All Dunn!!!
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +473 -0
all_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"total_flos": 1.9906356553640313e+19,
|
4 |
+
"train_loss": 0.47363961749354666,
|
5 |
+
"train_runtime": 3260.3696,
|
6 |
+
"train_samples_per_second": 4.125,
|
7 |
+
"train_steps_per_second": 0.516
|
8 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"total_flos": 1.9906356553640313e+19,
|
4 |
+
"train_loss": 0.47363961749354666,
|
5 |
+
"train_runtime": 3260.3696,
|
6 |
+
"train_samples_per_second": 4.125,
|
7 |
+
"train_steps_per_second": 0.516
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,473 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1682,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0011890606420927466,
|
13 |
+
"grad_norm": 115.31934356689453,
|
14 |
+
"learning_rate": 4.9970273483947685e-05,
|
15 |
+
"loss": 9.3546,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.034482758620689655,
|
20 |
+
"grad_norm": 18.599306106567383,
|
21 |
+
"learning_rate": 4.913793103448276e-05,
|
22 |
+
"loss": 2.4091,
|
23 |
+
"step": 29
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.06896551724137931,
|
27 |
+
"grad_norm": 12.393567085266113,
|
28 |
+
"learning_rate": 4.827586206896552e-05,
|
29 |
+
"loss": 1.1241,
|
30 |
+
"step": 58
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.10344827586206896,
|
34 |
+
"grad_norm": 50.094051361083984,
|
35 |
+
"learning_rate": 4.741379310344828e-05,
|
36 |
+
"loss": 1.0125,
|
37 |
+
"step": 87
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.13793103448275862,
|
41 |
+
"grad_norm": 39.95038986206055,
|
42 |
+
"learning_rate": 4.655172413793104e-05,
|
43 |
+
"loss": 0.9273,
|
44 |
+
"step": 116
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.1724137931034483,
|
48 |
+
"grad_norm": 13.281441688537598,
|
49 |
+
"learning_rate": 4.5689655172413794e-05,
|
50 |
+
"loss": 0.8734,
|
51 |
+
"step": 145
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.20689655172413793,
|
55 |
+
"grad_norm": 29.820255279541016,
|
56 |
+
"learning_rate": 4.482758620689655e-05,
|
57 |
+
"loss": 0.9182,
|
58 |
+
"step": 174
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.2413793103448276,
|
62 |
+
"grad_norm": 14.038665771484375,
|
63 |
+
"learning_rate": 4.396551724137931e-05,
|
64 |
+
"loss": 0.9129,
|
65 |
+
"step": 203
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.27586206896551724,
|
69 |
+
"grad_norm": 13.406801223754883,
|
70 |
+
"learning_rate": 4.3103448275862066e-05,
|
71 |
+
"loss": 0.7973,
|
72 |
+
"step": 232
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.3103448275862069,
|
76 |
+
"grad_norm": 15.321793556213379,
|
77 |
+
"learning_rate": 4.224137931034483e-05,
|
78 |
+
"loss": 0.8415,
|
79 |
+
"step": 261
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.3448275862068966,
|
83 |
+
"grad_norm": 27.121288299560547,
|
84 |
+
"learning_rate": 4.1379310344827587e-05,
|
85 |
+
"loss": 0.7995,
|
86 |
+
"step": 290
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.3793103448275862,
|
90 |
+
"grad_norm": 36.715858459472656,
|
91 |
+
"learning_rate": 4.0517241379310344e-05,
|
92 |
+
"loss": 0.6301,
|
93 |
+
"step": 319
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.41379310344827586,
|
97 |
+
"grad_norm": 16.650150299072266,
|
98 |
+
"learning_rate": 3.965517241379311e-05,
|
99 |
+
"loss": 0.5675,
|
100 |
+
"step": 348
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.4482758620689655,
|
104 |
+
"grad_norm": 230.8839569091797,
|
105 |
+
"learning_rate": 3.8793103448275865e-05,
|
106 |
+
"loss": 0.6154,
|
107 |
+
"step": 377
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.4827586206896552,
|
111 |
+
"grad_norm": 6.414034366607666,
|
112 |
+
"learning_rate": 3.793103448275862e-05,
|
113 |
+
"loss": 0.6439,
|
114 |
+
"step": 406
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.5172413793103449,
|
118 |
+
"grad_norm": 9.441433906555176,
|
119 |
+
"learning_rate": 3.7068965517241385e-05,
|
120 |
+
"loss": 0.5776,
|
121 |
+
"step": 435
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.5517241379310345,
|
125 |
+
"grad_norm": 23.88970947265625,
|
126 |
+
"learning_rate": 3.620689655172414e-05,
|
127 |
+
"loss": 0.6126,
|
128 |
+
"step": 464
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.5862068965517241,
|
132 |
+
"grad_norm": 10.72873592376709,
|
133 |
+
"learning_rate": 3.53448275862069e-05,
|
134 |
+
"loss": 0.529,
|
135 |
+
"step": 493
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.6206896551724138,
|
139 |
+
"grad_norm": 22.285120010375977,
|
140 |
+
"learning_rate": 3.4482758620689657e-05,
|
141 |
+
"loss": 0.5418,
|
142 |
+
"step": 522
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.6551724137931034,
|
146 |
+
"grad_norm": 28.53759002685547,
|
147 |
+
"learning_rate": 3.3620689655172414e-05,
|
148 |
+
"loss": 0.4864,
|
149 |
+
"step": 551
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.6896551724137931,
|
153 |
+
"grad_norm": 15.215431213378906,
|
154 |
+
"learning_rate": 3.275862068965517e-05,
|
155 |
+
"loss": 0.5474,
|
156 |
+
"step": 580
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.7241379310344828,
|
160 |
+
"grad_norm": 5.726695537567139,
|
161 |
+
"learning_rate": 3.1896551724137935e-05,
|
162 |
+
"loss": 0.4492,
|
163 |
+
"step": 609
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.7586206896551724,
|
167 |
+
"grad_norm": 20.181598663330078,
|
168 |
+
"learning_rate": 3.103448275862069e-05,
|
169 |
+
"loss": 0.4925,
|
170 |
+
"step": 638
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.7931034482758621,
|
174 |
+
"grad_norm": 16.79625701904297,
|
175 |
+
"learning_rate": 3.017241379310345e-05,
|
176 |
+
"loss": 0.4405,
|
177 |
+
"step": 667
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.8275862068965517,
|
181 |
+
"grad_norm": 5.420645236968994,
|
182 |
+
"learning_rate": 2.9310344827586206e-05,
|
183 |
+
"loss": 0.5411,
|
184 |
+
"step": 696
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.8620689655172413,
|
188 |
+
"grad_norm": 6.218174457550049,
|
189 |
+
"learning_rate": 2.844827586206897e-05,
|
190 |
+
"loss": 0.5516,
|
191 |
+
"step": 725
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.896551724137931,
|
195 |
+
"grad_norm": 27.71928596496582,
|
196 |
+
"learning_rate": 2.7586206896551727e-05,
|
197 |
+
"loss": 0.4341,
|
198 |
+
"step": 754
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.9310344827586207,
|
202 |
+
"grad_norm": 31.43971824645996,
|
203 |
+
"learning_rate": 2.672413793103448e-05,
|
204 |
+
"loss": 0.3883,
|
205 |
+
"step": 783
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.9655172413793104,
|
209 |
+
"grad_norm": 7.160225868225098,
|
210 |
+
"learning_rate": 2.5862068965517244e-05,
|
211 |
+
"loss": 0.4214,
|
212 |
+
"step": 812
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 1.0,
|
216 |
+
"grad_norm": 7.139071464538574,
|
217 |
+
"learning_rate": 2.5e-05,
|
218 |
+
"loss": 0.3728,
|
219 |
+
"step": 841
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 1.0,
|
223 |
+
"eval_cer": 0.019845431962462048,
|
224 |
+
"eval_loss": 0.2980102300643921,
|
225 |
+
"eval_runtime": 887.0299,
|
226 |
+
"eval_samples_per_second": 1.896,
|
227 |
+
"eval_steps_per_second": 0.238,
|
228 |
+
"step": 841
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 1.0344827586206897,
|
232 |
+
"grad_norm": 14.740863800048828,
|
233 |
+
"learning_rate": 2.413793103448276e-05,
|
234 |
+
"loss": 0.3343,
|
235 |
+
"step": 870
|
236 |
+
},
|
237 |
+
{
|
238 |
+
"epoch": 1.0689655172413792,
|
239 |
+
"grad_norm": 22.295618057250977,
|
240 |
+
"learning_rate": 2.327586206896552e-05,
|
241 |
+
"loss": 0.3419,
|
242 |
+
"step": 899
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 1.103448275862069,
|
246 |
+
"grad_norm": 2.4551053047180176,
|
247 |
+
"learning_rate": 2.2413793103448276e-05,
|
248 |
+
"loss": 0.3123,
|
249 |
+
"step": 928
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 1.1379310344827587,
|
253 |
+
"grad_norm": 7.861540794372559,
|
254 |
+
"learning_rate": 2.1551724137931033e-05,
|
255 |
+
"loss": 0.349,
|
256 |
+
"step": 957
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 1.1724137931034484,
|
260 |
+
"grad_norm": 7.037637233734131,
|
261 |
+
"learning_rate": 2.0689655172413793e-05,
|
262 |
+
"loss": 0.2869,
|
263 |
+
"step": 986
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 1.206896551724138,
|
267 |
+
"grad_norm": 11.390913009643555,
|
268 |
+
"learning_rate": 1.9827586206896554e-05,
|
269 |
+
"loss": 0.2826,
|
270 |
+
"step": 1015
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 1.2413793103448276,
|
274 |
+
"grad_norm": 2.8627371788024902,
|
275 |
+
"learning_rate": 1.896551724137931e-05,
|
276 |
+
"loss": 0.2573,
|
277 |
+
"step": 1044
|
278 |
+
},
|
279 |
+
{
|
280 |
+
"epoch": 1.2758620689655173,
|
281 |
+
"grad_norm": 16.846853256225586,
|
282 |
+
"learning_rate": 1.810344827586207e-05,
|
283 |
+
"loss": 0.2752,
|
284 |
+
"step": 1073
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 1.3103448275862069,
|
288 |
+
"grad_norm": 3.9017491340637207,
|
289 |
+
"learning_rate": 1.7241379310344828e-05,
|
290 |
+
"loss": 0.2555,
|
291 |
+
"step": 1102
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 1.3448275862068966,
|
295 |
+
"grad_norm": 5.074669361114502,
|
296 |
+
"learning_rate": 1.6379310344827585e-05,
|
297 |
+
"loss": 0.2452,
|
298 |
+
"step": 1131
|
299 |
+
},
|
300 |
+
{
|
301 |
+
"epoch": 1.3793103448275863,
|
302 |
+
"grad_norm": 18.69165802001953,
|
303 |
+
"learning_rate": 1.5517241379310346e-05,
|
304 |
+
"loss": 0.2666,
|
305 |
+
"step": 1160
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 1.4137931034482758,
|
309 |
+
"grad_norm": 5.539306640625,
|
310 |
+
"learning_rate": 1.4655172413793103e-05,
|
311 |
+
"loss": 0.2741,
|
312 |
+
"step": 1189
|
313 |
+
},
|
314 |
+
{
|
315 |
+
"epoch": 1.4482758620689655,
|
316 |
+
"grad_norm": 5.758066654205322,
|
317 |
+
"learning_rate": 1.3793103448275863e-05,
|
318 |
+
"loss": 0.2365,
|
319 |
+
"step": 1218
|
320 |
+
},
|
321 |
+
{
|
322 |
+
"epoch": 1.4827586206896552,
|
323 |
+
"grad_norm": 4.248748779296875,
|
324 |
+
"learning_rate": 1.2931034482758622e-05,
|
325 |
+
"loss": 0.2521,
|
326 |
+
"step": 1247
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 1.5172413793103448,
|
330 |
+
"grad_norm": 11.129036903381348,
|
331 |
+
"learning_rate": 1.206896551724138e-05,
|
332 |
+
"loss": 0.2257,
|
333 |
+
"step": 1276
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 1.5517241379310345,
|
337 |
+
"grad_norm": 2.8656809329986572,
|
338 |
+
"learning_rate": 1.1206896551724138e-05,
|
339 |
+
"loss": 0.2106,
|
340 |
+
"step": 1305
|
341 |
+
},
|
342 |
+
{
|
343 |
+
"epoch": 1.5862068965517242,
|
344 |
+
"grad_norm": 2.041245460510254,
|
345 |
+
"learning_rate": 1.0344827586206897e-05,
|
346 |
+
"loss": 0.227,
|
347 |
+
"step": 1334
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 1.6206896551724137,
|
351 |
+
"grad_norm": 1.451377034187317,
|
352 |
+
"learning_rate": 9.482758620689655e-06,
|
353 |
+
"loss": 0.1961,
|
354 |
+
"step": 1363
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 1.6551724137931034,
|
358 |
+
"grad_norm": 4.332240104675293,
|
359 |
+
"learning_rate": 8.620689655172414e-06,
|
360 |
+
"loss": 0.195,
|
361 |
+
"step": 1392
|
362 |
+
},
|
363 |
+
{
|
364 |
+
"epoch": 1.6896551724137931,
|
365 |
+
"grad_norm": 5.428757667541504,
|
366 |
+
"learning_rate": 7.758620689655173e-06,
|
367 |
+
"loss": 0.1929,
|
368 |
+
"step": 1421
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 1.7241379310344827,
|
372 |
+
"grad_norm": 2.5995798110961914,
|
373 |
+
"learning_rate": 6.896551724137932e-06,
|
374 |
+
"loss": 0.1991,
|
375 |
+
"step": 1450
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 1.7586206896551724,
|
379 |
+
"grad_norm": 2.2909860610961914,
|
380 |
+
"learning_rate": 6.03448275862069e-06,
|
381 |
+
"loss": 0.168,
|
382 |
+
"step": 1479
|
383 |
+
},
|
384 |
+
{
|
385 |
+
"epoch": 1.793103448275862,
|
386 |
+
"grad_norm": 2.049956798553467,
|
387 |
+
"learning_rate": 5.172413793103448e-06,
|
388 |
+
"loss": 0.18,
|
389 |
+
"step": 1508
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 1.8275862068965516,
|
393 |
+
"grad_norm": 2.291414737701416,
|
394 |
+
"learning_rate": 4.310344827586207e-06,
|
395 |
+
"loss": 0.1642,
|
396 |
+
"step": 1537
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 1.8620689655172413,
|
400 |
+
"grad_norm": 6.125259876251221,
|
401 |
+
"learning_rate": 3.448275862068966e-06,
|
402 |
+
"loss": 0.197,
|
403 |
+
"step": 1566
|
404 |
+
},
|
405 |
+
{
|
406 |
+
"epoch": 1.896551724137931,
|
407 |
+
"grad_norm": 2.359379291534424,
|
408 |
+
"learning_rate": 2.586206896551724e-06,
|
409 |
+
"loss": 0.1615,
|
410 |
+
"step": 1595
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 1.9310344827586206,
|
414 |
+
"grad_norm": 2.1470141410827637,
|
415 |
+
"learning_rate": 1.724137931034483e-06,
|
416 |
+
"loss": 0.1669,
|
417 |
+
"step": 1624
|
418 |
+
},
|
419 |
+
{
|
420 |
+
"epoch": 1.9655172413793105,
|
421 |
+
"grad_norm": 13.040664672851562,
|
422 |
+
"learning_rate": 8.620689655172415e-07,
|
423 |
+
"loss": 0.1641,
|
424 |
+
"step": 1653
|
425 |
+
},
|
426 |
+
{
|
427 |
+
"epoch": 2.0,
|
428 |
+
"grad_norm": 1.5605217218399048,
|
429 |
+
"learning_rate": 0.0,
|
430 |
+
"loss": 0.1551,
|
431 |
+
"step": 1682
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 2.0,
|
435 |
+
"eval_cer": 0.003201766491857577,
|
436 |
+
"eval_loss": 0.17416498064994812,
|
437 |
+
"eval_runtime": 882.6649,
|
438 |
+
"eval_samples_per_second": 1.906,
|
439 |
+
"eval_steps_per_second": 0.239,
|
440 |
+
"step": 1682
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 2.0,
|
444 |
+
"step": 1682,
|
445 |
+
"total_flos": 1.9906356553640313e+19,
|
446 |
+
"train_loss": 0.47363961749354666,
|
447 |
+
"train_runtime": 3260.3696,
|
448 |
+
"train_samples_per_second": 4.125,
|
449 |
+
"train_steps_per_second": 0.516
|
450 |
+
}
|
451 |
+
],
|
452 |
+
"logging_steps": 29,
|
453 |
+
"max_steps": 1682,
|
454 |
+
"num_input_tokens_seen": 0,
|
455 |
+
"num_train_epochs": 2,
|
456 |
+
"save_steps": 500,
|
457 |
+
"stateful_callbacks": {
|
458 |
+
"TrainerControl": {
|
459 |
+
"args": {
|
460 |
+
"should_epoch_stop": false,
|
461 |
+
"should_evaluate": false,
|
462 |
+
"should_log": false,
|
463 |
+
"should_save": true,
|
464 |
+
"should_training_stop": true
|
465 |
+
},
|
466 |
+
"attributes": {}
|
467 |
+
}
|
468 |
+
},
|
469 |
+
"total_flos": 1.9906356553640313e+19,
|
470 |
+
"train_batch_size": 8,
|
471 |
+
"trial_name": null,
|
472 |
+
"trial_params": null
|
473 |
+
}
|