End of training
Browse files- README.md +1 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- runs/Apr17_20-11-24_cf8be2e8a73d/events.out.tfevents.1713395353.cf8be2e8a73d.34.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +938 -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: 0.0200
|
21 |
|
22 |
## Model description
|
23 |
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_loss": 0.019963504746556282,
|
4 |
+
"eval_runtime": 89.035,
|
5 |
+
"eval_samples_per_second": 59.392,
|
6 |
+
"eval_steps_per_second": 1.247,
|
7 |
+
"total_flos": 5.914482579184435e+18,
|
8 |
+
"train_loss": 0.04452576920570385,
|
9 |
+
"train_runtime": 5366.5614,
|
10 |
+
"train_samples_per_second": 44.339,
|
11 |
+
"train_steps_per_second": 0.231
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_loss": 0.019963504746556282,
|
4 |
+
"eval_runtime": 89.035,
|
5 |
+
"eval_samples_per_second": 59.392,
|
6 |
+
"eval_steps_per_second": 1.247
|
7 |
+
}
|
runs/Apr17_20-11-24_cf8be2e8a73d/events.out.tfevents.1713395353.cf8be2e8a73d.34.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:05a65b76b5905816338c010536c233166e4a388ffe512b4b8428e5906a4cd01c
|
3 |
+
size 359
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"total_flos": 5.914482579184435e+18,
|
4 |
+
"train_loss": 0.04452576920570385,
|
5 |
+
"train_runtime": 5366.5614,
|
6 |
+
"train_samples_per_second": 44.339,
|
7 |
+
"train_steps_per_second": 0.231
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,938 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.019963504746556282,
|
3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-Kontur-competition-52K/checkpoint-496",
|
4 |
+
"epoch": 5.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1240,
|
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": 3.4303202629089355,
|
14 |
+
"learning_rate": 4.032258064516129e-06,
|
15 |
+
"loss": 0.7196,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.08,
|
20 |
+
"grad_norm": 2.8987796306610107,
|
21 |
+
"learning_rate": 8.064516129032258e-06,
|
22 |
+
"loss": 0.6457,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.12,
|
27 |
+
"grad_norm": 2.8838603496551514,
|
28 |
+
"learning_rate": 1.2096774193548388e-05,
|
29 |
+
"loss": 0.5219,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.16,
|
34 |
+
"grad_norm": 2.1233789920806885,
|
35 |
+
"learning_rate": 1.6129032258064517e-05,
|
36 |
+
"loss": 0.4099,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.2,
|
41 |
+
"grad_norm": 2.453838348388672,
|
42 |
+
"learning_rate": 2.0161290322580645e-05,
|
43 |
+
"loss": 0.2775,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.24,
|
48 |
+
"grad_norm": 5.15359354019165,
|
49 |
+
"learning_rate": 2.4193548387096777e-05,
|
50 |
+
"loss": 0.1655,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.28,
|
55 |
+
"grad_norm": 5.333171367645264,
|
56 |
+
"learning_rate": 2.822580645161291e-05,
|
57 |
+
"loss": 0.112,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.32,
|
62 |
+
"grad_norm": 3.0966763496398926,
|
63 |
+
"learning_rate": 3.2258064516129034e-05,
|
64 |
+
"loss": 0.0818,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.36,
|
69 |
+
"grad_norm": 13.956466674804688,
|
70 |
+
"learning_rate": 3.6290322580645165e-05,
|
71 |
+
"loss": 0.0884,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.4,
|
76 |
+
"grad_norm": 3.109863758087158,
|
77 |
+
"learning_rate": 4.032258064516129e-05,
|
78 |
+
"loss": 0.0755,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.44,
|
83 |
+
"grad_norm": 2.274998664855957,
|
84 |
+
"learning_rate": 4.435483870967742e-05,
|
85 |
+
"loss": 0.0562,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.48,
|
90 |
+
"grad_norm": 5.711202621459961,
|
91 |
+
"learning_rate": 4.8387096774193554e-05,
|
92 |
+
"loss": 0.0622,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.52,
|
97 |
+
"grad_norm": 9.838350296020508,
|
98 |
+
"learning_rate": 4.973118279569893e-05,
|
99 |
+
"loss": 0.0646,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.56,
|
104 |
+
"grad_norm": 8.744378089904785,
|
105 |
+
"learning_rate": 4.92831541218638e-05,
|
106 |
+
"loss": 0.0631,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.6,
|
111 |
+
"grad_norm": 13.287856101989746,
|
112 |
+
"learning_rate": 4.8835125448028677e-05,
|
113 |
+
"loss": 0.0732,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.65,
|
118 |
+
"grad_norm": 4.635748863220215,
|
119 |
+
"learning_rate": 4.8387096774193554e-05,
|
120 |
+
"loss": 0.049,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.69,
|
125 |
+
"grad_norm": 5.990776062011719,
|
126 |
+
"learning_rate": 4.7939068100358424e-05,
|
127 |
+
"loss": 0.0709,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.73,
|
132 |
+
"grad_norm": 6.079687595367432,
|
133 |
+
"learning_rate": 4.74910394265233e-05,
|
134 |
+
"loss": 0.0492,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.77,
|
139 |
+
"grad_norm": 1.6072505712509155,
|
140 |
+
"learning_rate": 4.704301075268818e-05,
|
141 |
+
"loss": 0.0527,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.81,
|
146 |
+
"grad_norm": 1.9706329107284546,
|
147 |
+
"learning_rate": 4.659498207885305e-05,
|
148 |
+
"loss": 0.0609,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.85,
|
153 |
+
"grad_norm": 5.284497261047363,
|
154 |
+
"learning_rate": 4.614695340501792e-05,
|
155 |
+
"loss": 0.0431,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.89,
|
160 |
+
"grad_norm": 3.325084924697876,
|
161 |
+
"learning_rate": 4.56989247311828e-05,
|
162 |
+
"loss": 0.0321,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.93,
|
167 |
+
"grad_norm": 9.077991485595703,
|
168 |
+
"learning_rate": 4.5250896057347674e-05,
|
169 |
+
"loss": 0.0375,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.97,
|
174 |
+
"grad_norm": 4.708994388580322,
|
175 |
+
"learning_rate": 4.4802867383512545e-05,
|
176 |
+
"loss": 0.0376,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 1.0,
|
181 |
+
"eval_loss": 0.06623569875955582,
|
182 |
+
"eval_runtime": 113.2602,
|
183 |
+
"eval_samples_per_second": 46.689,
|
184 |
+
"eval_steps_per_second": 0.98,
|
185 |
+
"step": 248
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 1.01,
|
189 |
+
"grad_norm": 4.394409656524658,
|
190 |
+
"learning_rate": 4.435483870967742e-05,
|
191 |
+
"loss": 0.0348,
|
192 |
+
"step": 250
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 1.05,
|
196 |
+
"grad_norm": 20.027359008789062,
|
197 |
+
"learning_rate": 4.390681003584229e-05,
|
198 |
+
"loss": 0.0685,
|
199 |
+
"step": 260
|
200 |
+
},
|
201 |
+
{
|
202 |
+
"epoch": 1.09,
|
203 |
+
"grad_norm": 2.6204428672790527,
|
204 |
+
"learning_rate": 4.345878136200717e-05,
|
205 |
+
"loss": 0.0445,
|
206 |
+
"step": 270
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 1.13,
|
210 |
+
"grad_norm": 6.732935905456543,
|
211 |
+
"learning_rate": 4.301075268817205e-05,
|
212 |
+
"loss": 0.0357,
|
213 |
+
"step": 280
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 1.17,
|
217 |
+
"grad_norm": 9.537686347961426,
|
218 |
+
"learning_rate": 4.256272401433692e-05,
|
219 |
+
"loss": 0.0367,
|
220 |
+
"step": 290
|
221 |
+
},
|
222 |
+
{
|
223 |
+
"epoch": 1.21,
|
224 |
+
"grad_norm": 4.934083938598633,
|
225 |
+
"learning_rate": 4.2114695340501795e-05,
|
226 |
+
"loss": 0.0249,
|
227 |
+
"step": 300
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 1.25,
|
231 |
+
"grad_norm": 2.7931134700775146,
|
232 |
+
"learning_rate": 4.166666666666667e-05,
|
233 |
+
"loss": 0.0312,
|
234 |
+
"step": 310
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 1.29,
|
238 |
+
"grad_norm": 1.5536555051803589,
|
239 |
+
"learning_rate": 4.121863799283154e-05,
|
240 |
+
"loss": 0.0329,
|
241 |
+
"step": 320
|
242 |
+
},
|
243 |
+
{
|
244 |
+
"epoch": 1.33,
|
245 |
+
"grad_norm": 4.139862060546875,
|
246 |
+
"learning_rate": 4.077060931899642e-05,
|
247 |
+
"loss": 0.0306,
|
248 |
+
"step": 330
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 1.37,
|
252 |
+
"grad_norm": 2.881141185760498,
|
253 |
+
"learning_rate": 4.032258064516129e-05,
|
254 |
+
"loss": 0.0272,
|
255 |
+
"step": 340
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 1.41,
|
259 |
+
"grad_norm": 4.688608646392822,
|
260 |
+
"learning_rate": 3.987455197132617e-05,
|
261 |
+
"loss": 0.0332,
|
262 |
+
"step": 350
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 1.45,
|
266 |
+
"grad_norm": 3.6917848587036133,
|
267 |
+
"learning_rate": 3.9426523297491045e-05,
|
268 |
+
"loss": 0.0365,
|
269 |
+
"step": 360
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 1.49,
|
273 |
+
"grad_norm": 3.620156764984131,
|
274 |
+
"learning_rate": 3.8978494623655915e-05,
|
275 |
+
"loss": 0.0232,
|
276 |
+
"step": 370
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 1.53,
|
280 |
+
"grad_norm": 5.679781913757324,
|
281 |
+
"learning_rate": 3.8530465949820786e-05,
|
282 |
+
"loss": 0.0259,
|
283 |
+
"step": 380
|
284 |
+
},
|
285 |
+
{
|
286 |
+
"epoch": 1.57,
|
287 |
+
"grad_norm": 3.670245885848999,
|
288 |
+
"learning_rate": 3.808243727598566e-05,
|
289 |
+
"loss": 0.0296,
|
290 |
+
"step": 390
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 1.61,
|
294 |
+
"grad_norm": 5.259506702423096,
|
295 |
+
"learning_rate": 3.763440860215054e-05,
|
296 |
+
"loss": 0.0155,
|
297 |
+
"step": 400
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 1.65,
|
301 |
+
"grad_norm": 2.921619176864624,
|
302 |
+
"learning_rate": 3.718637992831541e-05,
|
303 |
+
"loss": 0.0228,
|
304 |
+
"step": 410
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 1.69,
|
308 |
+
"grad_norm": 1.523906946182251,
|
309 |
+
"learning_rate": 3.673835125448029e-05,
|
310 |
+
"loss": 0.0197,
|
311 |
+
"step": 420
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 1.73,
|
315 |
+
"grad_norm": 4.975411891937256,
|
316 |
+
"learning_rate": 3.6290322580645165e-05,
|
317 |
+
"loss": 0.0296,
|
318 |
+
"step": 430
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 1.77,
|
322 |
+
"grad_norm": 3.5619239807128906,
|
323 |
+
"learning_rate": 3.5842293906810036e-05,
|
324 |
+
"loss": 0.0265,
|
325 |
+
"step": 440
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 1.81,
|
329 |
+
"grad_norm": 2.4044506549835205,
|
330 |
+
"learning_rate": 3.539426523297491e-05,
|
331 |
+
"loss": 0.0217,
|
332 |
+
"step": 450
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 1.85,
|
336 |
+
"grad_norm": 8.599495887756348,
|
337 |
+
"learning_rate": 3.494623655913979e-05,
|
338 |
+
"loss": 0.034,
|
339 |
+
"step": 460
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 1.9,
|
343 |
+
"grad_norm": 4.817391395568848,
|
344 |
+
"learning_rate": 3.449820788530466e-05,
|
345 |
+
"loss": 0.0244,
|
346 |
+
"step": 470
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 1.94,
|
350 |
+
"grad_norm": 0.9919489622116089,
|
351 |
+
"learning_rate": 3.405017921146954e-05,
|
352 |
+
"loss": 0.0285,
|
353 |
+
"step": 480
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 1.98,
|
357 |
+
"grad_norm": 1.3462491035461426,
|
358 |
+
"learning_rate": 3.360215053763441e-05,
|
359 |
+
"loss": 0.0213,
|
360 |
+
"step": 490
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 2.0,
|
364 |
+
"eval_loss": 0.019963504746556282,
|
365 |
+
"eval_runtime": 83.0975,
|
366 |
+
"eval_samples_per_second": 63.636,
|
367 |
+
"eval_steps_per_second": 1.336,
|
368 |
+
"step": 496
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 2.02,
|
372 |
+
"grad_norm": 2.6078598499298096,
|
373 |
+
"learning_rate": 3.3154121863799286e-05,
|
374 |
+
"loss": 0.029,
|
375 |
+
"step": 500
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 2.06,
|
379 |
+
"grad_norm": 0.9603502750396729,
|
380 |
+
"learning_rate": 3.270609318996416e-05,
|
381 |
+
"loss": 0.0263,
|
382 |
+
"step": 510
|
383 |
+
},
|
384 |
+
{
|
385 |
+
"epoch": 2.1,
|
386 |
+
"grad_norm": 11.797562599182129,
|
387 |
+
"learning_rate": 3.2258064516129034e-05,
|
388 |
+
"loss": 0.0256,
|
389 |
+
"step": 520
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 2.14,
|
393 |
+
"grad_norm": 6.010237693786621,
|
394 |
+
"learning_rate": 3.1810035842293904e-05,
|
395 |
+
"loss": 0.0259,
|
396 |
+
"step": 530
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 2.18,
|
400 |
+
"grad_norm": 11.067808151245117,
|
401 |
+
"learning_rate": 3.136200716845878e-05,
|
402 |
+
"loss": 0.0244,
|
403 |
+
"step": 540
|
404 |
+
},
|
405 |
+
{
|
406 |
+
"epoch": 2.22,
|
407 |
+
"grad_norm": 8.755197525024414,
|
408 |
+
"learning_rate": 3.091397849462366e-05,
|
409 |
+
"loss": 0.0172,
|
410 |
+
"step": 550
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 2.26,
|
414 |
+
"grad_norm": 5.274145603179932,
|
415 |
+
"learning_rate": 3.046594982078853e-05,
|
416 |
+
"loss": 0.0219,
|
417 |
+
"step": 560
|
418 |
+
},
|
419 |
+
{
|
420 |
+
"epoch": 2.3,
|
421 |
+
"grad_norm": 3.9025063514709473,
|
422 |
+
"learning_rate": 3.0017921146953403e-05,
|
423 |
+
"loss": 0.0121,
|
424 |
+
"step": 570
|
425 |
+
},
|
426 |
+
{
|
427 |
+
"epoch": 2.34,
|
428 |
+
"grad_norm": 0.16941958665847778,
|
429 |
+
"learning_rate": 2.9569892473118284e-05,
|
430 |
+
"loss": 0.0151,
|
431 |
+
"step": 580
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 2.38,
|
435 |
+
"grad_norm": 5.620195388793945,
|
436 |
+
"learning_rate": 2.9121863799283154e-05,
|
437 |
+
"loss": 0.02,
|
438 |
+
"step": 590
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 2.42,
|
442 |
+
"grad_norm": 4.685577869415283,
|
443 |
+
"learning_rate": 2.8673835125448028e-05,
|
444 |
+
"loss": 0.0212,
|
445 |
+
"step": 600
|
446 |
+
},
|
447 |
+
{
|
448 |
+
"epoch": 2.46,
|
449 |
+
"grad_norm": 0.9184539914131165,
|
450 |
+
"learning_rate": 2.822580645161291e-05,
|
451 |
+
"loss": 0.0243,
|
452 |
+
"step": 610
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 2.5,
|
456 |
+
"grad_norm": 4.498030662536621,
|
457 |
+
"learning_rate": 2.777777777777778e-05,
|
458 |
+
"loss": 0.0199,
|
459 |
+
"step": 620
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 2.54,
|
463 |
+
"grad_norm": 0.5963709354400635,
|
464 |
+
"learning_rate": 2.7329749103942653e-05,
|
465 |
+
"loss": 0.0189,
|
466 |
+
"step": 630
|
467 |
+
},
|
468 |
+
{
|
469 |
+
"epoch": 2.58,
|
470 |
+
"grad_norm": 0.7427639961242676,
|
471 |
+
"learning_rate": 2.6881720430107527e-05,
|
472 |
+
"loss": 0.01,
|
473 |
+
"step": 640
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 2.62,
|
477 |
+
"grad_norm": 1.3474091291427612,
|
478 |
+
"learning_rate": 2.6433691756272404e-05,
|
479 |
+
"loss": 0.0121,
|
480 |
+
"step": 650
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 2.66,
|
484 |
+
"grad_norm": 2.8091518878936768,
|
485 |
+
"learning_rate": 2.5985663082437278e-05,
|
486 |
+
"loss": 0.0252,
|
487 |
+
"step": 660
|
488 |
+
},
|
489 |
+
{
|
490 |
+
"epoch": 2.7,
|
491 |
+
"grad_norm": 1.0539556741714478,
|
492 |
+
"learning_rate": 2.5537634408602152e-05,
|
493 |
+
"loss": 0.0192,
|
494 |
+
"step": 670
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 2.74,
|
498 |
+
"grad_norm": 0.38558557629585266,
|
499 |
+
"learning_rate": 2.5089605734767026e-05,
|
500 |
+
"loss": 0.0115,
|
501 |
+
"step": 680
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 2.78,
|
505 |
+
"grad_norm": 2.709695339202881,
|
506 |
+
"learning_rate": 2.46415770609319e-05,
|
507 |
+
"loss": 0.0178,
|
508 |
+
"step": 690
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 2.82,
|
512 |
+
"grad_norm": 4.6164751052856445,
|
513 |
+
"learning_rate": 2.4193548387096777e-05,
|
514 |
+
"loss": 0.0118,
|
515 |
+
"step": 700
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 2.86,
|
519 |
+
"grad_norm": 2.706479549407959,
|
520 |
+
"learning_rate": 2.374551971326165e-05,
|
521 |
+
"loss": 0.0165,
|
522 |
+
"step": 710
|
523 |
+
},
|
524 |
+
{
|
525 |
+
"epoch": 2.9,
|
526 |
+
"grad_norm": 2.981581211090088,
|
527 |
+
"learning_rate": 2.3297491039426525e-05,
|
528 |
+
"loss": 0.0206,
|
529 |
+
"step": 720
|
530 |
+
},
|
531 |
+
{
|
532 |
+
"epoch": 2.94,
|
533 |
+
"grad_norm": 1.4225656986236572,
|
534 |
+
"learning_rate": 2.28494623655914e-05,
|
535 |
+
"loss": 0.0092,
|
536 |
+
"step": 730
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 2.98,
|
540 |
+
"grad_norm": 1.1643812656402588,
|
541 |
+
"learning_rate": 2.2401433691756272e-05,
|
542 |
+
"loss": 0.0094,
|
543 |
+
"step": 740
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 3.0,
|
547 |
+
"eval_loss": 0.13013091683387756,
|
548 |
+
"eval_runtime": 83.9945,
|
549 |
+
"eval_samples_per_second": 62.957,
|
550 |
+
"eval_steps_per_second": 1.322,
|
551 |
+
"step": 744
|
552 |
+
},
|
553 |
+
{
|
554 |
+
"epoch": 3.02,
|
555 |
+
"grad_norm": 2.4861252307891846,
|
556 |
+
"learning_rate": 2.1953405017921146e-05,
|
557 |
+
"loss": 0.0198,
|
558 |
+
"step": 750
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 3.06,
|
562 |
+
"grad_norm": 2.1413233280181885,
|
563 |
+
"learning_rate": 2.1505376344086024e-05,
|
564 |
+
"loss": 0.0142,
|
565 |
+
"step": 760
|
566 |
+
},
|
567 |
+
{
|
568 |
+
"epoch": 3.1,
|
569 |
+
"grad_norm": 2.2505719661712646,
|
570 |
+
"learning_rate": 2.1057347670250897e-05,
|
571 |
+
"loss": 0.0151,
|
572 |
+
"step": 770
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 3.15,
|
576 |
+
"grad_norm": 2.7683489322662354,
|
577 |
+
"learning_rate": 2.060931899641577e-05,
|
578 |
+
"loss": 0.0202,
|
579 |
+
"step": 780
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 3.19,
|
583 |
+
"grad_norm": 1.9615308046340942,
|
584 |
+
"learning_rate": 2.0161290322580645e-05,
|
585 |
+
"loss": 0.0076,
|
586 |
+
"step": 790
|
587 |
+
},
|
588 |
+
{
|
589 |
+
"epoch": 3.23,
|
590 |
+
"grad_norm": 2.115872621536255,
|
591 |
+
"learning_rate": 1.9713261648745522e-05,
|
592 |
+
"loss": 0.0133,
|
593 |
+
"step": 800
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 3.27,
|
597 |
+
"grad_norm": 0.3355218768119812,
|
598 |
+
"learning_rate": 1.9265232974910393e-05,
|
599 |
+
"loss": 0.0149,
|
600 |
+
"step": 810
|
601 |
+
},
|
602 |
+
{
|
603 |
+
"epoch": 3.31,
|
604 |
+
"grad_norm": 1.101788878440857,
|
605 |
+
"learning_rate": 1.881720430107527e-05,
|
606 |
+
"loss": 0.0168,
|
607 |
+
"step": 820
|
608 |
+
},
|
609 |
+
{
|
610 |
+
"epoch": 3.35,
|
611 |
+
"grad_norm": 4.441117763519287,
|
612 |
+
"learning_rate": 1.8369175627240144e-05,
|
613 |
+
"loss": 0.0101,
|
614 |
+
"step": 830
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 3.39,
|
618 |
+
"grad_norm": 0.4963151216506958,
|
619 |
+
"learning_rate": 1.7921146953405018e-05,
|
620 |
+
"loss": 0.0079,
|
621 |
+
"step": 840
|
622 |
+
},
|
623 |
+
{
|
624 |
+
"epoch": 3.43,
|
625 |
+
"grad_norm": 0.14219801127910614,
|
626 |
+
"learning_rate": 1.7473118279569895e-05,
|
627 |
+
"loss": 0.0117,
|
628 |
+
"step": 850
|
629 |
+
},
|
630 |
+
{
|
631 |
+
"epoch": 3.47,
|
632 |
+
"grad_norm": 0.04577568918466568,
|
633 |
+
"learning_rate": 1.702508960573477e-05,
|
634 |
+
"loss": 0.0065,
|
635 |
+
"step": 860
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 3.51,
|
639 |
+
"grad_norm": 3.448432207107544,
|
640 |
+
"learning_rate": 1.6577060931899643e-05,
|
641 |
+
"loss": 0.0141,
|
642 |
+
"step": 870
|
643 |
+
},
|
644 |
+
{
|
645 |
+
"epoch": 3.55,
|
646 |
+
"grad_norm": 0.43813198804855347,
|
647 |
+
"learning_rate": 1.6129032258064517e-05,
|
648 |
+
"loss": 0.0121,
|
649 |
+
"step": 880
|
650 |
+
},
|
651 |
+
{
|
652 |
+
"epoch": 3.59,
|
653 |
+
"grad_norm": 1.8220750093460083,
|
654 |
+
"learning_rate": 1.568100358422939e-05,
|
655 |
+
"loss": 0.0057,
|
656 |
+
"step": 890
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 3.63,
|
660 |
+
"grad_norm": 4.196650505065918,
|
661 |
+
"learning_rate": 1.5232974910394265e-05,
|
662 |
+
"loss": 0.0082,
|
663 |
+
"step": 900
|
664 |
+
},
|
665 |
+
{
|
666 |
+
"epoch": 3.67,
|
667 |
+
"grad_norm": 1.0396842956542969,
|
668 |
+
"learning_rate": 1.4784946236559142e-05,
|
669 |
+
"loss": 0.0113,
|
670 |
+
"step": 910
|
671 |
+
},
|
672 |
+
{
|
673 |
+
"epoch": 3.71,
|
674 |
+
"grad_norm": 3.6332926750183105,
|
675 |
+
"learning_rate": 1.4336917562724014e-05,
|
676 |
+
"loss": 0.0056,
|
677 |
+
"step": 920
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 3.75,
|
681 |
+
"grad_norm": 6.7421369552612305,
|
682 |
+
"learning_rate": 1.388888888888889e-05,
|
683 |
+
"loss": 0.0085,
|
684 |
+
"step": 930
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 3.79,
|
688 |
+
"grad_norm": 1.0675561428070068,
|
689 |
+
"learning_rate": 1.3440860215053763e-05,
|
690 |
+
"loss": 0.0061,
|
691 |
+
"step": 940
|
692 |
+
},
|
693 |
+
{
|
694 |
+
"epoch": 3.83,
|
695 |
+
"grad_norm": 3.8124799728393555,
|
696 |
+
"learning_rate": 1.2992831541218639e-05,
|
697 |
+
"loss": 0.0087,
|
698 |
+
"step": 950
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 3.87,
|
702 |
+
"grad_norm": 4.4822540283203125,
|
703 |
+
"learning_rate": 1.2544802867383513e-05,
|
704 |
+
"loss": 0.0048,
|
705 |
+
"step": 960
|
706 |
+
},
|
707 |
+
{
|
708 |
+
"epoch": 3.91,
|
709 |
+
"grad_norm": 1.812424898147583,
|
710 |
+
"learning_rate": 1.2096774193548388e-05,
|
711 |
+
"loss": 0.0074,
|
712 |
+
"step": 970
|
713 |
+
},
|
714 |
+
{
|
715 |
+
"epoch": 3.95,
|
716 |
+
"grad_norm": 6.007579803466797,
|
717 |
+
"learning_rate": 1.1648745519713262e-05,
|
718 |
+
"loss": 0.0116,
|
719 |
+
"step": 980
|
720 |
+
},
|
721 |
+
{
|
722 |
+
"epoch": 3.99,
|
723 |
+
"grad_norm": 0.2030833214521408,
|
724 |
+
"learning_rate": 1.1200716845878136e-05,
|
725 |
+
"loss": 0.0103,
|
726 |
+
"step": 990
|
727 |
+
},
|
728 |
+
{
|
729 |
+
"epoch": 4.0,
|
730 |
+
"eval_loss": 0.060427818447351456,
|
731 |
+
"eval_runtime": 84.6044,
|
732 |
+
"eval_samples_per_second": 62.503,
|
733 |
+
"eval_steps_per_second": 1.312,
|
734 |
+
"step": 992
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 4.03,
|
738 |
+
"grad_norm": 0.05641289800405502,
|
739 |
+
"learning_rate": 1.0752688172043012e-05,
|
740 |
+
"loss": 0.0085,
|
741 |
+
"step": 1000
|
742 |
+
},
|
743 |
+
{
|
744 |
+
"epoch": 4.07,
|
745 |
+
"grad_norm": 0.805799126625061,
|
746 |
+
"learning_rate": 1.0304659498207886e-05,
|
747 |
+
"loss": 0.0077,
|
748 |
+
"step": 1010
|
749 |
+
},
|
750 |
+
{
|
751 |
+
"epoch": 4.11,
|
752 |
+
"grad_norm": 4.972249984741211,
|
753 |
+
"learning_rate": 9.856630824372761e-06,
|
754 |
+
"loss": 0.0113,
|
755 |
+
"step": 1020
|
756 |
+
},
|
757 |
+
{
|
758 |
+
"epoch": 4.15,
|
759 |
+
"grad_norm": 0.03549932688474655,
|
760 |
+
"learning_rate": 9.408602150537635e-06,
|
761 |
+
"loss": 0.0052,
|
762 |
+
"step": 1030
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 4.19,
|
766 |
+
"grad_norm": 1.279232144355774,
|
767 |
+
"learning_rate": 8.960573476702509e-06,
|
768 |
+
"loss": 0.0077,
|
769 |
+
"step": 1040
|
770 |
+
},
|
771 |
+
{
|
772 |
+
"epoch": 4.23,
|
773 |
+
"grad_norm": 0.22537721693515778,
|
774 |
+
"learning_rate": 8.512544802867385e-06,
|
775 |
+
"loss": 0.0082,
|
776 |
+
"step": 1050
|
777 |
+
},
|
778 |
+
{
|
779 |
+
"epoch": 4.27,
|
780 |
+
"grad_norm": 1.7011988162994385,
|
781 |
+
"learning_rate": 8.064516129032258e-06,
|
782 |
+
"loss": 0.0036,
|
783 |
+
"step": 1060
|
784 |
+
},
|
785 |
+
{
|
786 |
+
"epoch": 4.31,
|
787 |
+
"grad_norm": 4.37709379196167,
|
788 |
+
"learning_rate": 7.616487455197132e-06,
|
789 |
+
"loss": 0.0041,
|
790 |
+
"step": 1070
|
791 |
+
},
|
792 |
+
{
|
793 |
+
"epoch": 4.35,
|
794 |
+
"grad_norm": 2.2503268718719482,
|
795 |
+
"learning_rate": 7.168458781362007e-06,
|
796 |
+
"loss": 0.0152,
|
797 |
+
"step": 1080
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 4.4,
|
801 |
+
"grad_norm": 3.610405683517456,
|
802 |
+
"learning_rate": 6.720430107526882e-06,
|
803 |
+
"loss": 0.0107,
|
804 |
+
"step": 1090
|
805 |
+
},
|
806 |
+
{
|
807 |
+
"epoch": 4.44,
|
808 |
+
"grad_norm": 0.22053413093090057,
|
809 |
+
"learning_rate": 6.2724014336917564e-06,
|
810 |
+
"loss": 0.0021,
|
811 |
+
"step": 1100
|
812 |
+
},
|
813 |
+
{
|
814 |
+
"epoch": 4.48,
|
815 |
+
"grad_norm": 8.507258415222168,
|
816 |
+
"learning_rate": 5.824372759856631e-06,
|
817 |
+
"loss": 0.0141,
|
818 |
+
"step": 1110
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 4.52,
|
822 |
+
"grad_norm": 5.034768581390381,
|
823 |
+
"learning_rate": 5.376344086021506e-06,
|
824 |
+
"loss": 0.0142,
|
825 |
+
"step": 1120
|
826 |
+
},
|
827 |
+
{
|
828 |
+
"epoch": 4.56,
|
829 |
+
"grad_norm": 0.22189994156360626,
|
830 |
+
"learning_rate": 4.928315412186381e-06,
|
831 |
+
"loss": 0.0094,
|
832 |
+
"step": 1130
|
833 |
+
},
|
834 |
+
{
|
835 |
+
"epoch": 4.6,
|
836 |
+
"grad_norm": 2.6591525077819824,
|
837 |
+
"learning_rate": 4.4802867383512545e-06,
|
838 |
+
"loss": 0.0039,
|
839 |
+
"step": 1140
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 4.64,
|
843 |
+
"grad_norm": 4.526949405670166,
|
844 |
+
"learning_rate": 4.032258064516129e-06,
|
845 |
+
"loss": 0.0032,
|
846 |
+
"step": 1150
|
847 |
+
},
|
848 |
+
{
|
849 |
+
"epoch": 4.68,
|
850 |
+
"grad_norm": 3.290435791015625,
|
851 |
+
"learning_rate": 3.5842293906810035e-06,
|
852 |
+
"loss": 0.0031,
|
853 |
+
"step": 1160
|
854 |
+
},
|
855 |
+
{
|
856 |
+
"epoch": 4.72,
|
857 |
+
"grad_norm": 0.6715773344039917,
|
858 |
+
"learning_rate": 3.1362007168458782e-06,
|
859 |
+
"loss": 0.0028,
|
860 |
+
"step": 1170
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 4.76,
|
864 |
+
"grad_norm": 0.7959145903587341,
|
865 |
+
"learning_rate": 2.688172043010753e-06,
|
866 |
+
"loss": 0.0063,
|
867 |
+
"step": 1180
|
868 |
+
},
|
869 |
+
{
|
870 |
+
"epoch": 4.8,
|
871 |
+
"grad_norm": 0.10818018019199371,
|
872 |
+
"learning_rate": 2.2401433691756272e-06,
|
873 |
+
"loss": 0.0045,
|
874 |
+
"step": 1190
|
875 |
+
},
|
876 |
+
{
|
877 |
+
"epoch": 4.84,
|
878 |
+
"grad_norm": 0.03141075000166893,
|
879 |
+
"learning_rate": 1.7921146953405017e-06,
|
880 |
+
"loss": 0.0068,
|
881 |
+
"step": 1200
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 4.88,
|
885 |
+
"grad_norm": 5.118274688720703,
|
886 |
+
"learning_rate": 1.3440860215053765e-06,
|
887 |
+
"loss": 0.0071,
|
888 |
+
"step": 1210
|
889 |
+
},
|
890 |
+
{
|
891 |
+
"epoch": 4.92,
|
892 |
+
"grad_norm": 1.3246444463729858,
|
893 |
+
"learning_rate": 8.960573476702509e-07,
|
894 |
+
"loss": 0.0026,
|
895 |
+
"step": 1220
|
896 |
+
},
|
897 |
+
{
|
898 |
+
"epoch": 4.96,
|
899 |
+
"grad_norm": 2.687753915786743,
|
900 |
+
"learning_rate": 4.4802867383512544e-07,
|
901 |
+
"loss": 0.0046,
|
902 |
+
"step": 1230
|
903 |
+
},
|
904 |
+
{
|
905 |
+
"epoch": 5.0,
|
906 |
+
"grad_norm": 0.166173055768013,
|
907 |
+
"learning_rate": 0.0,
|
908 |
+
"loss": 0.0073,
|
909 |
+
"step": 1240
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 5.0,
|
913 |
+
"eval_loss": 0.07113554328680038,
|
914 |
+
"eval_runtime": 85.5998,
|
915 |
+
"eval_samples_per_second": 61.776,
|
916 |
+
"eval_steps_per_second": 1.297,
|
917 |
+
"step": 1240
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 5.0,
|
921 |
+
"step": 1240,
|
922 |
+
"total_flos": 5.914482579184435e+18,
|
923 |
+
"train_loss": 0.04452576920570385,
|
924 |
+
"train_runtime": 5366.5614,
|
925 |
+
"train_samples_per_second": 44.339,
|
926 |
+
"train_steps_per_second": 0.231
|
927 |
+
}
|
928 |
+
],
|
929 |
+
"logging_steps": 10,
|
930 |
+
"max_steps": 1240,
|
931 |
+
"num_input_tokens_seen": 0,
|
932 |
+
"num_train_epochs": 5,
|
933 |
+
"save_steps": 500,
|
934 |
+
"total_flos": 5.914482579184435e+18,
|
935 |
+
"train_batch_size": 48,
|
936 |
+
"trial_name": null,
|
937 |
+
"trial_params": null
|
938 |
+
}
|