🍻 cheers
Browse files- README.md +3 -2
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/Apr16_13-47-24_a962527840ae/events.out.tfevents.1713275740.a962527840ae.34.1 +3 -0
- train_results.json +6 -6
- trainer_state.json +74 -213
README.md
CHANGED
@@ -2,6 +2,7 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: google/vit-base-patch16-224-in21k
|
4 |
tags:
|
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
- imagefolder
|
@@ -14,7 +15,7 @@ model-index:
|
|
14 |
name: Image Classification
|
15 |
type: image-classification
|
16 |
dataset:
|
17 |
-
name:
|
18 |
type: imagefolder
|
19 |
config: default
|
20 |
split: train
|
@@ -30,7 +31,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
30 |
|
31 |
# Action_model
|
32 |
|
33 |
-
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
|
34 |
It achieves the following results on the evaluation set:
|
35 |
- Loss: 0.7479
|
36 |
- Accuracy: 0.8095
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: google/vit-base-patch16-224-in21k
|
4 |
tags:
|
5 |
+
- image-classification
|
6 |
- generated_from_trainer
|
7 |
datasets:
|
8 |
- imagefolder
|
|
|
15 |
name: Image Classification
|
16 |
type: image-classification
|
17 |
dataset:
|
18 |
+
name: action_class
|
19 |
type: imagefolder
|
20 |
config: default
|
21 |
split: train
|
|
|
31 |
|
32 |
# Action_model
|
33 |
|
34 |
+
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 action_class dataset.
|
35 |
It achieves the following results on the evaluation set:
|
36 |
- Loss: 0.7479
|
37 |
- Accuracy: 0.8095
|
all_results.json
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime":
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
-
"total_flos":
|
9 |
-
"train_loss":
|
10 |
-
"train_runtime":
|
11 |
-
"train_samples_per_second":
|
12 |
-
"train_steps_per_second":
|
13 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_accuracy": 0.8095238095238095,
|
4 |
+
"eval_loss": 0.747850775718689,
|
5 |
+
"eval_runtime": 14.6564,
|
6 |
+
"eval_samples_per_second": 71.641,
|
7 |
+
"eval_steps_per_second": 9.006,
|
8 |
+
"total_flos": 7.776878731479245e+17,
|
9 |
+
"train_loss": 1.0910153146002703,
|
10 |
+
"train_runtime": 357.1216,
|
11 |
+
"train_samples_per_second": 28.1,
|
12 |
+
"train_steps_per_second": 0.879
|
13 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime":
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"eval_accuracy": 0.8095238095238095,
|
4 |
+
"eval_loss": 0.747850775718689,
|
5 |
+
"eval_runtime": 14.6564,
|
6 |
+
"eval_samples_per_second": 71.641,
|
7 |
+
"eval_steps_per_second": 9.006
|
8 |
}
|
runs/Apr16_13-47-24_a962527840ae/events.out.tfevents.1713275740.a962527840ae.34.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5c2014389d8906c0e5e1087f2252c30400d17433e90d1c62c17dfd64c22e33f3
|
3 |
+
size 411
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"total_flos":
|
4 |
-
"train_loss":
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 7.776878731479245e+17,
|
4 |
+
"train_loss": 1.0910153146002703,
|
5 |
+
"train_runtime": 357.1216,
|
6 |
+
"train_samples_per_second": 28.1,
|
7 |
+
"train_steps_per_second": 0.879
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,300 +1,161 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "Action_model/checkpoint-
|
4 |
-
"epoch":
|
5 |
"eval_steps": 100,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.06,
|
13 |
-
"grad_norm":
|
14 |
-
"learning_rate": 9.
|
15 |
-
"loss":
|
16 |
"step": 20
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.13,
|
20 |
-
"grad_norm":
|
21 |
-
"learning_rate":
|
22 |
-
"loss":
|
23 |
"step": 40
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.19,
|
27 |
-
"grad_norm":
|
28 |
-
"learning_rate":
|
29 |
-
"loss":
|
30 |
"step": 60
|
31 |
},
|
32 |
{
|
33 |
"epoch": 0.25,
|
34 |
-
"grad_norm":
|
35 |
-
"learning_rate":
|
36 |
-
"loss":
|
37 |
"step": 80
|
38 |
},
|
39 |
{
|
40 |
"epoch": 0.32,
|
41 |
-
"grad_norm":
|
42 |
-
"learning_rate":
|
43 |
-
"loss":
|
44 |
"step": 100
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.32,
|
48 |
-
"eval_accuracy": 0.
|
49 |
-
"eval_loss":
|
50 |
-
"eval_runtime":
|
51 |
-
"eval_samples_per_second":
|
52 |
-
"eval_steps_per_second":
|
53 |
"step": 100
|
54 |
},
|
55 |
{
|
56 |
"epoch": 0.38,
|
57 |
-
"grad_norm":
|
58 |
-
"learning_rate":
|
59 |
-
"loss":
|
60 |
"step": 120
|
61 |
},
|
62 |
{
|
63 |
"epoch": 0.45,
|
64 |
-
"grad_norm":
|
65 |
-
"learning_rate":
|
66 |
-
"loss":
|
67 |
"step": 140
|
68 |
},
|
69 |
{
|
70 |
"epoch": 0.51,
|
71 |
-
"grad_norm":
|
72 |
-
"learning_rate":
|
73 |
-
"loss": 0.
|
74 |
"step": 160
|
75 |
},
|
76 |
{
|
77 |
"epoch": 0.57,
|
78 |
-
"grad_norm":
|
79 |
-
"learning_rate":
|
80 |
-
"loss": 0.
|
81 |
"step": 180
|
82 |
},
|
83 |
{
|
84 |
"epoch": 0.64,
|
85 |
-
"grad_norm": 3.
|
86 |
-
"learning_rate":
|
87 |
-
"loss": 0.
|
88 |
"step": 200
|
89 |
},
|
90 |
{
|
91 |
"epoch": 0.64,
|
92 |
-
"eval_accuracy": 0.
|
93 |
-
"eval_loss": 0.
|
94 |
-
"eval_runtime": 14.
|
95 |
-
"eval_samples_per_second":
|
96 |
-
"eval_steps_per_second":
|
97 |
"step": 200
|
98 |
},
|
99 |
{
|
100 |
"epoch": 0.7,
|
101 |
-
"grad_norm":
|
102 |
-
"learning_rate":
|
103 |
-
"loss": 0.
|
104 |
"step": 220
|
105 |
},
|
106 |
{
|
107 |
"epoch": 0.76,
|
108 |
-
"grad_norm":
|
109 |
-
"learning_rate":
|
110 |
-
"loss": 0.
|
111 |
"step": 240
|
112 |
},
|
113 |
{
|
114 |
"epoch": 0.83,
|
115 |
-
"grad_norm":
|
116 |
-
"learning_rate":
|
117 |
-
"loss": 0.
|
118 |
"step": 260
|
119 |
},
|
120 |
{
|
121 |
"epoch": 0.89,
|
122 |
-
"grad_norm":
|
123 |
-
"learning_rate":
|
124 |
-
"loss": 0.
|
125 |
"step": 280
|
126 |
},
|
127 |
{
|
128 |
"epoch": 0.96,
|
129 |
-
"grad_norm":
|
130 |
-
"learning_rate":
|
131 |
-
"loss": 0.
|
132 |
"step": 300
|
133 |
},
|
134 |
{
|
135 |
"epoch": 0.96,
|
136 |
-
"eval_accuracy": 0.
|
137 |
-
"eval_loss": 0.
|
138 |
-
"eval_runtime": 14.
|
139 |
-
"eval_samples_per_second": 71.
|
140 |
-
"eval_steps_per_second": 8.
|
141 |
"step": 300
|
142 |
},
|
143 |
{
|
144 |
-
"epoch": 1.
|
145 |
-
"
|
146 |
-
"
|
147 |
-
"
|
148 |
-
"
|
149 |
-
|
150 |
-
|
151 |
-
"epoch": 1.08,
|
152 |
-
"grad_norm": 4.398645401000977,
|
153 |
-
"learning_rate": 4.601910828025478e-05,
|
154 |
-
"loss": 0.2851,
|
155 |
-
"step": 340
|
156 |
-
},
|
157 |
-
{
|
158 |
-
"epoch": 1.15,
|
159 |
-
"grad_norm": 4.354762077331543,
|
160 |
-
"learning_rate": 4.2834394904458604e-05,
|
161 |
-
"loss": 0.3029,
|
162 |
-
"step": 360
|
163 |
-
},
|
164 |
-
{
|
165 |
-
"epoch": 1.21,
|
166 |
-
"grad_norm": 1.9630823135375977,
|
167 |
-
"learning_rate": 3.964968152866242e-05,
|
168 |
-
"loss": 0.2774,
|
169 |
-
"step": 380
|
170 |
-
},
|
171 |
-
{
|
172 |
-
"epoch": 1.27,
|
173 |
-
"grad_norm": 2.4799087047576904,
|
174 |
-
"learning_rate": 3.646496815286624e-05,
|
175 |
-
"loss": 0.2812,
|
176 |
-
"step": 400
|
177 |
-
},
|
178 |
-
{
|
179 |
-
"epoch": 1.27,
|
180 |
-
"eval_accuracy": 0.7676190476190476,
|
181 |
-
"eval_loss": 0.7738763689994812,
|
182 |
-
"eval_runtime": 14.6761,
|
183 |
-
"eval_samples_per_second": 71.545,
|
184 |
-
"eval_steps_per_second": 8.994,
|
185 |
-
"step": 400
|
186 |
-
},
|
187 |
-
{
|
188 |
-
"epoch": 1.34,
|
189 |
-
"grad_norm": 5.377834320068359,
|
190 |
-
"learning_rate": 3.328025477707007e-05,
|
191 |
-
"loss": 0.2768,
|
192 |
-
"step": 420
|
193 |
-
},
|
194 |
-
{
|
195 |
-
"epoch": 1.4,
|
196 |
-
"grad_norm": 4.753965854644775,
|
197 |
-
"learning_rate": 3.0095541401273885e-05,
|
198 |
-
"loss": 0.3242,
|
199 |
-
"step": 440
|
200 |
-
},
|
201 |
-
{
|
202 |
-
"epoch": 1.46,
|
203 |
-
"grad_norm": 4.039095401763916,
|
204 |
-
"learning_rate": 2.6910828025477707e-05,
|
205 |
-
"loss": 0.2168,
|
206 |
-
"step": 460
|
207 |
-
},
|
208 |
-
{
|
209 |
-
"epoch": 1.53,
|
210 |
-
"grad_norm": 3.443063497543335,
|
211 |
-
"learning_rate": 2.372611464968153e-05,
|
212 |
-
"loss": 0.2584,
|
213 |
-
"step": 480
|
214 |
-
},
|
215 |
-
{
|
216 |
-
"epoch": 1.59,
|
217 |
-
"grad_norm": 1.6640291213989258,
|
218 |
-
"learning_rate": 2.054140127388535e-05,
|
219 |
-
"loss": 0.251,
|
220 |
-
"step": 500
|
221 |
-
},
|
222 |
-
{
|
223 |
-
"epoch": 1.59,
|
224 |
-
"eval_accuracy": 0.7857142857142857,
|
225 |
-
"eval_loss": 0.7165716886520386,
|
226 |
-
"eval_runtime": 14.6301,
|
227 |
-
"eval_samples_per_second": 71.77,
|
228 |
-
"eval_steps_per_second": 9.022,
|
229 |
-
"step": 500
|
230 |
-
},
|
231 |
-
{
|
232 |
-
"epoch": 1.66,
|
233 |
-
"grad_norm": 3.264369487762451,
|
234 |
-
"learning_rate": 1.7356687898089173e-05,
|
235 |
-
"loss": 0.2804,
|
236 |
-
"step": 520
|
237 |
-
},
|
238 |
-
{
|
239 |
-
"epoch": 1.72,
|
240 |
-
"grad_norm": 3.547678232192993,
|
241 |
-
"learning_rate": 1.4171974522292993e-05,
|
242 |
-
"loss": 0.2737,
|
243 |
-
"step": 540
|
244 |
-
},
|
245 |
-
{
|
246 |
-
"epoch": 1.78,
|
247 |
-
"grad_norm": 5.212786674499512,
|
248 |
-
"learning_rate": 1.0987261146496815e-05,
|
249 |
-
"loss": 0.2503,
|
250 |
-
"step": 560
|
251 |
-
},
|
252 |
-
{
|
253 |
-
"epoch": 1.85,
|
254 |
-
"grad_norm": 3.4162983894348145,
|
255 |
-
"learning_rate": 7.802547770700637e-06,
|
256 |
-
"loss": 0.253,
|
257 |
-
"step": 580
|
258 |
-
},
|
259 |
-
{
|
260 |
-
"epoch": 1.91,
|
261 |
-
"grad_norm": 5.042104244232178,
|
262 |
-
"learning_rate": 4.6178343949044585e-06,
|
263 |
-
"loss": 0.2425,
|
264 |
-
"step": 600
|
265 |
-
},
|
266 |
-
{
|
267 |
-
"epoch": 1.91,
|
268 |
-
"eval_accuracy": 0.7742857142857142,
|
269 |
-
"eval_loss": 0.7691510915756226,
|
270 |
-
"eval_runtime": 14.587,
|
271 |
-
"eval_samples_per_second": 71.982,
|
272 |
-
"eval_steps_per_second": 9.049,
|
273 |
-
"step": 600
|
274 |
-
},
|
275 |
-
{
|
276 |
-
"epoch": 1.97,
|
277 |
-
"grad_norm": 5.002872467041016,
|
278 |
-
"learning_rate": 1.4331210191082802e-06,
|
279 |
-
"loss": 0.2493,
|
280 |
-
"step": 620
|
281 |
-
},
|
282 |
-
{
|
283 |
-
"epoch": 2.0,
|
284 |
-
"step": 628,
|
285 |
-
"total_flos": 1.555375746295849e+18,
|
286 |
-
"train_loss": 0.33379530375170857,
|
287 |
-
"train_runtime": 590.5675,
|
288 |
-
"train_samples_per_second": 33.984,
|
289 |
-
"train_steps_per_second": 1.063
|
290 |
}
|
291 |
],
|
292 |
"logging_steps": 20,
|
293 |
-
"max_steps":
|
294 |
"num_input_tokens_seen": 0,
|
295 |
-
"num_train_epochs":
|
296 |
"save_steps": 100,
|
297 |
-
"total_flos":
|
298 |
"train_batch_size": 32,
|
299 |
"trial_name": null,
|
300 |
"trial_params": null
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.747850775718689,
|
3 |
+
"best_model_checkpoint": "Action_model/checkpoint-300",
|
4 |
+
"epoch": 1.0,
|
5 |
"eval_steps": 100,
|
6 |
+
"global_step": 314,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.06,
|
13 |
+
"grad_norm": 1.3472708463668823,
|
14 |
+
"learning_rate": 9.363057324840766e-05,
|
15 |
+
"loss": 2.1856,
|
16 |
"step": 20
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.13,
|
20 |
+
"grad_norm": 1.6540354490280151,
|
21 |
+
"learning_rate": 8.726114649681529e-05,
|
22 |
+
"loss": 1.9121,
|
23 |
"step": 40
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.19,
|
27 |
+
"grad_norm": 1.9076273441314697,
|
28 |
+
"learning_rate": 8.089171974522294e-05,
|
29 |
+
"loss": 1.5695,
|
30 |
"step": 60
|
31 |
},
|
32 |
{
|
33 |
"epoch": 0.25,
|
34 |
+
"grad_norm": 2.171828508377075,
|
35 |
+
"learning_rate": 7.452229299363057e-05,
|
36 |
+
"loss": 1.3273,
|
37 |
"step": 80
|
38 |
},
|
39 |
{
|
40 |
"epoch": 0.32,
|
41 |
+
"grad_norm": 3.170124053955078,
|
42 |
+
"learning_rate": 6.815286624203822e-05,
|
43 |
+
"loss": 1.1778,
|
44 |
"step": 100
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.32,
|
48 |
+
"eval_accuracy": 0.78,
|
49 |
+
"eval_loss": 1.0651038885116577,
|
50 |
+
"eval_runtime": 20.8448,
|
51 |
+
"eval_samples_per_second": 50.372,
|
52 |
+
"eval_steps_per_second": 6.333,
|
53 |
"step": 100
|
54 |
},
|
55 |
{
|
56 |
"epoch": 0.38,
|
57 |
+
"grad_norm": 2.705270767211914,
|
58 |
+
"learning_rate": 6.178343949044585e-05,
|
59 |
+
"loss": 1.0757,
|
60 |
"step": 120
|
61 |
},
|
62 |
{
|
63 |
"epoch": 0.45,
|
64 |
+
"grad_norm": 2.980999231338501,
|
65 |
+
"learning_rate": 5.5414012738853505e-05,
|
66 |
+
"loss": 1.0115,
|
67 |
"step": 140
|
68 |
},
|
69 |
{
|
70 |
"epoch": 0.51,
|
71 |
+
"grad_norm": 4.483489990234375,
|
72 |
+
"learning_rate": 4.904458598726115e-05,
|
73 |
+
"loss": 0.9806,
|
74 |
"step": 160
|
75 |
},
|
76 |
{
|
77 |
"epoch": 0.57,
|
78 |
+
"grad_norm": 3.3504586219787598,
|
79 |
+
"learning_rate": 4.267515923566879e-05,
|
80 |
+
"loss": 0.8805,
|
81 |
"step": 180
|
82 |
},
|
83 |
{
|
84 |
"epoch": 0.64,
|
85 |
+
"grad_norm": 3.248554229736328,
|
86 |
+
"learning_rate": 3.630573248407643e-05,
|
87 |
+
"loss": 0.8527,
|
88 |
"step": 200
|
89 |
},
|
90 |
{
|
91 |
"epoch": 0.64,
|
92 |
+
"eval_accuracy": 0.8133333333333334,
|
93 |
+
"eval_loss": 0.7873561978340149,
|
94 |
+
"eval_runtime": 14.6757,
|
95 |
+
"eval_samples_per_second": 71.547,
|
96 |
+
"eval_steps_per_second": 8.994,
|
97 |
"step": 200
|
98 |
},
|
99 |
{
|
100 |
"epoch": 0.7,
|
101 |
+
"grad_norm": 2.6971187591552734,
|
102 |
+
"learning_rate": 2.9936305732484078e-05,
|
103 |
+
"loss": 0.7773,
|
104 |
"step": 220
|
105 |
},
|
106 |
{
|
107 |
"epoch": 0.76,
|
108 |
+
"grad_norm": 2.3084142208099365,
|
109 |
+
"learning_rate": 2.356687898089172e-05,
|
110 |
+
"loss": 0.6965,
|
111 |
"step": 240
|
112 |
},
|
113 |
{
|
114 |
"epoch": 0.83,
|
115 |
+
"grad_norm": 3.3731772899627686,
|
116 |
+
"learning_rate": 1.7197452229299362e-05,
|
117 |
+
"loss": 0.77,
|
118 |
"step": 260
|
119 |
},
|
120 |
{
|
121 |
"epoch": 0.89,
|
122 |
+
"grad_norm": 1.8178930282592773,
|
123 |
+
"learning_rate": 1.0828025477707008e-05,
|
124 |
+
"loss": 0.7106,
|
125 |
"step": 280
|
126 |
},
|
127 |
{
|
128 |
"epoch": 0.96,
|
129 |
+
"grad_norm": 2.819934368133545,
|
130 |
+
"learning_rate": 4.45859872611465e-06,
|
131 |
+
"loss": 0.7114,
|
132 |
"step": 300
|
133 |
},
|
134 |
{
|
135 |
"epoch": 0.96,
|
136 |
+
"eval_accuracy": 0.8095238095238095,
|
137 |
+
"eval_loss": 0.747850775718689,
|
138 |
+
"eval_runtime": 14.738,
|
139 |
+
"eval_samples_per_second": 71.244,
|
140 |
+
"eval_steps_per_second": 8.956,
|
141 |
"step": 300
|
142 |
},
|
143 |
{
|
144 |
+
"epoch": 1.0,
|
145 |
+
"step": 314,
|
146 |
+
"total_flos": 7.776878731479245e+17,
|
147 |
+
"train_loss": 1.0910153146002703,
|
148 |
+
"train_runtime": 357.1216,
|
149 |
+
"train_samples_per_second": 28.1,
|
150 |
+
"train_steps_per_second": 0.879
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
151 |
}
|
152 |
],
|
153 |
"logging_steps": 20,
|
154 |
+
"max_steps": 314,
|
155 |
"num_input_tokens_seen": 0,
|
156 |
+
"num_train_epochs": 1,
|
157 |
"save_steps": 100,
|
158 |
+
"total_flos": 7.776878731479245e+17,
|
159 |
"train_batch_size": 32,
|
160 |
"trial_name": null,
|
161 |
"trial_params": null
|