Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +456 -0
trainer_state.json
ADDED
@@ -0,0 +1,456 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 50,
|
6 |
+
"global_step": 320,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.03125,
|
13 |
+
"grad_norm": 159.0,
|
14 |
+
"learning_rate": 1.8750000000000002e-05,
|
15 |
+
"loss": 1.1247,
|
16 |
+
"num_input_tokens_seen": 1465248,
|
17 |
+
"step": 10
|
18 |
+
},
|
19 |
+
{
|
20 |
+
"epoch": 0.0625,
|
21 |
+
"grad_norm": 65.0,
|
22 |
+
"learning_rate": 2.9987186375809513e-05,
|
23 |
+
"loss": 0.638,
|
24 |
+
"num_input_tokens_seen": 2893200,
|
25 |
+
"step": 20
|
26 |
+
},
|
27 |
+
{
|
28 |
+
"epoch": 0.09375,
|
29 |
+
"grad_norm": 10.0625,
|
30 |
+
"learning_rate": 2.984328439990804e-05,
|
31 |
+
"loss": 0.2887,
|
32 |
+
"num_input_tokens_seen": 4373616,
|
33 |
+
"step": 30
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.125,
|
37 |
+
"grad_norm": 14.875,
|
38 |
+
"learning_rate": 2.9541003989089956e-05,
|
39 |
+
"loss": 0.2392,
|
40 |
+
"num_input_tokens_seen": 5755248,
|
41 |
+
"step": 40
|
42 |
+
},
|
43 |
+
{
|
44 |
+
"epoch": 0.15625,
|
45 |
+
"grad_norm": 9.8125,
|
46 |
+
"learning_rate": 2.9083570487361445e-05,
|
47 |
+
"loss": 0.226,
|
48 |
+
"num_input_tokens_seen": 7213808,
|
49 |
+
"step": 50
|
50 |
+
},
|
51 |
+
{
|
52 |
+
"epoch": 0.1875,
|
53 |
+
"grad_norm": 19.0,
|
54 |
+
"learning_rate": 2.8475864728379682e-05,
|
55 |
+
"loss": 0.2489,
|
56 |
+
"num_input_tokens_seen": 8649584,
|
57 |
+
"step": 60
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.21875,
|
61 |
+
"grad_norm": 27.75,
|
62 |
+
"learning_rate": 2.7724370956763605e-05,
|
63 |
+
"loss": 0.2462,
|
64 |
+
"num_input_tokens_seen": 10054416,
|
65 |
+
"step": 70
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.25,
|
69 |
+
"grad_norm": 12.5,
|
70 |
+
"learning_rate": 2.6837107640945904e-05,
|
71 |
+
"loss": 0.2133,
|
72 |
+
"num_input_tokens_seen": 11521216,
|
73 |
+
"step": 80
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 0.28125,
|
77 |
+
"grad_norm": 8.625,
|
78 |
+
"learning_rate": 2.5823541915795932e-05,
|
79 |
+
"loss": 0.2035,
|
80 |
+
"num_input_tokens_seen": 12912576,
|
81 |
+
"step": 90
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.3125,
|
85 |
+
"grad_norm": 4.25,
|
86 |
+
"learning_rate": 2.469448856791411e-05,
|
87 |
+
"loss": 0.2226,
|
88 |
+
"num_input_tokens_seen": 14369728,
|
89 |
+
"step": 100
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.34375,
|
93 |
+
"grad_norm": 22.625,
|
94 |
+
"learning_rate": 2.3461994641428768e-05,
|
95 |
+
"loss": 0.2314,
|
96 |
+
"num_input_tokens_seen": 15796256,
|
97 |
+
"step": 110
|
98 |
+
},
|
99 |
+
{
|
100 |
+
"epoch": 0.375,
|
101 |
+
"grad_norm": 19.875,
|
102 |
+
"learning_rate": 2.2139210895556104e-05,
|
103 |
+
"loss": 0.2564,
|
104 |
+
"num_input_tokens_seen": 17240400,
|
105 |
+
"step": 120
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.40625,
|
109 |
+
"grad_norm": 7.78125,
|
110 |
+
"learning_rate": 2.074025148547635e-05,
|
111 |
+
"loss": 0.2442,
|
112 |
+
"num_input_tokens_seen": 18663216,
|
113 |
+
"step": 130
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 0.4375,
|
117 |
+
"grad_norm": 12.125,
|
118 |
+
"learning_rate": 1.928004336373658e-05,
|
119 |
+
"loss": 0.2115,
|
120 |
+
"num_input_tokens_seen": 20069056,
|
121 |
+
"step": 140
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.46875,
|
125 |
+
"grad_norm": 21.0,
|
126 |
+
"learning_rate": 1.777416700907338e-05,
|
127 |
+
"loss": 0.1656,
|
128 |
+
"num_input_tokens_seen": 21471248,
|
129 |
+
"step": 150
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.5,
|
133 |
+
"grad_norm": 10.6875,
|
134 |
+
"learning_rate": 1.623869018208499e-05,
|
135 |
+
"loss": 0.1906,
|
136 |
+
"num_input_tokens_seen": 22898704,
|
137 |
+
"step": 160
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.53125,
|
141 |
+
"grad_norm": 8.4375,
|
142 |
+
"learning_rate": 1.4689996481586688e-05,
|
143 |
+
"loss": 0.1883,
|
144 |
+
"num_input_tokens_seen": 24365264,
|
145 |
+
"step": 170
|
146 |
+
},
|
147 |
+
{
|
148 |
+
"epoch": 0.5625,
|
149 |
+
"grad_norm": 14.0,
|
150 |
+
"learning_rate": 1.3144610530959784e-05,
|
151 |
+
"loss": 0.2115,
|
152 |
+
"num_input_tokens_seen": 25870176,
|
153 |
+
"step": 180
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 0.59375,
|
157 |
+
"grad_norm": 16.25,
|
158 |
+
"learning_rate": 1.1619021659762912e-05,
|
159 |
+
"loss": 0.1644,
|
160 |
+
"num_input_tokens_seen": 27346880,
|
161 |
+
"step": 190
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 0.625,
|
165 |
+
"grad_norm": 8.375,
|
166 |
+
"learning_rate": 1.0129507961929749e-05,
|
167 |
+
"loss": 0.2046,
|
168 |
+
"num_input_tokens_seen": 28825968,
|
169 |
+
"step": 200
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"epoch": 0.65625,
|
173 |
+
"grad_norm": 13.3125,
|
174 |
+
"learning_rate": 8.691962607859386e-06,
|
175 |
+
"loss": 0.1767,
|
176 |
+
"num_input_tokens_seen": 30278096,
|
177 |
+
"step": 210
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.6875,
|
181 |
+
"grad_norm": 6.96875,
|
182 |
+
"learning_rate": 7.321724263655989e-06,
|
183 |
+
"loss": 0.1739,
|
184 |
+
"num_input_tokens_seen": 31701248,
|
185 |
+
"step": 220
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 0.71875,
|
189 |
+
"grad_norm": 3.765625,
|
190 |
+
"learning_rate": 6.0334134269513865e-06,
|
191 |
+
"loss": 0.1648,
|
192 |
+
"num_input_tokens_seen": 33175280,
|
193 |
+
"step": 230
|
194 |
+
},
|
195 |
+
{
|
196 |
+
"epoch": 0.75,
|
197 |
+
"grad_norm": 37.5,
|
198 |
+
"learning_rate": 4.840776425613887e-06,
|
199 |
+
"loss": 0.1662,
|
200 |
+
"num_input_tokens_seen": 34680976,
|
201 |
+
"step": 240
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 0.78125,
|
205 |
+
"grad_norm": 6.3125,
|
206 |
+
"learning_rate": 3.756538743883111e-06,
|
207 |
+
"loss": 0.1867,
|
208 |
+
"num_input_tokens_seen": 36137360,
|
209 |
+
"step": 250
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 0.8125,
|
213 |
+
"grad_norm": 3.46875,
|
214 |
+
"learning_rate": 2.792269240947076e-06,
|
215 |
+
"loss": 0.1601,
|
216 |
+
"num_input_tokens_seen": 37577552,
|
217 |
+
"step": 260
|
218 |
+
},
|
219 |
+
{
|
220 |
+
"epoch": 0.84375,
|
221 |
+
"grad_norm": 6.0,
|
222 |
+
"learning_rate": 1.958256710754496e-06,
|
223 |
+
"loss": 0.1679,
|
224 |
+
"num_input_tokens_seen": 39032832,
|
225 |
+
"step": 270
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 0.875,
|
229 |
+
"grad_norm": 8.125,
|
230 |
+
"learning_rate": 1.2634001001741375e-06,
|
231 |
+
"loss": 0.1518,
|
232 |
+
"num_input_tokens_seen": 40447376,
|
233 |
+
"step": 280
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.90625,
|
237 |
+
"grad_norm": 5.96875,
|
238 |
+
"learning_rate": 7.151135568777839e-07,
|
239 |
+
"loss": 0.171,
|
240 |
+
"num_input_tokens_seen": 41890144,
|
241 |
+
"step": 290
|
242 |
+
},
|
243 |
+
{
|
244 |
+
"epoch": 0.9375,
|
245 |
+
"grad_norm": 17.5,
|
246 |
+
"learning_rate": 3.192473200896828e-07,
|
247 |
+
"loss": 0.1702,
|
248 |
+
"num_input_tokens_seen": 43291456,
|
249 |
+
"step": 300
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 0.96875,
|
253 |
+
"grad_norm": 20.5,
|
254 |
+
"learning_rate": 8.002529830136163e-08,
|
255 |
+
"loss": 0.1595,
|
256 |
+
"num_input_tokens_seen": 44726672,
|
257 |
+
"step": 310
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 1.0,
|
261 |
+
"grad_norm": 13.3125,
|
262 |
+
"learning_rate": 0.0,
|
263 |
+
"loss": 0.1722,
|
264 |
+
"num_input_tokens_seen": 46209888,
|
265 |
+
"step": 320
|
266 |
+
},
|
267 |
+
{
|
268 |
+
"epoch": 1.0,
|
269 |
+
"eval_cnn_balanced_accuracy": 0.5759533564450047,
|
270 |
+
"eval_cnn_f1_score": 0.2608695652173913,
|
271 |
+
"eval_cnn_loss": 0.31287845969200134,
|
272 |
+
"eval_cnn_pr_auc": 0.42626378805395937,
|
273 |
+
"eval_cnn_roc_auc": 0.7679027909094093,
|
274 |
+
"eval_cnn_runtime": 10.722,
|
275 |
+
"eval_cnn_samples_per_second": 52.043,
|
276 |
+
"eval_cnn_steps_per_second": 0.839,
|
277 |
+
"num_input_tokens_seen": 46209888,
|
278 |
+
"step": 320
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 1.0,
|
282 |
+
"eval_xsum_balanced_accuracy": 0.749277038750723,
|
283 |
+
"eval_xsum_f1_score": 0.7258382642998027,
|
284 |
+
"eval_xsum_loss": 0.5868125557899475,
|
285 |
+
"eval_xsum_pr_auc": 0.7976212402287128,
|
286 |
+
"eval_xsum_roc_auc": 0.8217916586337639,
|
287 |
+
"eval_xsum_runtime": 9.7581,
|
288 |
+
"eval_xsum_samples_per_second": 57.183,
|
289 |
+
"eval_xsum_steps_per_second": 0.922,
|
290 |
+
"num_input_tokens_seen": 46209888,
|
291 |
+
"step": 320
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 1.0,
|
295 |
+
"eval_medias_balanced_accuracy": 0.7074867769288893,
|
296 |
+
"eval_medias_f1_score": 0.5704225352112676,
|
297 |
+
"eval_medias_loss": 0.45626452565193176,
|
298 |
+
"eval_medias_pr_auc": 0.6584352779944264,
|
299 |
+
"eval_medias_roc_auc": 0.7960236336159854,
|
300 |
+
"eval_medias_runtime": 9.9504,
|
301 |
+
"eval_medias_samples_per_second": 72.962,
|
302 |
+
"eval_medias_steps_per_second": 1.206,
|
303 |
+
"num_input_tokens_seen": 46209888,
|
304 |
+
"step": 320
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 1.0,
|
308 |
+
"eval_meetb_balanced_accuracy": 0.7702572347266881,
|
309 |
+
"eval_meetb_f1_score": 0.6498194945848376,
|
310 |
+
"eval_meetb_loss": 0.316259503364563,
|
311 |
+
"eval_meetb_pr_auc": 0.7402648937803408,
|
312 |
+
"eval_meetb_roc_auc": 0.8713504823151126,
|
313 |
+
"eval_meetb_runtime": 10.6662,
|
314 |
+
"eval_meetb_samples_per_second": 72.378,
|
315 |
+
"eval_meetb_steps_per_second": 1.219,
|
316 |
+
"num_input_tokens_seen": 46209888,
|
317 |
+
"step": 320
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 1.0,
|
321 |
+
"eval_wice_balanced_accuracy": 0.8154247364773681,
|
322 |
+
"eval_wice_f1_score": 0.8679245283018868,
|
323 |
+
"eval_wice_loss": 0.4058681130409241,
|
324 |
+
"eval_wice_pr_auc": 0.9493679290266217,
|
325 |
+
"eval_wice_roc_auc": 0.8869679395995185,
|
326 |
+
"eval_wice_runtime": 123.4927,
|
327 |
+
"eval_wice_samples_per_second": 2.899,
|
328 |
+
"eval_wice_steps_per_second": 0.049,
|
329 |
+
"num_input_tokens_seen": 46209888,
|
330 |
+
"step": 320
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 1.0,
|
334 |
+
"eval_reveal_balanced_accuracy": 0.8941412213740458,
|
335 |
+
"eval_reveal_f1_score": 0.9196141479099679,
|
336 |
+
"eval_reveal_loss": 0.2713985741138458,
|
337 |
+
"eval_reveal_pr_auc": 0.9884053490966663,
|
338 |
+
"eval_reveal_roc_auc": 0.960442748091603,
|
339 |
+
"eval_reveal_runtime": 7.8148,
|
340 |
+
"eval_reveal_samples_per_second": 218.815,
|
341 |
+
"eval_reveal_steps_per_second": 3.455,
|
342 |
+
"num_input_tokens_seen": 46209888,
|
343 |
+
"step": 320
|
344 |
+
},
|
345 |
+
{
|
346 |
+
"epoch": 1.0,
|
347 |
+
"eval_claim_verify_balanced_accuracy": 0.7654793545023335,
|
348 |
+
"eval_claim_verify_f1_score": 0.6621160409556314,
|
349 |
+
"eval_claim_verify_loss": 0.47898775339126587,
|
350 |
+
"eval_claim_verify_pr_auc": 0.6755500302254687,
|
351 |
+
"eval_claim_verify_roc_auc": 0.8519738375912951,
|
352 |
+
"eval_claim_verify_runtime": 261.7647,
|
353 |
+
"eval_claim_verify_samples_per_second": 4.156,
|
354 |
+
"eval_claim_verify_steps_per_second": 0.065,
|
355 |
+
"num_input_tokens_seen": 46209888,
|
356 |
+
"step": 320
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 1.0,
|
360 |
+
"eval_fact_check_balanced_accuracy": 0.7718129805113534,
|
361 |
+
"eval_fact_check_f1_score": 0.9008264462809917,
|
362 |
+
"eval_fact_check_loss": 0.39186760783195496,
|
363 |
+
"eval_fact_check_pr_auc": 0.9477958353480682,
|
364 |
+
"eval_fact_check_roc_auc": 0.8752134364384051,
|
365 |
+
"eval_fact_check_runtime": 5.283,
|
366 |
+
"eval_fact_check_samples_per_second": 296.421,
|
367 |
+
"eval_fact_check_steps_per_second": 4.732,
|
368 |
+
"num_input_tokens_seen": 46209888,
|
369 |
+
"step": 320
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 1.0,
|
373 |
+
"eval_expertqa_balanced_accuracy": 0.5946286054753636,
|
374 |
+
"eval_expertqa_f1_score": 0.36883320281910725,
|
375 |
+
"eval_expertqa_loss": 1.1040500402450562,
|
376 |
+
"eval_expertqa_pr_auc": 0.2890662808738864,
|
377 |
+
"eval_expertqa_roc_auc": 0.6258950520788967,
|
378 |
+
"eval_expertqa_runtime": 536.4175,
|
379 |
+
"eval_expertqa_samples_per_second": 6.901,
|
380 |
+
"eval_expertqa_steps_per_second": 0.108,
|
381 |
+
"num_input_tokens_seen": 46209888,
|
382 |
+
"step": 320
|
383 |
+
},
|
384 |
+
{
|
385 |
+
"epoch": 1.0,
|
386 |
+
"eval_lfqa_balanced_accuracy": 0.8775737079235312,
|
387 |
+
"eval_lfqa_f1_score": 0.8557275541795665,
|
388 |
+
"eval_lfqa_loss": 0.27868813276290894,
|
389 |
+
"eval_lfqa_pr_auc": 0.9362703923849833,
|
390 |
+
"eval_lfqa_roc_auc": 0.9523769464424847,
|
391 |
+
"eval_lfqa_runtime": 12.7663,
|
392 |
+
"eval_lfqa_samples_per_second": 149.691,
|
393 |
+
"eval_lfqa_steps_per_second": 2.35,
|
394 |
+
"num_input_tokens_seen": 46209888,
|
395 |
+
"step": 320
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 1.0,
|
399 |
+
"eval_ragtruth_balanced_accuracy": 0.7859350741200968,
|
400 |
+
"eval_ragtruth_f1_score": 0.5078776645041705,
|
401 |
+
"eval_ragtruth_loss": 0.26025664806365967,
|
402 |
+
"eval_ragtruth_pr_auc": 0.4400226298715947,
|
403 |
+
"eval_ragtruth_roc_auc": 0.8972721245465168,
|
404 |
+
"eval_ragtruth_runtime": 149.8477,
|
405 |
+
"eval_ragtruth_samples_per_second": 109.251,
|
406 |
+
"eval_ragtruth_steps_per_second": 1.708,
|
407 |
+
"num_input_tokens_seen": 46209888,
|
408 |
+
"step": 320
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 1.0,
|
412 |
+
"eval_halloumi_synthetic_balanced_accuracy": 0.7681937112954458,
|
413 |
+
"eval_halloumi_synthetic_f1_score": 0.6775431861804223,
|
414 |
+
"eval_halloumi_synthetic_loss": 0.34176450967788696,
|
415 |
+
"eval_halloumi_synthetic_pr_auc": 0.7894696954251237,
|
416 |
+
"eval_halloumi_synthetic_roc_auc": 0.9044560224857751,
|
417 |
+
"eval_halloumi_synthetic_runtime": 38.8927,
|
418 |
+
"eval_halloumi_synthetic_samples_per_second": 53.712,
|
419 |
+
"eval_halloumi_synthetic_steps_per_second": 0.848,
|
420 |
+
"num_input_tokens_seen": 46209888,
|
421 |
+
"step": 320
|
422 |
+
},
|
423 |
+
{
|
424 |
+
"epoch": 1.0,
|
425 |
+
"num_input_tokens_seen": 46209888,
|
426 |
+
"step": 320,
|
427 |
+
"total_flos": 2.418947678713938e+17,
|
428 |
+
"train_loss": 0.2418923556804657,
|
429 |
+
"train_runtime": 3758.6466,
|
430 |
+
"train_samples_per_second": 10.887,
|
431 |
+
"train_steps_per_second": 0.085,
|
432 |
+
"train_tokens_per_second": 1545.744
|
433 |
+
}
|
434 |
+
],
|
435 |
+
"logging_steps": 10,
|
436 |
+
"max_steps": 320,
|
437 |
+
"num_input_tokens_seen": 46209888,
|
438 |
+
"num_train_epochs": 1,
|
439 |
+
"save_steps": 0,
|
440 |
+
"stateful_callbacks": {
|
441 |
+
"TrainerControl": {
|
442 |
+
"args": {
|
443 |
+
"should_epoch_stop": false,
|
444 |
+
"should_evaluate": false,
|
445 |
+
"should_log": false,
|
446 |
+
"should_save": false,
|
447 |
+
"should_training_stop": false
|
448 |
+
},
|
449 |
+
"attributes": {}
|
450 |
+
}
|
451 |
+
},
|
452 |
+
"total_flos": 2.418947678713938e+17,
|
453 |
+
"train_batch_size": 16,
|
454 |
+
"trial_name": null,
|
455 |
+
"trial_params": null
|
456 |
+
}
|