End of training
Browse files
README.md
CHANGED
@@ -2,6 +2,8 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: google/vit-base-patch16-224-in21k
|
4 |
tags:
|
|
|
|
|
5 |
- generated_from_trainer
|
6 |
metrics:
|
7 |
- accuracy
|
@@ -15,10 +17,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
# beans-vit-base-patch16-224-in21k
|
17 |
|
18 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
- Loss: 0.
|
21 |
-
- Accuracy: 0.
|
22 |
|
23 |
## Model description
|
24 |
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: google/vit-base-patch16-224-in21k
|
4 |
tags:
|
5 |
+
- image-classification
|
6 |
+
- vision
|
7 |
- generated_from_trainer
|
8 |
metrics:
|
9 |
- accuracy
|
|
|
17 |
|
18 |
# beans-vit-base-patch16-224-in21k
|
19 |
|
20 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.0657
|
23 |
+
- Accuracy: 0.9850
|
24 |
|
25 |
## Model description
|
26 |
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_accuracy": 0.9849624060150376,
|
4 |
+
"eval_loss": 0.06574746966362,
|
5 |
+
"eval_runtime": 1.0946,
|
6 |
+
"eval_samples_per_second": 121.505,
|
7 |
+
"eval_steps_per_second": 15.531,
|
8 |
+
"train_loss": 0.2192855260005364,
|
9 |
+
"train_runtime": 96.5552,
|
10 |
+
"train_samples_per_second": 53.544,
|
11 |
+
"train_steps_per_second": 6.732
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_accuracy": 0.9849624060150376,
|
4 |
+
"eval_loss": 0.06574746966362,
|
5 |
+
"eval_runtime": 1.0946,
|
6 |
+
"eval_samples_per_second": 121.505,
|
7 |
+
"eval_steps_per_second": 15.531
|
8 |
+
}
|
runs/Aug10_14-46-59_kwangryeol-OMEN-30L-Desktop-GT13-1xxx/events.out.tfevents.1723268952.kwangryeol-OMEN-30L-Desktop-GT13-1xxx.3538458.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:65d287f8f17eb1af7aa46db1ce4e3d137679c77c489cb59215d9e799a2c5af30
|
3 |
+
size 411
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"train_loss": 0.2192855260005364,
|
4 |
+
"train_runtime": 96.5552,
|
5 |
+
"train_samples_per_second": 53.544,
|
6 |
+
"train_steps_per_second": 6.732
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,530 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.06574746966362,
|
3 |
+
"best_model_checkpoint": "./beans_outputs/checkpoint-520",
|
4 |
+
"epoch": 5.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 650,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.08,
|
13 |
+
"grad_norm": 2.219902515411377,
|
14 |
+
"learning_rate": 1.9692307692307696e-05,
|
15 |
+
"loss": 1.0246,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.15,
|
20 |
+
"grad_norm": 1.863097906112671,
|
21 |
+
"learning_rate": 1.9384615384615386e-05,
|
22 |
+
"loss": 0.9458,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.23,
|
27 |
+
"grad_norm": 1.9998500347137451,
|
28 |
+
"learning_rate": 1.907692307692308e-05,
|
29 |
+
"loss": 0.8433,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.31,
|
34 |
+
"grad_norm": 2.666048288345337,
|
35 |
+
"learning_rate": 1.876923076923077e-05,
|
36 |
+
"loss": 0.6995,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.38,
|
41 |
+
"grad_norm": 4.233841419219971,
|
42 |
+
"learning_rate": 1.8461538461538465e-05,
|
43 |
+
"loss": 0.663,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.46,
|
48 |
+
"grad_norm": 2.192035436630249,
|
49 |
+
"learning_rate": 1.8153846153846155e-05,
|
50 |
+
"loss": 0.585,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.54,
|
55 |
+
"grad_norm": 1.8312418460845947,
|
56 |
+
"learning_rate": 1.784615384615385e-05,
|
57 |
+
"loss": 0.5225,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.62,
|
62 |
+
"grad_norm": 1.2257707118988037,
|
63 |
+
"learning_rate": 1.753846153846154e-05,
|
64 |
+
"loss": 0.3896,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.69,
|
69 |
+
"grad_norm": 2.517259359359741,
|
70 |
+
"learning_rate": 1.7230769230769234e-05,
|
71 |
+
"loss": 0.3501,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.77,
|
76 |
+
"grad_norm": 3.061749219894409,
|
77 |
+
"learning_rate": 1.6923076923076924e-05,
|
78 |
+
"loss": 0.3671,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.85,
|
83 |
+
"grad_norm": 1.758705973625183,
|
84 |
+
"learning_rate": 1.6615384615384618e-05,
|
85 |
+
"loss": 0.3307,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.92,
|
90 |
+
"grad_norm": 2.4563472270965576,
|
91 |
+
"learning_rate": 1.630769230769231e-05,
|
92 |
+
"loss": 0.2444,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 1.0,
|
97 |
+
"grad_norm": 7.120365619659424,
|
98 |
+
"learning_rate": 1.6000000000000003e-05,
|
99 |
+
"loss": 0.2814,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 1.0,
|
104 |
+
"eval_accuracy": 0.9624060150375939,
|
105 |
+
"eval_loss": 0.21468126773834229,
|
106 |
+
"eval_runtime": 1.1414,
|
107 |
+
"eval_samples_per_second": 116.529,
|
108 |
+
"eval_steps_per_second": 14.895,
|
109 |
+
"step": 130
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 1.08,
|
113 |
+
"grad_norm": 1.7768912315368652,
|
114 |
+
"learning_rate": 1.5692307692307693e-05,
|
115 |
+
"loss": 0.2,
|
116 |
+
"step": 140
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 1.15,
|
120 |
+
"grad_norm": 3.082838535308838,
|
121 |
+
"learning_rate": 1.5384615384615387e-05,
|
122 |
+
"loss": 0.2219,
|
123 |
+
"step": 150
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 1.23,
|
127 |
+
"grad_norm": 1.386346459388733,
|
128 |
+
"learning_rate": 1.5076923076923078e-05,
|
129 |
+
"loss": 0.2001,
|
130 |
+
"step": 160
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 1.31,
|
134 |
+
"grad_norm": 0.5072073340415955,
|
135 |
+
"learning_rate": 1.4769230769230772e-05,
|
136 |
+
"loss": 0.2183,
|
137 |
+
"step": 170
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 1.38,
|
141 |
+
"grad_norm": 0.6016378998756409,
|
142 |
+
"learning_rate": 1.4461538461538462e-05,
|
143 |
+
"loss": 0.2069,
|
144 |
+
"step": 180
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 1.46,
|
148 |
+
"grad_norm": 2.8826181888580322,
|
149 |
+
"learning_rate": 1.4153846153846156e-05,
|
150 |
+
"loss": 0.1539,
|
151 |
+
"step": 190
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 1.54,
|
155 |
+
"grad_norm": 0.43352940678596497,
|
156 |
+
"learning_rate": 1.3846153846153847e-05,
|
157 |
+
"loss": 0.2524,
|
158 |
+
"step": 200
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 1.62,
|
162 |
+
"grad_norm": 0.5819545984268188,
|
163 |
+
"learning_rate": 1.353846153846154e-05,
|
164 |
+
"loss": 0.1836,
|
165 |
+
"step": 210
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 1.69,
|
169 |
+
"grad_norm": 2.1941072940826416,
|
170 |
+
"learning_rate": 1.3230769230769231e-05,
|
171 |
+
"loss": 0.2198,
|
172 |
+
"step": 220
|
173 |
+
},
|
174 |
+
{
|
175 |
+
"epoch": 1.77,
|
176 |
+
"grad_norm": 1.9567605257034302,
|
177 |
+
"learning_rate": 1.2923076923076925e-05,
|
178 |
+
"loss": 0.1557,
|
179 |
+
"step": 230
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 1.85,
|
183 |
+
"grad_norm": 2.7541744709014893,
|
184 |
+
"learning_rate": 1.2615384615384616e-05,
|
185 |
+
"loss": 0.2846,
|
186 |
+
"step": 240
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 1.92,
|
190 |
+
"grad_norm": 0.38941043615341187,
|
191 |
+
"learning_rate": 1.230769230769231e-05,
|
192 |
+
"loss": 0.1364,
|
193 |
+
"step": 250
|
194 |
+
},
|
195 |
+
{
|
196 |
+
"epoch": 2.0,
|
197 |
+
"grad_norm": 0.34662145376205444,
|
198 |
+
"learning_rate": 1.2e-05,
|
199 |
+
"loss": 0.1326,
|
200 |
+
"step": 260
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 2.0,
|
204 |
+
"eval_accuracy": 0.9699248120300752,
|
205 |
+
"eval_loss": 0.12728814780712128,
|
206 |
+
"eval_runtime": 1.0735,
|
207 |
+
"eval_samples_per_second": 123.898,
|
208 |
+
"eval_steps_per_second": 15.837,
|
209 |
+
"step": 260
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 2.08,
|
213 |
+
"grad_norm": 0.4577018618583679,
|
214 |
+
"learning_rate": 1.1692307692307694e-05,
|
215 |
+
"loss": 0.0934,
|
216 |
+
"step": 270
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 2.15,
|
220 |
+
"grad_norm": 0.3331770598888397,
|
221 |
+
"learning_rate": 1.1384615384615385e-05,
|
222 |
+
"loss": 0.1416,
|
223 |
+
"step": 280
|
224 |
+
},
|
225 |
+
{
|
226 |
+
"epoch": 2.23,
|
227 |
+
"grad_norm": 7.138263702392578,
|
228 |
+
"learning_rate": 1.1076923076923079e-05,
|
229 |
+
"loss": 0.1296,
|
230 |
+
"step": 290
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 2.31,
|
234 |
+
"grad_norm": 8.594639778137207,
|
235 |
+
"learning_rate": 1.076923076923077e-05,
|
236 |
+
"loss": 0.1417,
|
237 |
+
"step": 300
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 2.38,
|
241 |
+
"grad_norm": 5.681203365325928,
|
242 |
+
"learning_rate": 1.0461538461538463e-05,
|
243 |
+
"loss": 0.1707,
|
244 |
+
"step": 310
|
245 |
+
},
|
246 |
+
{
|
247 |
+
"epoch": 2.46,
|
248 |
+
"grad_norm": 0.3230597674846649,
|
249 |
+
"learning_rate": 1.0153846153846154e-05,
|
250 |
+
"loss": 0.1496,
|
251 |
+
"step": 320
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 2.54,
|
255 |
+
"grad_norm": 6.502106189727783,
|
256 |
+
"learning_rate": 9.846153846153848e-06,
|
257 |
+
"loss": 0.1758,
|
258 |
+
"step": 330
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 2.62,
|
262 |
+
"grad_norm": 1.060750126838684,
|
263 |
+
"learning_rate": 9.53846153846154e-06,
|
264 |
+
"loss": 0.0753,
|
265 |
+
"step": 340
|
266 |
+
},
|
267 |
+
{
|
268 |
+
"epoch": 2.69,
|
269 |
+
"grad_norm": 0.3386625051498413,
|
270 |
+
"learning_rate": 9.230769230769232e-06,
|
271 |
+
"loss": 0.1345,
|
272 |
+
"step": 350
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 2.77,
|
276 |
+
"grad_norm": 6.454658031463623,
|
277 |
+
"learning_rate": 8.923076923076925e-06,
|
278 |
+
"loss": 0.117,
|
279 |
+
"step": 360
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 2.85,
|
283 |
+
"grad_norm": 0.3338567614555359,
|
284 |
+
"learning_rate": 8.615384615384617e-06,
|
285 |
+
"loss": 0.1511,
|
286 |
+
"step": 370
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 2.92,
|
290 |
+
"grad_norm": 1.0864334106445312,
|
291 |
+
"learning_rate": 8.307692307692309e-06,
|
292 |
+
"loss": 0.1284,
|
293 |
+
"step": 380
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 3.0,
|
297 |
+
"grad_norm": 0.30264168977737427,
|
298 |
+
"learning_rate": 8.000000000000001e-06,
|
299 |
+
"loss": 0.1413,
|
300 |
+
"step": 390
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 3.0,
|
304 |
+
"eval_accuracy": 0.9774436090225563,
|
305 |
+
"eval_loss": 0.09794322401285172,
|
306 |
+
"eval_runtime": 1.1225,
|
307 |
+
"eval_samples_per_second": 118.486,
|
308 |
+
"eval_steps_per_second": 15.145,
|
309 |
+
"step": 390
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 3.08,
|
313 |
+
"grad_norm": 3.8275489807128906,
|
314 |
+
"learning_rate": 7.692307692307694e-06,
|
315 |
+
"loss": 0.1396,
|
316 |
+
"step": 400
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 3.15,
|
320 |
+
"grad_norm": 6.871574401855469,
|
321 |
+
"learning_rate": 7.384615384615386e-06,
|
322 |
+
"loss": 0.1577,
|
323 |
+
"step": 410
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 3.23,
|
327 |
+
"grad_norm": 8.653078079223633,
|
328 |
+
"learning_rate": 7.076923076923078e-06,
|
329 |
+
"loss": 0.1429,
|
330 |
+
"step": 420
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 3.31,
|
334 |
+
"grad_norm": 0.23040761053562164,
|
335 |
+
"learning_rate": 6.76923076923077e-06,
|
336 |
+
"loss": 0.0676,
|
337 |
+
"step": 430
|
338 |
+
},
|
339 |
+
{
|
340 |
+
"epoch": 3.38,
|
341 |
+
"grad_norm": 4.652414798736572,
|
342 |
+
"learning_rate": 6.461538461538463e-06,
|
343 |
+
"loss": 0.1215,
|
344 |
+
"step": 440
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 3.46,
|
348 |
+
"grad_norm": 11.157665252685547,
|
349 |
+
"learning_rate": 6.153846153846155e-06,
|
350 |
+
"loss": 0.0796,
|
351 |
+
"step": 450
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 3.54,
|
355 |
+
"grad_norm": 4.33862829208374,
|
356 |
+
"learning_rate": 5.846153846153847e-06,
|
357 |
+
"loss": 0.0967,
|
358 |
+
"step": 460
|
359 |
+
},
|
360 |
+
{
|
361 |
+
"epoch": 3.62,
|
362 |
+
"grad_norm": 7.9337921142578125,
|
363 |
+
"learning_rate": 5.538461538461539e-06,
|
364 |
+
"loss": 0.0716,
|
365 |
+
"step": 470
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 3.69,
|
369 |
+
"grad_norm": 0.22305496037006378,
|
370 |
+
"learning_rate": 5.230769230769232e-06,
|
371 |
+
"loss": 0.0902,
|
372 |
+
"step": 480
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 3.77,
|
376 |
+
"grad_norm": 0.2107217162847519,
|
377 |
+
"learning_rate": 4.923076923076924e-06,
|
378 |
+
"loss": 0.072,
|
379 |
+
"step": 490
|
380 |
+
},
|
381 |
+
{
|
382 |
+
"epoch": 3.85,
|
383 |
+
"grad_norm": 0.9404855370521545,
|
384 |
+
"learning_rate": 4.615384615384616e-06,
|
385 |
+
"loss": 0.1483,
|
386 |
+
"step": 500
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 3.92,
|
390 |
+
"grad_norm": 0.21295365691184998,
|
391 |
+
"learning_rate": 4.307692307692308e-06,
|
392 |
+
"loss": 0.1349,
|
393 |
+
"step": 510
|
394 |
+
},
|
395 |
+
{
|
396 |
+
"epoch": 4.0,
|
397 |
+
"grad_norm": 0.3001614511013031,
|
398 |
+
"learning_rate": 4.000000000000001e-06,
|
399 |
+
"loss": 0.0855,
|
400 |
+
"step": 520
|
401 |
+
},
|
402 |
+
{
|
403 |
+
"epoch": 4.0,
|
404 |
+
"eval_accuracy": 0.9849624060150376,
|
405 |
+
"eval_loss": 0.06574746966362,
|
406 |
+
"eval_runtime": 1.128,
|
407 |
+
"eval_samples_per_second": 117.907,
|
408 |
+
"eval_steps_per_second": 15.071,
|
409 |
+
"step": 520
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 4.08,
|
413 |
+
"grad_norm": 8.45413589477539,
|
414 |
+
"learning_rate": 3.692307692307693e-06,
|
415 |
+
"loss": 0.0996,
|
416 |
+
"step": 530
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 4.15,
|
420 |
+
"grad_norm": 1.4845136404037476,
|
421 |
+
"learning_rate": 3.384615384615385e-06,
|
422 |
+
"loss": 0.1602,
|
423 |
+
"step": 540
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 4.23,
|
427 |
+
"grad_norm": 0.35962048172950745,
|
428 |
+
"learning_rate": 3.0769230769230774e-06,
|
429 |
+
"loss": 0.1495,
|
430 |
+
"step": 550
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 4.31,
|
434 |
+
"grad_norm": 1.3418010473251343,
|
435 |
+
"learning_rate": 2.7692307692307697e-06,
|
436 |
+
"loss": 0.0511,
|
437 |
+
"step": 560
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 4.38,
|
441 |
+
"grad_norm": 0.33496251702308655,
|
442 |
+
"learning_rate": 2.461538461538462e-06,
|
443 |
+
"loss": 0.1355,
|
444 |
+
"step": 570
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 4.46,
|
448 |
+
"grad_norm": 0.2655811011791229,
|
449 |
+
"learning_rate": 2.153846153846154e-06,
|
450 |
+
"loss": 0.0836,
|
451 |
+
"step": 580
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 4.54,
|
455 |
+
"grad_norm": 10.997761726379395,
|
456 |
+
"learning_rate": 1.8461538461538465e-06,
|
457 |
+
"loss": 0.0673,
|
458 |
+
"step": 590
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 4.62,
|
462 |
+
"grad_norm": 2.0622777938842773,
|
463 |
+
"learning_rate": 1.5384615384615387e-06,
|
464 |
+
"loss": 0.1054,
|
465 |
+
"step": 600
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 4.69,
|
469 |
+
"grad_norm": 5.778710842132568,
|
470 |
+
"learning_rate": 1.230769230769231e-06,
|
471 |
+
"loss": 0.092,
|
472 |
+
"step": 610
|
473 |
+
},
|
474 |
+
{
|
475 |
+
"epoch": 4.77,
|
476 |
+
"grad_norm": 0.2569364309310913,
|
477 |
+
"learning_rate": 9.230769230769232e-07,
|
478 |
+
"loss": 0.075,
|
479 |
+
"step": 620
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 4.85,
|
483 |
+
"grad_norm": 0.21514290571212769,
|
484 |
+
"learning_rate": 6.153846153846155e-07,
|
485 |
+
"loss": 0.0612,
|
486 |
+
"step": 630
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 4.92,
|
490 |
+
"grad_norm": 0.23064370453357697,
|
491 |
+
"learning_rate": 3.0769230769230774e-07,
|
492 |
+
"loss": 0.0783,
|
493 |
+
"step": 640
|
494 |
+
},
|
495 |
+
{
|
496 |
+
"epoch": 5.0,
|
497 |
+
"grad_norm": 0.6160356402397156,
|
498 |
+
"learning_rate": 0.0,
|
499 |
+
"loss": 0.1234,
|
500 |
+
"step": 650
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 5.0,
|
504 |
+
"eval_accuracy": 0.9774436090225563,
|
505 |
+
"eval_loss": 0.08280191570520401,
|
506 |
+
"eval_runtime": 1.0692,
|
507 |
+
"eval_samples_per_second": 124.39,
|
508 |
+
"eval_steps_per_second": 15.899,
|
509 |
+
"step": 650
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 5.0,
|
513 |
+
"step": 650,
|
514 |
+
"total_flos": 4.006371770595533e+17,
|
515 |
+
"train_loss": 0.2192855260005364,
|
516 |
+
"train_runtime": 96.5552,
|
517 |
+
"train_samples_per_second": 53.544,
|
518 |
+
"train_steps_per_second": 6.732
|
519 |
+
}
|
520 |
+
],
|
521 |
+
"logging_steps": 10,
|
522 |
+
"max_steps": 650,
|
523 |
+
"num_input_tokens_seen": 0,
|
524 |
+
"num_train_epochs": 5,
|
525 |
+
"save_steps": 500,
|
526 |
+
"total_flos": 4.006371770595533e+17,
|
527 |
+
"train_batch_size": 8,
|
528 |
+
"trial_name": null,
|
529 |
+
"trial_params": null
|
530 |
+
}
|