SivilTaram
commited on
Upload checkpoint-500/trainer_state.json with huggingface_hub
Browse files
checkpoint-500/trainer_state.json
ADDED
@@ -0,0 +1,399 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.921234454168586,
|
5 |
+
"eval_steps": 200,
|
6 |
+
"global_step": 500,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.018424689083371717,
|
13 |
+
"grad_norm": 12.213397932615418,
|
14 |
+
"learning_rate": 1.2269938650306748e-07,
|
15 |
+
"loss": 0.7896,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.036849378166743434,
|
20 |
+
"grad_norm": 12.364034738650293,
|
21 |
+
"learning_rate": 2.4539877300613496e-07,
|
22 |
+
"loss": 0.7591,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.055274067250115154,
|
27 |
+
"grad_norm": 3.33541668449751,
|
28 |
+
"learning_rate": 3.6809815950920245e-07,
|
29 |
+
"loss": 0.68,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.07369875633348687,
|
34 |
+
"grad_norm": 1.8325651149958935,
|
35 |
+
"learning_rate": 4.907975460122699e-07,
|
36 |
+
"loss": 0.6151,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.09212344541685859,
|
41 |
+
"grad_norm": 1.1850978860914436,
|
42 |
+
"learning_rate": 6.134969325153374e-07,
|
43 |
+
"loss": 0.5901,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.11054813450023031,
|
48 |
+
"grad_norm": 1.0238439765458438,
|
49 |
+
"learning_rate": 7.361963190184049e-07,
|
50 |
+
"loss": 0.5866,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.12897282358360201,
|
55 |
+
"grad_norm": 1.0859904158012907,
|
56 |
+
"learning_rate": 8.588957055214724e-07,
|
57 |
+
"loss": 0.5872,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.14739751266697373,
|
62 |
+
"grad_norm": 0.9445030962726604,
|
63 |
+
"learning_rate": 9.815950920245399e-07,
|
64 |
+
"loss": 0.5908,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.16582220175034545,
|
69 |
+
"grad_norm": 0.902903286685186,
|
70 |
+
"learning_rate": 1.1042944785276073e-06,
|
71 |
+
"loss": 0.585,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.18424689083371718,
|
76 |
+
"grad_norm": 0.9406769634542373,
|
77 |
+
"learning_rate": 1.2269938650306748e-06,
|
78 |
+
"loss": 0.5759,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.2026715799170889,
|
83 |
+
"grad_norm": 0.9379530599920557,
|
84 |
+
"learning_rate": 1.3496932515337423e-06,
|
85 |
+
"loss": 0.5782,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.22109626900046062,
|
90 |
+
"grad_norm": 0.9806028234388734,
|
91 |
+
"learning_rate": 1.4723926380368098e-06,
|
92 |
+
"loss": 0.5599,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.23952095808383234,
|
97 |
+
"grad_norm": 1.0489169076559612,
|
98 |
+
"learning_rate": 1.595092024539877e-06,
|
99 |
+
"loss": 0.5482,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.25794564716720403,
|
104 |
+
"grad_norm": 1.1182127736456162,
|
105 |
+
"learning_rate": 1.7177914110429447e-06,
|
106 |
+
"loss": 0.548,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.2763703362505758,
|
111 |
+
"grad_norm": 1.0633010115096506,
|
112 |
+
"learning_rate": 1.8404907975460122e-06,
|
113 |
+
"loss": 0.5378,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.29479502533394747,
|
118 |
+
"grad_norm": 0.9818550023254589,
|
119 |
+
"learning_rate": 1.9631901840490797e-06,
|
120 |
+
"loss": 0.5336,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.3132197144173192,
|
125 |
+
"grad_norm": 1.1126051299502253,
|
126 |
+
"learning_rate": 1.9998870284726965e-06,
|
127 |
+
"loss": 0.5265,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.3316444035006909,
|
132 |
+
"grad_norm": 1.2009782036779326,
|
133 |
+
"learning_rate": 1.9993337599895922e-06,
|
134 |
+
"loss": 0.5248,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.35006909258406266,
|
139 |
+
"grad_norm": 1.0946468762708952,
|
140 |
+
"learning_rate": 1.9983196994710604e-06,
|
141 |
+
"loss": 0.5196,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.36849378166743435,
|
146 |
+
"grad_norm": 1.0764331817764232,
|
147 |
+
"learning_rate": 1.996845314499634e-06,
|
148 |
+
"loss": 0.5176,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.36849378166743435,
|
153 |
+
"eval_loss": 0.48491016030311584,
|
154 |
+
"eval_runtime": 420.0216,
|
155 |
+
"eval_samples_per_second": 18.373,
|
156 |
+
"eval_steps_per_second": 2.298,
|
157 |
+
"step": 200
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"epoch": 0.3869184707508061,
|
161 |
+
"grad_norm": 1.1132997861385496,
|
162 |
+
"learning_rate": 1.9949112849131003e-06,
|
163 |
+
"loss": 0.5247,
|
164 |
+
"step": 210
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.4053431598341778,
|
168 |
+
"grad_norm": 1.2031973800065288,
|
169 |
+
"learning_rate": 1.9925185024910275e-06,
|
170 |
+
"loss": 0.4998,
|
171 |
+
"step": 220
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 0.42376784891754954,
|
175 |
+
"grad_norm": 1.3213263203662036,
|
176 |
+
"learning_rate": 1.9896680705435688e-06,
|
177 |
+
"loss": 0.4837,
|
178 |
+
"step": 230
|
179 |
+
},
|
180 |
+
{
|
181 |
+
"epoch": 0.44219253800092123,
|
182 |
+
"grad_norm": 1.2768991437360866,
|
183 |
+
"learning_rate": 1.9863613034027223e-06,
|
184 |
+
"loss": 0.5022,
|
185 |
+
"step": 240
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 0.460617227084293,
|
189 |
+
"grad_norm": 1.125849818584755,
|
190 |
+
"learning_rate": 1.9825997258162984e-06,
|
191 |
+
"loss": 0.4894,
|
192 |
+
"step": 250
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 0.47904191616766467,
|
196 |
+
"grad_norm": 1.2472284470521988,
|
197 |
+
"learning_rate": 1.9783850722448565e-06,
|
198 |
+
"loss": 0.4858,
|
199 |
+
"step": 260
|
200 |
+
},
|
201 |
+
{
|
202 |
+
"epoch": 0.49746660525103636,
|
203 |
+
"grad_norm": 1.2644974012951764,
|
204 |
+
"learning_rate": 1.9737192860619475e-06,
|
205 |
+
"loss": 0.4777,
|
206 |
+
"step": 270
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.5158912943344081,
|
210 |
+
"grad_norm": 1.3849217540568468,
|
211 |
+
"learning_rate": 1.9686045186580258e-06,
|
212 |
+
"loss": 0.4649,
|
213 |
+
"step": 280
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 0.5343159834177799,
|
217 |
+
"grad_norm": 1.3250529799695097,
|
218 |
+
"learning_rate": 1.9630431284484445e-06,
|
219 |
+
"loss": 0.4611,
|
220 |
+
"step": 290
|
221 |
+
},
|
222 |
+
{
|
223 |
+
"epoch": 0.5527406725011516,
|
224 |
+
"grad_norm": 1.2452480422858616,
|
225 |
+
"learning_rate": 1.9570376797859937e-06,
|
226 |
+
"loss": 0.4656,
|
227 |
+
"step": 300
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 0.5711653615845232,
|
231 |
+
"grad_norm": 1.3984211841107805,
|
232 |
+
"learning_rate": 1.9505909417784754e-06,
|
233 |
+
"loss": 0.4431,
|
234 |
+
"step": 310
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 0.5895900506678949,
|
238 |
+
"grad_norm": 1.223702182223073,
|
239 |
+
"learning_rate": 1.943705887011874e-06,
|
240 |
+
"loss": 0.4554,
|
241 |
+
"step": 320
|
242 |
+
},
|
243 |
+
{
|
244 |
+
"epoch": 0.6080147397512667,
|
245 |
+
"grad_norm": 1.3622081330675895,
|
246 |
+
"learning_rate": 1.936385690179698e-06,
|
247 |
+
"loss": 0.4411,
|
248 |
+
"step": 330
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.6264394288346384,
|
252 |
+
"grad_norm": 1.519239936734184,
|
253 |
+
"learning_rate": 1.928633726619129e-06,
|
254 |
+
"loss": 0.4272,
|
255 |
+
"step": 340
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 0.6448641179180101,
|
259 |
+
"grad_norm": 1.567409748872381,
|
260 |
+
"learning_rate": 1.92045357075466e-06,
|
261 |
+
"loss": 0.4171,
|
262 |
+
"step": 350
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 0.6632888070013818,
|
266 |
+
"grad_norm": 1.4683737283437361,
|
267 |
+
"learning_rate": 1.9118489944499284e-06,
|
268 |
+
"loss": 0.4282,
|
269 |
+
"step": 360
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.6817134960847536,
|
273 |
+
"grad_norm": 1.3169044491538355,
|
274 |
+
"learning_rate": 1.9028239652685126e-06,
|
275 |
+
"loss": 0.4202,
|
276 |
+
"step": 370
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 0.7001381851681253,
|
280 |
+
"grad_norm": 1.4350164646007288,
|
281 |
+
"learning_rate": 1.893382644644493e-06,
|
282 |
+
"loss": 0.4159,
|
283 |
+
"step": 380
|
284 |
+
},
|
285 |
+
{
|
286 |
+
"epoch": 0.718562874251497,
|
287 |
+
"grad_norm": 1.5280431086824893,
|
288 |
+
"learning_rate": 1.8835293859636177e-06,
|
289 |
+
"loss": 0.3997,
|
290 |
+
"step": 390
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 0.7369875633348687,
|
294 |
+
"grad_norm": 1.6589930030103581,
|
295 |
+
"learning_rate": 1.8732687325559566e-06,
|
296 |
+
"loss": 0.4079,
|
297 |
+
"step": 400
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 0.7369875633348687,
|
301 |
+
"eval_loss": 0.3841191828250885,
|
302 |
+
"eval_runtime": 422.8229,
|
303 |
+
"eval_samples_per_second": 18.251,
|
304 |
+
"eval_steps_per_second": 2.282,
|
305 |
+
"step": 400
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 0.7554122524182404,
|
309 |
+
"grad_norm": 1.6654473824863523,
|
310 |
+
"learning_rate": 1.8626054156009805e-06,
|
311 |
+
"loss": 0.404,
|
312 |
+
"step": 410
|
313 |
+
},
|
314 |
+
{
|
315 |
+
"epoch": 0.7738369415016122,
|
316 |
+
"grad_norm": 2.3979929087591527,
|
317 |
+
"learning_rate": 1.8515443519460136e-06,
|
318 |
+
"loss": 0.4003,
|
319 |
+
"step": 420
|
320 |
+
},
|
321 |
+
{
|
322 |
+
"epoch": 0.7922616305849839,
|
323 |
+
"grad_norm": 1.3611905196878527,
|
324 |
+
"learning_rate": 1.8400906418390804e-06,
|
325 |
+
"loss": 0.3799,
|
326 |
+
"step": 430
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 0.8106863196683556,
|
330 |
+
"grad_norm": 1.4783642041172977,
|
331 |
+
"learning_rate": 1.8282495665771862e-06,
|
332 |
+
"loss": 0.387,
|
333 |
+
"step": 440
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 0.8291110087517273,
|
337 |
+
"grad_norm": 1.3634768413482825,
|
338 |
+
"learning_rate": 1.8160265860711132e-06,
|
339 |
+
"loss": 0.3824,
|
340 |
+
"step": 450
|
341 |
+
},
|
342 |
+
{
|
343 |
+
"epoch": 0.8475356978350991,
|
344 |
+
"grad_norm": 1.6646421151787196,
|
345 |
+
"learning_rate": 1.803427336327861e-06,
|
346 |
+
"loss": 0.3875,
|
347 |
+
"step": 460
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 0.8659603869184708,
|
351 |
+
"grad_norm": 1.5372444888922498,
|
352 |
+
"learning_rate": 1.7904576268518884e-06,
|
353 |
+
"loss": 0.3906,
|
354 |
+
"step": 470
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 0.8843850760018425,
|
358 |
+
"grad_norm": 1.6088504472102723,
|
359 |
+
"learning_rate": 1.7771234379663542e-06,
|
360 |
+
"loss": 0.3712,
|
361 |
+
"step": 480
|
362 |
+
},
|
363 |
+
{
|
364 |
+
"epoch": 0.9028097650852142,
|
365 |
+
"grad_norm": 1.764162466911846,
|
366 |
+
"learning_rate": 1.763430918055595e-06,
|
367 |
+
"loss": 0.3606,
|
368 |
+
"step": 490
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 0.921234454168586,
|
372 |
+
"grad_norm": 1.71583814988475,
|
373 |
+
"learning_rate": 1.7493863807301113e-06,
|
374 |
+
"loss": 0.3596,
|
375 |
+
"step": 500
|
376 |
+
}
|
377 |
+
],
|
378 |
+
"logging_steps": 10,
|
379 |
+
"max_steps": 1626,
|
380 |
+
"num_input_tokens_seen": 0,
|
381 |
+
"num_train_epochs": 3,
|
382 |
+
"save_steps": 500,
|
383 |
+
"stateful_callbacks": {
|
384 |
+
"TrainerControl": {
|
385 |
+
"args": {
|
386 |
+
"should_epoch_stop": false,
|
387 |
+
"should_evaluate": false,
|
388 |
+
"should_log": false,
|
389 |
+
"should_save": true,
|
390 |
+
"should_training_stop": false
|
391 |
+
},
|
392 |
+
"attributes": {}
|
393 |
+
}
|
394 |
+
},
|
395 |
+
"total_flos": 418549931704320.0,
|
396 |
+
"train_batch_size": 4,
|
397 |
+
"trial_name": null,
|
398 |
+
"trial_params": null
|
399 |
+
}
|