 0.0123,
|
367 |
+
"step": 500
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 1.9,
|
371 |
+
"grad_norm": 5.7476348876953125,
|
372 |
+
"learning_rate": 3.4504132231404964e-05,
|
373 |
+
"loss": 0.0096,
|
374 |
+
"step": 510
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 1.93,
|
378 |
+
"grad_norm": 27.465803146362305,
|
379 |
+
"learning_rate": 3.409090909090909e-05,
|
380 |
+
"loss": 0.028,
|
381 |
+
"step": 520
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 1.97,
|
385 |
+
"grad_norm": 0.047597743570804596,
|
386 |
+
"learning_rate": 3.367768595041322e-05,
|
387 |
+
"loss": 0.0286,
|
388 |
+
"step": 530
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 2.0,
|
392 |
+
"eval_loss": 0.0467231348156929,
|
393 |
+
"eval_runtime": 43.1197,
|
394 |
+
"eval_samples_per_second": 88.707,
|
395 |
+
"eval_steps_per_second": 2.783,
|
396 |
+
"step": 538
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 2.01,
|
400 |
+
"grad_norm": 0.10430701822042465,
|
401 |
+
"learning_rate": 3.3264462809917355e-05,
|
402 |
+
"loss": 0.009,
|
403 |
+
"step": 540
|
404 |
+
},
|
405 |
+
{
|
406 |
+
"epoch": 2.04,
|
407 |
+
"grad_norm": 4.288072109222412,
|
408 |
+
"learning_rate": 3.285123966942149e-05,
|
409 |
+
"loss": 0.0041,
|
410 |
+
"step": 550
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 2.08,
|
414 |
+
"grad_norm": 1.3560419082641602,
|
415 |
+
"learning_rate": 3.243801652892562e-05,
|
416 |
+
"loss": 0.0175,
|
417 |
+
"step": 560
|
418 |
+
},
|
419 |
+
{
|
420 |
+
"epoch": 2.12,
|
421 |
+
"grad_norm": 5.84924840927124,
|
422 |
+
"learning_rate": 3.202479338842975e-05,
|
423 |
+
"loss": 0.0095,
|
424 |
+
"step": 570
|
425 |
+
},
|
426 |
+
{
|
427 |
+
"epoch": 2.16,
|
428 |
+
"grad_norm": 8.711688041687012,
|
429 |
+
"learning_rate": 3.161157024793389e-05,
|
430 |
+
"loss": 0.028,
|
431 |
+
"step": 580
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 2.19,
|
435 |
+
"grad_norm": 1.378418207168579,
|
436 |
+
"learning_rate": 3.119834710743802e-05,
|
437 |
+
"loss": 0.0244,
|
438 |
+
"step": 590
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 2.23,
|
442 |
+
"grad_norm": 4.585868835449219,
|
443 |
+
"learning_rate": 3.078512396694215e-05,
|
444 |
+
"loss": 0.0045,
|
445 |
+
"step": 600
|
446 |
+
},
|
447 |
+
{
|
448 |
+
"epoch": 2.27,
|
449 |
+
"grad_norm": 14.25820541381836,
|
450 |
+
"learning_rate": 3.0371900826446282e-05,
|
451 |
+
"loss": 0.0217,
|
452 |
+
"step": 610
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 2.3,
|
456 |
+
"grad_norm": 0.5124061107635498,
|
457 |
+
"learning_rate": 2.9958677685950414e-05,
|
458 |
+
"loss": 0.0015,
|
459 |
+
"step": 620
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 2.34,
|
463 |
+
"grad_norm": 5.597954750061035,
|
464 |
+
"learning_rate": 2.954545454545455e-05,
|
465 |
+
"loss": 0.0096,
|
466 |
+
"step": 630
|
467 |
+
},
|
468 |
+
{
|
469 |
+
"epoch": 2.38,
|
470 |
+
"grad_norm": 4.180160045623779,
|
471 |
+
"learning_rate": 2.9132231404958676e-05,
|
472 |
+
"loss": 0.016,
|
473 |
+
"step": 640
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 2.42,
|
477 |
+
"grad_norm": 1.3298313617706299,
|
478 |
+
"learning_rate": 2.871900826446281e-05,
|
479 |
+
"loss": 0.0104,
|
480 |
+
"step": 650
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 2.45,
|
484 |
+
"grad_norm": 0.11605699360370636,
|
485 |
+
"learning_rate": 2.8305785123966943e-05,
|
486 |
+
"loss": 0.0176,
|
487 |
+
"step": 660
|
488 |
+
},
|
489 |
+
{
|
490 |
+
"epoch": 2.49,
|
491 |
+
"grad_norm": 0.0669424757361412,
|
492 |
+
"learning_rate": 2.7892561983471078e-05,
|
493 |
+
"loss": 0.0043,
|
494 |
+
"step": 670
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 2.53,
|
498 |
+
"grad_norm": 0.7822141647338867,
|
499 |
+
"learning_rate": 2.7479338842975206e-05,
|
500 |
+
"loss": 0.0041,
|
501 |
+
"step": 680
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 2.57,
|
505 |
+
"grad_norm": 0.9650640487670898,
|
506 |
+
"learning_rate": 2.7066115702479337e-05,
|
507 |
+
"loss": 0.0096,
|
508 |
+
"step": 690
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 2.6,
|
512 |
+
"grad_norm": 10.423120498657227,
|
513 |
+
"learning_rate": 2.6652892561983472e-05,
|
514 |
+
"loss": 0.0065,
|
515 |
+
"step": 700
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 2.64,
|
519 |
+
"grad_norm": 10.540702819824219,
|
520 |
+
"learning_rate": 2.6239669421487607e-05,
|
521 |
+
"loss": 0.0068,
|
522 |
+
"step": 710
|
523 |
+
},
|
524 |
+
{
|
525 |
+
"epoch": 2.68,
|
526 |
+
"grad_norm": 0.06544263660907745,
|
527 |
+
"learning_rate": 2.5826446280991735e-05,
|
528 |
+
"loss": 0.0142,
|
529 |
+
"step": 720
|
530 |
+
},
|
531 |
+
{
|
532 |
+
"epoch": 2.71,
|
533 |
+
"grad_norm": 1.903037428855896,
|
534 |
+
"learning_rate": 2.5413223140495866e-05,
|
535 |
+
"loss": 0.0079,
|
536 |
+
"step": 730
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 2.75,
|
540 |
+
"grad_norm": 8.562331199645996,
|
541 |
+
"learning_rate": 2.5e-05,
|
542 |
+
"loss": 0.0109,
|
543 |
+
"step": 740
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 2.79,
|
547 |
+
"grad_norm": 2.383655548095703,
|
548 |
+
"learning_rate": 2.4586776859504136e-05,
|
549 |
+
"loss": 0.0063,
|
550 |
+
"step": 750
|
551 |
+
},
|
552 |
+
{
|
553 |
+
"epoch": 2.83,
|
554 |
+
"grad_norm": 6.162371635437012,
|
555 |
+
"learning_rate": 2.4173553719008264e-05,
|
556 |
+
"loss": 0.0118,
|
557 |
+
"step": 760
|
558 |
+
},
|
559 |
+
{
|
560 |
+
"epoch": 2.86,
|
561 |
+
"grad_norm": 0.7068905234336853,
|
562 |
+
"learning_rate": 2.37603305785124e-05,
|
563 |
+
"loss": 0.016,
|
564 |
+
"step": 770
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 2.9,
|
568 |
+
"grad_norm": 0.07612281292676926,
|
569 |
+
"learning_rate": 2.334710743801653e-05,
|
570 |
+
"loss": 0.008,
|
571 |
+
"step": 780
|
572 |
+
},
|
573 |
+
{
|
574 |
+
"epoch": 2.94,
|
575 |
+
"grad_norm": 5.8349199295043945,
|
576 |
+
"learning_rate": 2.2933884297520665e-05,
|
577 |
+
"loss": 0.0091,
|
578 |
+
"step": 790
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 2.97,
|
582 |
+
"grad_norm": 0.30163004994392395,
|
583 |
+
"learning_rate": 2.2520661157024793e-05,
|
584 |
+
"loss": 0.0252,
|
585 |
+
"step": 800
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 3.0,
|
589 |
+
"eval_loss": 0.008771373890340328,
|
590 |
+
"eval_runtime": 43.1769,
|
591 |
+
"eval_samples_per_second": 88.589,
|
592 |
+
"eval_steps_per_second": 2.779,
|
593 |
+
"step": 807
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 3.01,
|
597 |
+
"grad_norm": 1.3629869222640991,
|
598 |
+
"learning_rate": 2.2107438016528928e-05,
|
599 |
+
"loss": 0.0095,
|
600 |
+
"step": 810
|
601 |
+
},
|
602 |
+
{
|
603 |
+
"epoch": 3.05,
|
604 |
+
"grad_norm": 1.119672179222107,
|
605 |
+
"learning_rate": 2.169421487603306e-05,
|
606 |
+
"loss": 0.006,
|
607 |
+
"step": 820
|
608 |
+
},
|
609 |
+
{
|
610 |
+
"epoch": 3.09,
|
611 |
+
"grad_norm": 8.719965934753418,
|
612 |
+
"learning_rate": 2.128099173553719e-05,
|
613 |
+
"loss": 0.0144,
|
614 |
+
"step": 830
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 3.12,
|
618 |
+
"grad_norm": 0.12227091938257217,
|
619 |
+
"learning_rate": 2.0867768595041323e-05,
|
620 |
+
"loss": 0.0033,
|
621 |
+
"step": 840
|
622 |
+
},
|
623 |
+
{
|
624 |
+
"epoch": 3.16,
|
625 |
+
"grad_norm": 1.8717501163482666,
|
626 |
+
"learning_rate": 2.0454545454545457e-05,
|
627 |
+
"loss": 0.0027,
|
628 |
+
"step": 850
|
629 |
+
},
|
630 |
+
{
|
631 |
+
"epoch": 3.2,
|
632 |
+
"grad_norm": 6.553406238555908,
|
633 |
+
"learning_rate": 2.004132231404959e-05,
|
634 |
+
"loss": 0.0133,
|
635 |
+
"step": 860
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 3.23,
|
639 |
+
"grad_norm": 2.087083101272583,
|
640 |
+
"learning_rate": 1.962809917355372e-05,
|
641 |
+
"loss": 0.0134,
|
642 |
+
"step": 870
|
643 |
+
},
|
644 |
+
{
|
645 |
+
"epoch": 3.27,
|
646 |
+
"grad_norm": 1.4037662744522095,
|
647 |
+
"learning_rate": 1.9214876033057852e-05,
|
648 |
+
"loss": 0.0031,
|
649 |
+
"step": 880
|
650 |
+
},
|
651 |
+
{
|
652 |
+
"epoch": 3.31,
|
653 |
+
"grad_norm": 2.297409772872925,
|
654 |
+
"learning_rate": 1.8801652892561987e-05,
|
655 |
+
"loss": 0.0162,
|
656 |
+
"step": 890
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 3.35,
|
660 |
+
"grad_norm": 0.025398077443242073,
|
661 |
+
"learning_rate": 1.8388429752066118e-05,
|
662 |
+
"loss": 0.0061,
|
663 |
+
"step": 900
|
664 |
+
},
|
665 |
+
{
|
666 |
+
"epoch": 3.38,
|
667 |
+
"grad_norm": 0.037510234862565994,
|
668 |
+
"learning_rate": 1.797520661157025e-05,
|
669 |
+
"loss": 0.0064,
|
670 |
+
"step": 910
|
671 |
+
},
|
672 |
+
{
|
673 |
+
"epoch": 3.42,
|
674 |
+
"grad_norm": 0.3090139627456665,
|
675 |
+
"learning_rate": 1.756198347107438e-05,
|
676 |
+
"loss": 0.004,
|
677 |
+
"step": 920
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 3.46,
|
681 |
+
"grad_norm": 3.6702754497528076,
|
682 |
+
"learning_rate": 1.7148760330578516e-05,
|
683 |
+
"loss": 0.0046,
|
684 |
+
"step": 930
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 3.49,
|
688 |
+
"grad_norm": 0.22104105353355408,
|
689 |
+
"learning_rate": 1.6735537190082644e-05,
|
690 |
+
"loss": 0.0003,
|
691 |
+
"step": 940
|
692 |
+
},
|
693 |
+
{
|
694 |
+
"epoch": 3.53,
|
695 |
+
"grad_norm": 2.2657175064086914,
|
696 |
+
"learning_rate": 1.632231404958678e-05,
|
697 |
+
"loss": 0.0048,
|
698 |
+
"step": 950
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 3.57,
|
702 |
+
"grad_norm": 0.0027564549818634987,
|
703 |
+
"learning_rate": 1.590909090909091e-05,
|
704 |
+
"loss": 0.0112,
|
705 |
+
"step": 960
|
706 |
+
},
|
707 |
+
{
|
708 |
+
"epoch": 3.61,
|
709 |
+
"grad_norm": 0.024479985237121582,
|
710 |
+
"learning_rate": 1.549586776859504e-05,
|
711 |
+
"loss": 0.0031,
|
712 |
+
"step": 970
|
713 |
+
},
|
714 |
+
{
|
715 |
+
"epoch": 3.64,
|
716 |
+
"grad_norm": 0.3030235767364502,
|
717 |
+
"learning_rate": 1.5082644628099175e-05,
|
718 |
+
"loss": 0.0031,
|
719 |
+
"step": 980
|
720 |
+
},
|
721 |
+
{
|
722 |
+
"epoch": 3.68,
|
723 |
+
"grad_norm": 0.01274044532328844,
|
724 |
+
"learning_rate": 1.4669421487603308e-05,
|
725 |
+
"loss": 0.0086,
|
726 |
+
"step": 990
|
727 |
+
},
|
728 |
+
{
|
729 |
+
"epoch": 3.72,
|
730 |
+
"grad_norm": 0.9196871519088745,
|
731 |
+
"learning_rate": 1.4256198347107438e-05,
|
732 |
+
"loss": 0.01,
|
733 |
+
"step": 1000
|
734 |
+
},
|
735 |
+
{
|
736 |
+
"epoch": 3.75,
|
737 |
+
"grad_norm": 0.09063485264778137,
|
738 |
+
"learning_rate": 1.3842975206611573e-05,
|
739 |
+
"loss": 0.0043,
|
740 |
+
"step": 1010
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 3.79,
|
744 |
+
"grad_norm": 0.008292087353765965,
|
745 |
+
"learning_rate": 1.3429752066115702e-05,
|
746 |
+
"loss": 0.0028,
|
747 |
+
"step": 1020
|
748 |
+
},
|
749 |
+
{
|
750 |
+
"epoch": 3.83,
|
751 |
+
"grad_norm": 0.17935311794281006,
|
752 |
+
"learning_rate": 1.3016528925619837e-05,
|
753 |
+
"loss": 0.006,
|
754 |
+
"step": 1030
|
755 |
+
},
|
756 |
+
{
|
757 |
+
"epoch": 3.87,
|
758 |
+
"grad_norm": 0.2591820955276489,
|
759 |
+
"learning_rate": 1.2603305785123967e-05,
|
760 |
+
"loss": 0.0072,
|
761 |
+
"step": 1040
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 3.9,
|
765 |
+
"grad_norm": 0.8280309438705444,
|
766 |
+
"learning_rate": 1.21900826446281e-05,
|
767 |
+
"loss": 0.0044,
|
768 |
+
"step": 1050
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 3.94,
|
772 |
+
"grad_norm": 1.8738752603530884,
|
773 |
+
"learning_rate": 1.1776859504132231e-05,
|
774 |
+
"loss": 0.0033,
|
775 |
+
"step": 1060
|
776 |
+
},
|
777 |
+
{
|
778 |
+
"epoch": 3.98,
|
779 |
+
"grad_norm": 0.01168906595557928,
|
780 |
+
"learning_rate": 1.1363636363636365e-05,
|
781 |
+
"loss": 0.0003,
|
782 |
+
"step": 1070
|
783 |
+
},
|
784 |
+
{
|
785 |
+
"epoch": 4.0,
|
786 |
+
"eval_loss": 0.033934369683265686,
|
787 |
+
"eval_runtime": 42.9173,
|
788 |
+
"eval_samples_per_second": 89.125,
|
789 |
+
"eval_steps_per_second": 2.796,
|
790 |
+
"step": 1076
|
791 |
+
},
|
792 |
+
{
|
793 |
+
"epoch": 4.01,
|
794 |
+
"grad_norm": 1.7786719799041748,
|
795 |
+
"learning_rate": 1.0950413223140496e-05,
|
796 |
+
"loss": 0.005,
|
797 |
+
"step": 1080
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 4.05,
|
801 |
+
"grad_norm": 0.44828712940216064,
|
802 |
+
"learning_rate": 1.0537190082644628e-05,
|
803 |
+
"loss": 0.001,
|
804 |
+
"step": 1090
|
805 |
+
},
|
806 |
+
{
|
807 |
+
"epoch": 4.09,
|
808 |
+
"grad_norm": 0.0037061686161905527,
|
809 |
+
"learning_rate": 1.012396694214876e-05,
|
810 |
+
"loss": 0.0005,
|
811 |
+
"step": 1100
|
812 |
+
},
|
813 |
+
{
|
814 |
+
"epoch": 4.13,
|
815 |
+
"grad_norm": 7.8463358879089355,
|
816 |
+
"learning_rate": 9.710743801652892e-06,
|
817 |
+
"loss": 0.0054,
|
818 |
+
"step": 1110
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 4.16,
|
822 |
+
"grad_norm": 0.07012557983398438,
|
823 |
+
"learning_rate": 9.297520661157025e-06,
|
824 |
+
"loss": 0.0037,
|
825 |
+
"step": 1120
|
826 |
+
},
|
827 |
+
{
|
828 |
+
"epoch": 4.2,
|
829 |
+
"grad_norm": 2.830775260925293,
|
830 |
+
"learning_rate": 8.884297520661157e-06,
|
831 |
+
"loss": 0.0015,
|
832 |
+
"step": 1130
|
833 |
+
},
|
834 |
+
{
|
835 |
+
"epoch": 4.24,
|
836 |
+
"grad_norm": 0.2861256003379822,
|
837 |
+
"learning_rate": 8.47107438016529e-06,
|
838 |
+
"loss": 0.0042,
|
839 |
+
"step": 1140
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 4.28,
|
843 |
+
"grad_norm": 0.9108858108520508,
|
844 |
+
"learning_rate": 8.057851239669421e-06,
|
845 |
+
"loss": 0.0003,
|
846 |
+
"step": 1150
|
847 |
+
},
|
848 |
+
{
|
849 |
+
"epoch": 4.31,
|
850 |
+
"grad_norm": 0.06316760182380676,
|
851 |
+
"learning_rate": 7.644628099173553e-06,
|
852 |
+
"loss": 0.0002,
|
853 |
+
"step": 1160
|
854 |
+
},
|
855 |
+
{
|
856 |
+
"epoch": 4.35,
|
857 |
+
"grad_norm": 0.006003058515489101,
|
858 |
+
"learning_rate": 7.231404958677686e-06,
|
859 |
+
"loss": 0.001,
|
860 |
+
"step": 1170
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 4.39,
|
864 |
+
"grad_norm": 0.0066665462218225,
|
865 |
+
"learning_rate": 6.818181818181818e-06,
|
866 |
+
"loss": 0.012,
|
867 |
+
"step": 1180
|
868 |
+
},
|
869 |
+
{
|
870 |
+
"epoch": 4.42,
|
871 |
+
"grad_norm": 7.692056179046631,
|
872 |
+
"learning_rate": 6.404958677685951e-06,
|
873 |
+
"loss": 0.0029,
|
874 |
+
"step": 1190
|
875 |
+
},
|
876 |
+
{
|
877 |
+
"epoch": 4.46,
|
878 |
+
"grad_norm": 0.01048464234918356,
|
879 |
+
"learning_rate": 5.991735537190083e-06,
|
880 |
+
"loss": 0.0007,
|
881 |
+
"step": 1200
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 4.5,
|
885 |
+
"grad_norm": 0.016681592911481857,
|
886 |
+
"learning_rate": 5.578512396694215e-06,
|
887 |
+
"loss": 0.0015,
|
888 |
+
"step": 1210
|
889 |
+
},
|
890 |
+
{
|
891 |
+
"epoch": 4.54,
|
892 |
+
"grad_norm": 0.035201072692871094,
|
893 |
+
"learning_rate": 5.1652892561983475e-06,
|
894 |
+
"loss": 0.0028,
|
895 |
+
"step": 1220
|
896 |
+
},
|
897 |
+
{
|
898 |
+
"epoch": 4.57,
|
899 |
+
"grad_norm": 0.05388149991631508,
|
900 |
+
"learning_rate": 4.75206611570248e-06,
|
901 |
+
"loss": 0.0008,
|
902 |
+
"step": 1230
|
903 |
+
},
|
904 |
+
{
|
905 |
+
"epoch": 4.61,
|
906 |
+
"grad_norm": 0.14326803386211395,
|
907 |
+
"learning_rate": 4.338842975206612e-06,
|
908 |
+
"loss": 0.0001,
|
909 |
+
"step": 1240
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 4.65,
|
913 |
+
"grad_norm": 0.001435384969227016,
|
914 |
+
"learning_rate": 3.925619834710744e-06,
|
915 |
+
"loss": 0.0048,
|
916 |
+
"step": 1250
|
917 |
+
},
|
918 |
+
{
|
919 |
+
"epoch": 4.68,
|
920 |
+
"grad_norm": 0.11075280606746674,
|
921 |
+
"learning_rate": 3.5123966942148763e-06,
|
922 |
+
"loss": 0.0006,
|
923 |
+
"step": 1260
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 4.72,
|
927 |
+
"grad_norm": 0.05017843842506409,
|
928 |
+
"learning_rate": 3.0991735537190086e-06,
|
929 |
+
"loss": 0.0001,
|
930 |
+
"step": 1270
|
931 |
+
},
|
932 |
+
{
|
933 |
+
"epoch": 4.76,
|
934 |
+
"grad_norm": 0.002463772427290678,
|
935 |
+
"learning_rate": 2.685950413223141e-06,
|
936 |
+
"loss": 0.0003,
|
937 |
+
"step": 1280
|
938 |
+
},
|
939 |
+
{
|
940 |
+
"epoch": 4.8,
|
941 |
+
"grad_norm": 13.101515769958496,
|
942 |
+
"learning_rate": 2.2727272727272728e-06,
|
943 |
+
"loss": 0.0011,
|
944 |
+
"step": 1290
|
945 |
+
},
|
946 |
+
{
|
947 |
+
"epoch": 4.83,
|
948 |
+
"grad_norm": 0.003497660392895341,
|
949 |
+
"learning_rate": 1.859504132231405e-06,
|
950 |
+
"loss": 0.0014,
|
951 |
+
"step": 1300
|
952 |
+
},
|
953 |
+
{
|
954 |
+
"epoch": 4.87,
|
955 |
+
"grad_norm": 0.07368556410074234,
|
956 |
+
"learning_rate": 1.4462809917355374e-06,
|
957 |
+
"loss": 0.0013,
|
958 |
+
"step": 1310
|
959 |
+
},
|
960 |
+
{
|
961 |
+
"epoch": 4.91,
|
962 |
+
"grad_norm": 0.0647512823343277,
|
963 |
+
"learning_rate": 1.0330578512396695e-06,
|
964 |
+
"loss": 0.0045,
|
965 |
+
"step": 1320
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 4.94,
|
969 |
+
"grad_norm": 0.0005362280062399805,
|
970 |
+
"learning_rate": 6.198347107438017e-07,
|
971 |
+
"loss": 0.0141,
|
972 |
+
"step": 1330
|
973 |
+
},
|
974 |
+
{
|
975 |
+
"epoch": 4.98,
|
976 |
+
"grad_norm": 0.001157268532551825,
|
977 |
+
"learning_rate": 2.066115702479339e-07,
|
978 |
+
"loss": 0.0019,
|
979 |
+
"step": 1340
|
980 |
+
},
|
981 |
+
{
|
982 |
+
"epoch": 5.0,
|
983 |
+
"eval_loss": 0.012286508455872536,
|
984 |
+
"eval_runtime": 42.4964,
|
985 |
+
"eval_samples_per_second": 90.008,
|
986 |
+
"eval_steps_per_second": 2.824,
|
987 |
+
"step": 1345
|
988 |
+
},
|
989 |
+
{
|
990 |
+
"epoch": 5.0,
|
991 |
+
"step": 1345,
|
992 |
+
"total_flos": 4.2775914245208883e+18,
|
993 |
+
"train_loss": 0.028647772959333094,
|
994 |
+
"train_runtime": 3141.2683,
|
995 |
+
"train_samples_per_second": 54.785,
|
996 |
+
"train_steps_per_second": 0.428
|
997 |
+
}
|
998 |
+
],
|
999 |
+
"logging_steps": 10,
|
1000 |
+
"max_steps": 1345,
|
1001 |
+
"num_input_tokens_seen": 0,
|
1002 |
+
"num_train_epochs": 5,
|
1003 |
+
"save_steps": 500,
|
1004 |
+
"total_flos": 4.2775914245208883e+18,
|
1005 |
+
"train_batch_size": 32,
|
1006 |
+
"trial_name": null,
|
1007 |
+
"trial_params": null
|
1008 |
+
}
|