Training in progress, step 50, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
@@ -20,12 +20,12 @@
|
|
20 |
"rank_pattern": {},
|
21 |
"revision": null,
|
22 |
"target_modules": [
|
23 |
-
"down_proj",
|
24 |
"up_proj",
|
25 |
-
"k_proj",
|
26 |
"v_proj",
|
27 |
-
"
|
|
|
28 |
"q_proj",
|
|
|
29 |
"o_proj"
|
30 |
],
|
31 |
"task_type": "CAUSAL_LM",
|
|
|
20 |
"rank_pattern": {},
|
21 |
"revision": null,
|
22 |
"target_modules": [
|
|
|
23 |
"up_proj",
|
|
|
24 |
"v_proj",
|
25 |
+
"down_proj",
|
26 |
+
"k_proj",
|
27 |
"q_proj",
|
28 |
+
"gate_proj",
|
29 |
"o_proj"
|
30 |
],
|
31 |
"task_type": "CAUSAL_LM",
|
last-checkpoint/adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 156926880
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:22f1bccf462e950147942e2f42336e85c24a531043ea1660a67b2e981a1e0dce
|
3 |
size 156926880
|
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 313998650
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bdf710d0d02c2e9314f982048011900cfd23d1e788a7a448122894d1532d73f2
|
3 |
size 313998650
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
"best_model_checkpoint": "miner_id_24/checkpoint-50",
|
4 |
"epoch": 0.05534417156693186,
|
5 |
"eval_steps": 25,
|
@@ -10,7 +10,7 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.0011068834313386372,
|
13 |
-
"grad_norm":
|
14 |
"learning_rate": 5e-05,
|
15 |
"loss": 14.1912,
|
16 |
"step": 1
|
@@ -18,368 +18,368 @@
|
|
18 |
{
|
19 |
"epoch": 0.0011068834313386372,
|
20 |
"eval_loss": 13.724494934082031,
|
21 |
-
"eval_runtime":
|
22 |
-
"eval_samples_per_second":
|
23 |
-
"eval_steps_per_second": 8.
|
24 |
"step": 1
|
25 |
},
|
26 |
{
|
27 |
"epoch": 0.0022137668626772745,
|
28 |
-
"grad_norm":
|
29 |
"learning_rate": 0.0001,
|
30 |
"loss": 13.95,
|
31 |
"step": 2
|
32 |
},
|
33 |
{
|
34 |
"epoch": 0.0033206502940159115,
|
35 |
-
"grad_norm":
|
36 |
"learning_rate": 9.989294616193017e-05,
|
37 |
-
"loss": 12.
|
38 |
"step": 3
|
39 |
},
|
40 |
{
|
41 |
"epoch": 0.004427533725354549,
|
42 |
-
"grad_norm":
|
43 |
"learning_rate": 9.957224306869053e-05,
|
44 |
-
"loss":
|
45 |
"step": 4
|
46 |
},
|
47 |
{
|
48 |
"epoch": 0.005534417156693186,
|
49 |
-
"grad_norm": 22.
|
50 |
"learning_rate": 9.903926402016153e-05,
|
51 |
-
"loss": 3.
|
52 |
"step": 5
|
53 |
},
|
54 |
{
|
55 |
"epoch": 0.006641300588031823,
|
56 |
-
"grad_norm":
|
57 |
"learning_rate": 9.829629131445342e-05,
|
58 |
-
"loss": 2.
|
59 |
"step": 6
|
60 |
},
|
61 |
{
|
62 |
"epoch": 0.00774818401937046,
|
63 |
-
"grad_norm": 10.
|
64 |
"learning_rate": 9.73465064747553e-05,
|
65 |
-
"loss": 1.
|
66 |
"step": 7
|
67 |
},
|
68 |
{
|
69 |
"epoch": 0.008855067450709098,
|
70 |
-
"grad_norm": 6.
|
71 |
"learning_rate": 9.619397662556435e-05,
|
72 |
-
"loss": 1.
|
73 |
"step": 8
|
74 |
},
|
75 |
{
|
76 |
"epoch": 0.009961950882047734,
|
77 |
-
"grad_norm":
|
78 |
"learning_rate": 9.484363707663442e-05,
|
79 |
-
"loss": 0.
|
80 |
"step": 9
|
81 |
},
|
82 |
{
|
83 |
"epoch": 0.011068834313386372,
|
84 |
-
"grad_norm":
|
85 |
"learning_rate": 9.330127018922194e-05,
|
86 |
-
"loss": 0.
|
87 |
"step": 10
|
88 |
},
|
89 |
{
|
90 |
"epoch": 0.012175717744725008,
|
91 |
-
"grad_norm":
|
92 |
"learning_rate": 9.157348061512727e-05,
|
93 |
-
"loss": 0.
|
94 |
"step": 11
|
95 |
},
|
96 |
{
|
97 |
"epoch": 0.013282601176063646,
|
98 |
-
"grad_norm":
|
99 |
"learning_rate": 8.966766701456177e-05,
|
100 |
-
"loss": 0.
|
101 |
"step": 12
|
102 |
},
|
103 |
{
|
104 |
"epoch": 0.014389484607402282,
|
105 |
-
"grad_norm":
|
106 |
"learning_rate": 8.759199037394887e-05,
|
107 |
-
"loss":
|
108 |
"step": 13
|
109 |
},
|
110 |
{
|
111 |
"epoch": 0.01549636803874092,
|
112 |
-
"grad_norm": 4.
|
113 |
"learning_rate": 8.535533905932738e-05,
|
114 |
-
"loss": 0.
|
115 |
"step": 14
|
116 |
},
|
117 |
{
|
118 |
"epoch": 0.016603251470079558,
|
119 |
-
"grad_norm": 2.
|
120 |
"learning_rate": 8.296729075500344e-05,
|
121 |
-
"loss": 0.
|
122 |
"step": 15
|
123 |
},
|
124 |
{
|
125 |
"epoch": 0.017710134901418196,
|
126 |
-
"grad_norm":
|
127 |
"learning_rate": 8.043807145043604e-05,
|
128 |
-
"loss": 0.
|
129 |
"step": 16
|
130 |
},
|
131 |
{
|
132 |
"epoch": 0.01881701833275683,
|
133 |
-
"grad_norm":
|
134 |
"learning_rate": 7.777851165098012e-05,
|
135 |
-
"loss": 0.
|
136 |
"step": 17
|
137 |
},
|
138 |
{
|
139 |
"epoch": 0.019923901764095468,
|
140 |
-
"grad_norm": 5.
|
141 |
"learning_rate": 7.500000000000001e-05,
|
142 |
-
"loss": 0.
|
143 |
"step": 18
|
144 |
},
|
145 |
{
|
146 |
"epoch": 0.021030785195434106,
|
147 |
-
"grad_norm":
|
148 |
"learning_rate": 7.211443451095007e-05,
|
149 |
-
"loss": 0.
|
150 |
"step": 19
|
151 |
},
|
152 |
{
|
153 |
"epoch": 0.022137668626772744,
|
154 |
-
"grad_norm":
|
155 |
"learning_rate": 6.91341716182545e-05,
|
156 |
-
"loss": 0.
|
157 |
"step": 20
|
158 |
},
|
159 |
{
|
160 |
"epoch": 0.02324455205811138,
|
161 |
-
"grad_norm":
|
162 |
"learning_rate": 6.607197326515808e-05,
|
163 |
-
"loss": 0.
|
164 |
"step": 21
|
165 |
},
|
166 |
{
|
167 |
"epoch": 0.024351435489450016,
|
168 |
-
"grad_norm": 2.
|
169 |
"learning_rate": 6.294095225512603e-05,
|
170 |
-
"loss": 0.
|
171 |
"step": 22
|
172 |
},
|
173 |
{
|
174 |
"epoch": 0.025458318920788654,
|
175 |
-
"grad_norm": 3.
|
176 |
"learning_rate": 5.9754516100806423e-05,
|
177 |
-
"loss": 0.
|
178 |
"step": 23
|
179 |
},
|
180 |
{
|
181 |
"epoch": 0.026565202352127292,
|
182 |
-
"grad_norm": 2.
|
183 |
"learning_rate": 5.6526309611002594e-05,
|
184 |
-
"loss": 0.
|
185 |
"step": 24
|
186 |
},
|
187 |
{
|
188 |
"epoch": 0.02767208578346593,
|
189 |
-
"grad_norm": 1.
|
190 |
"learning_rate": 5.327015646150716e-05,
|
191 |
-
"loss": 0.
|
192 |
"step": 25
|
193 |
},
|
194 |
{
|
195 |
"epoch": 0.02767208578346593,
|
196 |
-
"eval_loss": 0.
|
197 |
-
"eval_runtime":
|
198 |
-
"eval_samples_per_second":
|
199 |
-
"eval_steps_per_second": 8.
|
200 |
"step": 25
|
201 |
},
|
202 |
{
|
203 |
"epoch": 0.028778969214804564,
|
204 |
-
"grad_norm": 3.
|
205 |
"learning_rate": 5e-05,
|
206 |
-
"loss": 0.
|
207 |
"step": 26
|
208 |
},
|
209 |
{
|
210 |
"epoch": 0.029885852646143202,
|
211 |
-
"grad_norm": 2.
|
212 |
"learning_rate": 4.6729843538492847e-05,
|
213 |
-
"loss": 0.
|
214 |
"step": 27
|
215 |
},
|
216 |
{
|
217 |
"epoch": 0.03099273607748184,
|
218 |
-
"grad_norm":
|
219 |
"learning_rate": 4.347369038899744e-05,
|
220 |
-
"loss": 0.
|
221 |
"step": 28
|
222 |
},
|
223 |
{
|
224 |
"epoch": 0.032099619508820475,
|
225 |
-
"grad_norm": 1.
|
226 |
"learning_rate": 4.0245483899193595e-05,
|
227 |
-
"loss": 0.
|
228 |
"step": 29
|
229 |
},
|
230 |
{
|
231 |
"epoch": 0.033206502940159116,
|
232 |
-
"grad_norm": 0.
|
233 |
"learning_rate": 3.705904774487396e-05,
|
234 |
-
"loss": 0.
|
235 |
"step": 30
|
236 |
},
|
237 |
{
|
238 |
"epoch": 0.03431338637149775,
|
239 |
-
"grad_norm": 2.
|
240 |
"learning_rate": 3.392802673484193e-05,
|
241 |
-
"loss": 0.
|
242 |
"step": 31
|
243 |
},
|
244 |
{
|
245 |
"epoch": 0.03542026980283639,
|
246 |
-
"grad_norm": 1.
|
247 |
"learning_rate": 3.086582838174551e-05,
|
248 |
-
"loss": 0.
|
249 |
"step": 32
|
250 |
},
|
251 |
{
|
252 |
"epoch": 0.036527153234175026,
|
253 |
-
"grad_norm":
|
254 |
"learning_rate": 2.7885565489049946e-05,
|
255 |
-
"loss": 0.
|
256 |
"step": 33
|
257 |
},
|
258 |
{
|
259 |
"epoch": 0.03763403666551366,
|
260 |
-
"grad_norm":
|
261 |
"learning_rate": 2.500000000000001e-05,
|
262 |
-
"loss": 0.
|
263 |
"step": 34
|
264 |
},
|
265 |
{
|
266 |
"epoch": 0.0387409200968523,
|
267 |
-
"grad_norm":
|
268 |
"learning_rate": 2.2221488349019903e-05,
|
269 |
-
"loss": 0.
|
270 |
"step": 35
|
271 |
},
|
272 |
{
|
273 |
"epoch": 0.039847803528190937,
|
274 |
-
"grad_norm": 1.
|
275 |
"learning_rate": 1.9561928549563968e-05,
|
276 |
-
"loss": 0.
|
277 |
"step": 36
|
278 |
},
|
279 |
{
|
280 |
"epoch": 0.04095468695952958,
|
281 |
-
"grad_norm": 1.
|
282 |
"learning_rate": 1.703270924499656e-05,
|
283 |
-
"loss": 0.
|
284 |
"step": 37
|
285 |
},
|
286 |
{
|
287 |
"epoch": 0.04206157039086821,
|
288 |
-
"grad_norm":
|
289 |
"learning_rate": 1.4644660940672627e-05,
|
290 |
-
"loss": 0.
|
291 |
"step": 38
|
292 |
},
|
293 |
{
|
294 |
"epoch": 0.04316845382220685,
|
295 |
-
"grad_norm": 1.
|
296 |
"learning_rate": 1.2408009626051137e-05,
|
297 |
-
"loss": 0.
|
298 |
"step": 39
|
299 |
},
|
300 |
{
|
301 |
"epoch": 0.04427533725354549,
|
302 |
-
"grad_norm":
|
303 |
"learning_rate": 1.0332332985438248e-05,
|
304 |
-
"loss": 0.
|
305 |
"step": 40
|
306 |
},
|
307 |
{
|
308 |
"epoch": 0.04538222068488412,
|
309 |
-
"grad_norm": 1.
|
310 |
"learning_rate": 8.426519384872733e-06,
|
311 |
-
"loss": 0.
|
312 |
"step": 41
|
313 |
},
|
314 |
{
|
315 |
"epoch": 0.04648910411622276,
|
316 |
-
"grad_norm": 1.
|
317 |
"learning_rate": 6.698729810778065e-06,
|
318 |
-
"loss": 0.
|
319 |
"step": 42
|
320 |
},
|
321 |
{
|
322 |
"epoch": 0.0475959875475614,
|
323 |
-
"grad_norm":
|
324 |
"learning_rate": 5.156362923365588e-06,
|
325 |
-
"loss": 0.
|
326 |
"step": 43
|
327 |
},
|
328 |
{
|
329 |
"epoch": 0.04870287097890003,
|
330 |
-
"grad_norm": 2.
|
331 |
"learning_rate": 3.8060233744356633e-06,
|
332 |
-
"loss": 0.
|
333 |
"step": 44
|
334 |
},
|
335 |
{
|
336 |
"epoch": 0.049809754410238674,
|
337 |
-
"grad_norm": 1.
|
338 |
"learning_rate": 2.653493525244721e-06,
|
339 |
-
"loss": 0.
|
340 |
"step": 45
|
341 |
},
|
342 |
{
|
343 |
"epoch": 0.05091663784157731,
|
344 |
-
"grad_norm": 2.
|
345 |
"learning_rate": 1.70370868554659e-06,
|
346 |
-
"loss": 0.
|
347 |
"step": 46
|
348 |
},
|
349 |
{
|
350 |
"epoch": 0.05202352127291594,
|
351 |
-
"grad_norm": 1.
|
352 |
"learning_rate": 9.607359798384785e-07,
|
353 |
-
"loss": 0.
|
354 |
"step": 47
|
355 |
},
|
356 |
{
|
357 |
"epoch": 0.053130404704254584,
|
358 |
-
"grad_norm": 1.
|
359 |
"learning_rate": 4.277569313094809e-07,
|
360 |
-
"loss": 0.
|
361 |
"step": 48
|
362 |
},
|
363 |
{
|
364 |
"epoch": 0.05423728813559322,
|
365 |
-
"grad_norm": 1.
|
366 |
"learning_rate": 1.0705383806982606e-07,
|
367 |
-
"loss": 0.
|
368 |
"step": 49
|
369 |
},
|
370 |
{
|
371 |
"epoch": 0.05534417156693186,
|
372 |
-
"grad_norm": 1.
|
373 |
"learning_rate": 0.0,
|
374 |
-
"loss": 0.
|
375 |
"step": 50
|
376 |
},
|
377 |
{
|
378 |
"epoch": 0.05534417156693186,
|
379 |
-
"eval_loss": 0.
|
380 |
-
"eval_runtime":
|
381 |
-
"eval_samples_per_second":
|
382 |
-
"eval_steps_per_second": 8.
|
383 |
"step": 50
|
384 |
}
|
385 |
],
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.8104254007339478,
|
3 |
"best_model_checkpoint": "miner_id_24/checkpoint-50",
|
4 |
"epoch": 0.05534417156693186,
|
5 |
"eval_steps": 25,
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.0011068834313386372,
|
13 |
+
"grad_norm": 36.721168518066406,
|
14 |
"learning_rate": 5e-05,
|
15 |
"loss": 14.1912,
|
16 |
"step": 1
|
|
|
18 |
{
|
19 |
"epoch": 0.0011068834313386372,
|
20 |
"eval_loss": 13.724494934082031,
|
21 |
+
"eval_runtime": 91.0519,
|
22 |
+
"eval_samples_per_second": 66.852,
|
23 |
+
"eval_steps_per_second": 8.358,
|
24 |
"step": 1
|
25 |
},
|
26 |
{
|
27 |
"epoch": 0.0022137668626772745,
|
28 |
+
"grad_norm": 45.76126480102539,
|
29 |
"learning_rate": 0.0001,
|
30 |
"loss": 13.95,
|
31 |
"step": 2
|
32 |
},
|
33 |
{
|
34 |
"epoch": 0.0033206502940159115,
|
35 |
+
"grad_norm": 44.33281707763672,
|
36 |
"learning_rate": 9.989294616193017e-05,
|
37 |
+
"loss": 12.4936,
|
38 |
"step": 3
|
39 |
},
|
40 |
{
|
41 |
"epoch": 0.004427533725354549,
|
42 |
+
"grad_norm": 41.054508209228516,
|
43 |
"learning_rate": 9.957224306869053e-05,
|
44 |
+
"loss": 7.0168,
|
45 |
"step": 4
|
46 |
},
|
47 |
{
|
48 |
"epoch": 0.005534417156693186,
|
49 |
+
"grad_norm": 22.34103012084961,
|
50 |
"learning_rate": 9.903926402016153e-05,
|
51 |
+
"loss": 3.5822,
|
52 |
"step": 5
|
53 |
},
|
54 |
{
|
55 |
"epoch": 0.006641300588031823,
|
56 |
+
"grad_norm": 25.822521209716797,
|
57 |
"learning_rate": 9.829629131445342e-05,
|
58 |
+
"loss": 2.2788,
|
59 |
"step": 6
|
60 |
},
|
61 |
{
|
62 |
"epoch": 0.00774818401937046,
|
63 |
+
"grad_norm": 10.615482330322266,
|
64 |
"learning_rate": 9.73465064747553e-05,
|
65 |
+
"loss": 1.4014,
|
66 |
"step": 7
|
67 |
},
|
68 |
{
|
69 |
"epoch": 0.008855067450709098,
|
70 |
+
"grad_norm": 6.933248519897461,
|
71 |
"learning_rate": 9.619397662556435e-05,
|
72 |
+
"loss": 1.0412,
|
73 |
"step": 8
|
74 |
},
|
75 |
{
|
76 |
"epoch": 0.009961950882047734,
|
77 |
+
"grad_norm": 2.6085708141326904,
|
78 |
"learning_rate": 9.484363707663442e-05,
|
79 |
+
"loss": 0.8551,
|
80 |
"step": 9
|
81 |
},
|
82 |
{
|
83 |
"epoch": 0.011068834313386372,
|
84 |
+
"grad_norm": 3.8146591186523438,
|
85 |
"learning_rate": 9.330127018922194e-05,
|
86 |
+
"loss": 0.8501,
|
87 |
"step": 10
|
88 |
},
|
89 |
{
|
90 |
"epoch": 0.012175717744725008,
|
91 |
+
"grad_norm": 7.357889175415039,
|
92 |
"learning_rate": 9.157348061512727e-05,
|
93 |
+
"loss": 0.8831,
|
94 |
"step": 11
|
95 |
},
|
96 |
{
|
97 |
"epoch": 0.013282601176063646,
|
98 |
+
"grad_norm": 4.156259059906006,
|
99 |
"learning_rate": 8.966766701456177e-05,
|
100 |
+
"loss": 0.8321,
|
101 |
"step": 12
|
102 |
},
|
103 |
{
|
104 |
"epoch": 0.014389484607402282,
|
105 |
+
"grad_norm": 13.24997615814209,
|
106 |
"learning_rate": 8.759199037394887e-05,
|
107 |
+
"loss": 1.0533,
|
108 |
"step": 13
|
109 |
},
|
110 |
{
|
111 |
"epoch": 0.01549636803874092,
|
112 |
+
"grad_norm": 4.824550628662109,
|
113 |
"learning_rate": 8.535533905932738e-05,
|
114 |
+
"loss": 0.8486,
|
115 |
"step": 14
|
116 |
},
|
117 |
{
|
118 |
"epoch": 0.016603251470079558,
|
119 |
+
"grad_norm": 2.509622097015381,
|
120 |
"learning_rate": 8.296729075500344e-05,
|
121 |
+
"loss": 0.7962,
|
122 |
"step": 15
|
123 |
},
|
124 |
{
|
125 |
"epoch": 0.017710134901418196,
|
126 |
+
"grad_norm": 3.590033769607544,
|
127 |
"learning_rate": 8.043807145043604e-05,
|
128 |
+
"loss": 0.8333,
|
129 |
"step": 16
|
130 |
},
|
131 |
{
|
132 |
"epoch": 0.01881701833275683,
|
133 |
+
"grad_norm": 6.873021125793457,
|
134 |
"learning_rate": 7.777851165098012e-05,
|
135 |
+
"loss": 0.8969,
|
136 |
"step": 17
|
137 |
},
|
138 |
{
|
139 |
"epoch": 0.019923901764095468,
|
140 |
+
"grad_norm": 5.084534645080566,
|
141 |
"learning_rate": 7.500000000000001e-05,
|
142 |
+
"loss": 0.8647,
|
143 |
"step": 18
|
144 |
},
|
145 |
{
|
146 |
"epoch": 0.021030785195434106,
|
147 |
+
"grad_norm": 2.3888447284698486,
|
148 |
"learning_rate": 7.211443451095007e-05,
|
149 |
+
"loss": 0.8115,
|
150 |
"step": 19
|
151 |
},
|
152 |
{
|
153 |
"epoch": 0.022137668626772744,
|
154 |
+
"grad_norm": 1.8158559799194336,
|
155 |
"learning_rate": 6.91341716182545e-05,
|
156 |
+
"loss": 0.8075,
|
157 |
"step": 20
|
158 |
},
|
159 |
{
|
160 |
"epoch": 0.02324455205811138,
|
161 |
+
"grad_norm": 2.9052810668945312,
|
162 |
"learning_rate": 6.607197326515808e-05,
|
163 |
+
"loss": 0.8259,
|
164 |
"step": 21
|
165 |
},
|
166 |
{
|
167 |
"epoch": 0.024351435489450016,
|
168 |
+
"grad_norm": 2.515951156616211,
|
169 |
"learning_rate": 6.294095225512603e-05,
|
170 |
+
"loss": 0.8207,
|
171 |
"step": 22
|
172 |
},
|
173 |
{
|
174 |
"epoch": 0.025458318920788654,
|
175 |
+
"grad_norm": 3.8959262371063232,
|
176 |
"learning_rate": 5.9754516100806423e-05,
|
177 |
+
"loss": 0.8512,
|
178 |
"step": 23
|
179 |
},
|
180 |
{
|
181 |
"epoch": 0.026565202352127292,
|
182 |
+
"grad_norm": 2.965120792388916,
|
183 |
"learning_rate": 5.6526309611002594e-05,
|
184 |
+
"loss": 0.8395,
|
185 |
"step": 24
|
186 |
},
|
187 |
{
|
188 |
"epoch": 0.02767208578346593,
|
189 |
+
"grad_norm": 1.2737544775009155,
|
190 |
"learning_rate": 5.327015646150716e-05,
|
191 |
+
"loss": 0.8124,
|
192 |
"step": 25
|
193 |
},
|
194 |
{
|
195 |
"epoch": 0.02767208578346593,
|
196 |
+
"eval_loss": 0.8200458288192749,
|
197 |
+
"eval_runtime": 90.9598,
|
198 |
+
"eval_samples_per_second": 66.92,
|
199 |
+
"eval_steps_per_second": 8.366,
|
200 |
"step": 25
|
201 |
},
|
202 |
{
|
203 |
"epoch": 0.028778969214804564,
|
204 |
+
"grad_norm": 3.0076019763946533,
|
205 |
"learning_rate": 5e-05,
|
206 |
+
"loss": 0.914,
|
207 |
"step": 26
|
208 |
},
|
209 |
{
|
210 |
"epoch": 0.029885852646143202,
|
211 |
+
"grad_norm": 2.437142848968506,
|
212 |
"learning_rate": 4.6729843538492847e-05,
|
213 |
+
"loss": 0.8225,
|
214 |
"step": 27
|
215 |
},
|
216 |
{
|
217 |
"epoch": 0.03099273607748184,
|
218 |
+
"grad_norm": 2.596731185913086,
|
219 |
"learning_rate": 4.347369038899744e-05,
|
220 |
+
"loss": 0.8304,
|
221 |
"step": 28
|
222 |
},
|
223 |
{
|
224 |
"epoch": 0.032099619508820475,
|
225 |
+
"grad_norm": 1.6621183156967163,
|
226 |
"learning_rate": 4.0245483899193595e-05,
|
227 |
+
"loss": 0.8103,
|
228 |
"step": 29
|
229 |
},
|
230 |
{
|
231 |
"epoch": 0.033206502940159116,
|
232 |
+
"grad_norm": 0.6339697241783142,
|
233 |
"learning_rate": 3.705904774487396e-05,
|
234 |
+
"loss": 0.8156,
|
235 |
"step": 30
|
236 |
},
|
237 |
{
|
238 |
"epoch": 0.03431338637149775,
|
239 |
+
"grad_norm": 2.309584617614746,
|
240 |
"learning_rate": 3.392802673484193e-05,
|
241 |
+
"loss": 0.8227,
|
242 |
"step": 31
|
243 |
},
|
244 |
{
|
245 |
"epoch": 0.03542026980283639,
|
246 |
+
"grad_norm": 1.88184654712677,
|
247 |
"learning_rate": 3.086582838174551e-05,
|
248 |
+
"loss": 0.819,
|
249 |
"step": 32
|
250 |
},
|
251 |
{
|
252 |
"epoch": 0.036527153234175026,
|
253 |
+
"grad_norm": 1.8324432373046875,
|
254 |
"learning_rate": 2.7885565489049946e-05,
|
255 |
+
"loss": 0.8012,
|
256 |
"step": 33
|
257 |
},
|
258 |
{
|
259 |
"epoch": 0.03763403666551366,
|
260 |
+
"grad_norm": 1.9068471193313599,
|
261 |
"learning_rate": 2.500000000000001e-05,
|
262 |
+
"loss": 0.8098,
|
263 |
"step": 34
|
264 |
},
|
265 |
{
|
266 |
"epoch": 0.0387409200968523,
|
267 |
+
"grad_norm": 0.5746772885322571,
|
268 |
"learning_rate": 2.2221488349019903e-05,
|
269 |
+
"loss": 0.8211,
|
270 |
"step": 35
|
271 |
},
|
272 |
{
|
273 |
"epoch": 0.039847803528190937,
|
274 |
+
"grad_norm": 1.8682575225830078,
|
275 |
"learning_rate": 1.9561928549563968e-05,
|
276 |
+
"loss": 0.8151,
|
277 |
"step": 36
|
278 |
},
|
279 |
{
|
280 |
"epoch": 0.04095468695952958,
|
281 |
+
"grad_norm": 1.3192604780197144,
|
282 |
"learning_rate": 1.703270924499656e-05,
|
283 |
+
"loss": 0.8096,
|
284 |
"step": 37
|
285 |
},
|
286 |
{
|
287 |
"epoch": 0.04206157039086821,
|
288 |
+
"grad_norm": 10.249944686889648,
|
289 |
"learning_rate": 1.4644660940672627e-05,
|
290 |
+
"loss": 0.9605,
|
291 |
"step": 38
|
292 |
},
|
293 |
{
|
294 |
"epoch": 0.04316845382220685,
|
295 |
+
"grad_norm": 1.324076533317566,
|
296 |
"learning_rate": 1.2408009626051137e-05,
|
297 |
+
"loss": 0.8249,
|
298 |
"step": 39
|
299 |
},
|
300 |
{
|
301 |
"epoch": 0.04427533725354549,
|
302 |
+
"grad_norm": 0.9887760877609253,
|
303 |
"learning_rate": 1.0332332985438248e-05,
|
304 |
+
"loss": 0.8182,
|
305 |
"step": 40
|
306 |
},
|
307 |
{
|
308 |
"epoch": 0.04538222068488412,
|
309 |
+
"grad_norm": 1.1598294973373413,
|
310 |
"learning_rate": 8.426519384872733e-06,
|
311 |
+
"loss": 0.8137,
|
312 |
"step": 41
|
313 |
},
|
314 |
{
|
315 |
"epoch": 0.04648910411622276,
|
316 |
+
"grad_norm": 1.0750510692596436,
|
317 |
"learning_rate": 6.698729810778065e-06,
|
318 |
+
"loss": 0.8027,
|
319 |
"step": 42
|
320 |
},
|
321 |
{
|
322 |
"epoch": 0.0475959875475614,
|
323 |
+
"grad_norm": 1.1602271795272827,
|
324 |
"learning_rate": 5.156362923365588e-06,
|
325 |
+
"loss": 0.8111,
|
326 |
"step": 43
|
327 |
},
|
328 |
{
|
329 |
"epoch": 0.04870287097890003,
|
330 |
+
"grad_norm": 2.6957943439483643,
|
331 |
"learning_rate": 3.8060233744356633e-06,
|
332 |
+
"loss": 0.8029,
|
333 |
"step": 44
|
334 |
},
|
335 |
{
|
336 |
"epoch": 0.049809754410238674,
|
337 |
+
"grad_norm": 1.0971044301986694,
|
338 |
"learning_rate": 2.653493525244721e-06,
|
339 |
+
"loss": 0.807,
|
340 |
"step": 45
|
341 |
},
|
342 |
{
|
343 |
"epoch": 0.05091663784157731,
|
344 |
+
"grad_norm": 2.0505218505859375,
|
345 |
"learning_rate": 1.70370868554659e-06,
|
346 |
+
"loss": 0.8122,
|
347 |
"step": 46
|
348 |
},
|
349 |
{
|
350 |
"epoch": 0.05202352127291594,
|
351 |
+
"grad_norm": 1.332539439201355,
|
352 |
"learning_rate": 9.607359798384785e-07,
|
353 |
+
"loss": 0.8038,
|
354 |
"step": 47
|
355 |
},
|
356 |
{
|
357 |
"epoch": 0.053130404704254584,
|
358 |
+
"grad_norm": 1.3104387521743774,
|
359 |
"learning_rate": 4.277569313094809e-07,
|
360 |
+
"loss": 0.807,
|
361 |
"step": 48
|
362 |
},
|
363 |
{
|
364 |
"epoch": 0.05423728813559322,
|
365 |
+
"grad_norm": 1.1924852132797241,
|
366 |
"learning_rate": 1.0705383806982606e-07,
|
367 |
+
"loss": 0.8063,
|
368 |
"step": 49
|
369 |
},
|
370 |
{
|
371 |
"epoch": 0.05534417156693186,
|
372 |
+
"grad_norm": 1.5198986530303955,
|
373 |
"learning_rate": 0.0,
|
374 |
+
"loss": 0.812,
|
375 |
"step": 50
|
376 |
},
|
377 |
{
|
378 |
"epoch": 0.05534417156693186,
|
379 |
+
"eval_loss": 0.8104254007339478,
|
380 |
+
"eval_runtime": 91.4608,
|
381 |
+
"eval_samples_per_second": 66.553,
|
382 |
+
"eval_steps_per_second": 8.321,
|
383 |
"step": 50
|
384 |
}
|
385 |
],
|
last-checkpoint/training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 6776
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7585702810afab1a596eec3bf36188145c60d08d63b6a9e1c02b97bf5d9db974
|
3 |
size 6776
|