Training in progress, epoch 0
Browse files- adapter_config.json +4 -2
- adapter_model.safetensors +2 -2
- trainer_state.json +324 -0
- training_args.bin +1 -1
adapter_config.json
CHANGED
@@ -20,10 +20,12 @@
|
|
20 |
"rank_pattern": {},
|
21 |
"revision": null,
|
22 |
"target_modules": [
|
23 |
-
"k_proj",
|
24 |
"v_proj",
|
|
|
25 |
"q_proj",
|
26 |
-
"
|
|
|
|
|
27 |
],
|
28 |
"task_type": "CAUSAL_LM",
|
29 |
"use_dora": false,
|
|
|
20 |
"rank_pattern": {},
|
21 |
"revision": null,
|
22 |
"target_modules": [
|
|
|
23 |
"v_proj",
|
24 |
+
"fc1",
|
25 |
"q_proj",
|
26 |
+
"k_proj",
|
27 |
+
"fc2",
|
28 |
+
"dense"
|
29 |
],
|
30 |
"task_type": "CAUSAL_LM",
|
31 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:82158b6472a424bb8a813f866cc91a208f027b85104d960282b6fc6c43ef568a
|
3 |
+
size 47236352
|
trainer_state.json
ADDED
@@ -0,0 +1,324 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 9.824561403508772,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 420,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.23391812865497075,
|
13 |
+
"grad_norm": 11.1328125,
|
14 |
+
"learning_rate": 0.00019972037971811802,
|
15 |
+
"loss": 1.1875,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.4678362573099415,
|
20 |
+
"grad_norm": 9.84375,
|
21 |
+
"learning_rate": 0.00019888308262251285,
|
22 |
+
"loss": 1.0775,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.7017543859649122,
|
27 |
+
"grad_norm": 13.78125,
|
28 |
+
"learning_rate": 0.00019749279121818235,
|
29 |
+
"loss": 0.9793,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.935672514619883,
|
34 |
+
"grad_norm": 11.5859375,
|
35 |
+
"learning_rate": 0.0001955572805786141,
|
36 |
+
"loss": 0.9431,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 1.1695906432748537,
|
41 |
+
"grad_norm": 10.28125,
|
42 |
+
"learning_rate": 0.00019308737486442045,
|
43 |
+
"loss": 0.9035,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.4035087719298245,
|
48 |
+
"grad_norm": 10.578125,
|
49 |
+
"learning_rate": 0.0001900968867902419,
|
50 |
+
"loss": 0.9277,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 1.6374269005847952,
|
55 |
+
"grad_norm": 9.734375,
|
56 |
+
"learning_rate": 0.00018660254037844388,
|
57 |
+
"loss": 0.8933,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 1.871345029239766,
|
62 |
+
"grad_norm": 11.3984375,
|
63 |
+
"learning_rate": 0.0001826238774315995,
|
64 |
+
"loss": 0.8609,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 2.1052631578947367,
|
69 |
+
"grad_norm": 11.046875,
|
70 |
+
"learning_rate": 0.000178183148246803,
|
71 |
+
"loss": 0.8948,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 2.3391812865497075,
|
76 |
+
"grad_norm": 12.171875,
|
77 |
+
"learning_rate": 0.00017330518718298264,
|
78 |
+
"loss": 0.816,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 2.573099415204678,
|
83 |
+
"grad_norm": 12.5390625,
|
84 |
+
"learning_rate": 0.00016801727377709194,
|
85 |
+
"loss": 0.799,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 2.807017543859649,
|
90 |
+
"grad_norm": 11.2890625,
|
91 |
+
"learning_rate": 0.00016234898018587337,
|
92 |
+
"loss": 0.8065,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 3.0409356725146197,
|
97 |
+
"grad_norm": 13.59375,
|
98 |
+
"learning_rate": 0.0001563320058063622,
|
99 |
+
"loss": 0.83,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 3.2748538011695905,
|
104 |
+
"grad_norm": 11.03125,
|
105 |
+
"learning_rate": 0.00015000000000000001,
|
106 |
+
"loss": 0.8086,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 3.5087719298245617,
|
111 |
+
"grad_norm": 14.796875,
|
112 |
+
"learning_rate": 0.00014338837391175582,
|
113 |
+
"loss": 0.7818,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 3.742690058479532,
|
118 |
+
"grad_norm": 11.8359375,
|
119 |
+
"learning_rate": 0.00013653410243663952,
|
120 |
+
"loss": 0.8136,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 3.976608187134503,
|
125 |
+
"grad_norm": 12.0859375,
|
126 |
+
"learning_rate": 0.00012947551744109043,
|
127 |
+
"loss": 0.7879,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 4.2105263157894735,
|
132 |
+
"grad_norm": 13.9765625,
|
133 |
+
"learning_rate": 0.00012225209339563145,
|
134 |
+
"loss": 0.7898,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 4.444444444444445,
|
139 |
+
"grad_norm": 13.9921875,
|
140 |
+
"learning_rate": 0.00011490422661761744,
|
141 |
+
"loss": 0.7845,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 4.678362573099415,
|
146 |
+
"grad_norm": 12.3203125,
|
147 |
+
"learning_rate": 0.00010747300935864243,
|
148 |
+
"loss": 0.8112,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 4.912280701754386,
|
153 |
+
"grad_norm": 12.375,
|
154 |
+
"learning_rate": 0.0001,
|
155 |
+
"loss": 0.7769,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 5.146198830409356,
|
160 |
+
"grad_norm": 11.671875,
|
161 |
+
"learning_rate": 9.252699064135758e-05,
|
162 |
+
"loss": 0.7792,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 5.380116959064328,
|
167 |
+
"grad_norm": 11.34375,
|
168 |
+
"learning_rate": 8.509577338238255e-05,
|
169 |
+
"loss": 0.7844,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 5.614035087719298,
|
174 |
+
"grad_norm": 11.78125,
|
175 |
+
"learning_rate": 7.774790660436858e-05,
|
176 |
+
"loss": 0.7935,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 5.847953216374269,
|
181 |
+
"grad_norm": 14.96875,
|
182 |
+
"learning_rate": 7.052448255890957e-05,
|
183 |
+
"loss": 0.7794,
|
184 |
+
"step": 250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 6.081871345029239,
|
188 |
+
"grad_norm": 12.1171875,
|
189 |
+
"learning_rate": 6.34658975633605e-05,
|
190 |
+
"loss": 0.7809,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 6.315789473684211,
|
195 |
+
"grad_norm": 11.609375,
|
196 |
+
"learning_rate": 5.6611626088244194e-05,
|
197 |
+
"loss": 0.7735,
|
198 |
+
"step": 270
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 6.549707602339181,
|
202 |
+
"grad_norm": 14.2109375,
|
203 |
+
"learning_rate": 5.000000000000002e-05,
|
204 |
+
"loss": 0.7914,
|
205 |
+
"step": 280
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 6.783625730994152,
|
209 |
+
"grad_norm": 11.5234375,
|
210 |
+
"learning_rate": 4.3667994193637796e-05,
|
211 |
+
"loss": 0.7835,
|
212 |
+
"step": 290
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 7.017543859649122,
|
216 |
+
"grad_norm": 12.3046875,
|
217 |
+
"learning_rate": 3.7651019814126654e-05,
|
218 |
+
"loss": 0.766,
|
219 |
+
"step": 300
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 7.251461988304094,
|
223 |
+
"grad_norm": 10.6875,
|
224 |
+
"learning_rate": 3.198272622290804e-05,
|
225 |
+
"loss": 0.7917,
|
226 |
+
"step": 310
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 7.485380116959064,
|
230 |
+
"grad_norm": 11.7734375,
|
231 |
+
"learning_rate": 2.669481281701739e-05,
|
232 |
+
"loss": 0.7678,
|
233 |
+
"step": 320
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 7.719298245614035,
|
237 |
+
"grad_norm": 11.2734375,
|
238 |
+
"learning_rate": 2.181685175319702e-05,
|
239 |
+
"loss": 0.7525,
|
240 |
+
"step": 330
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 7.953216374269006,
|
244 |
+
"grad_norm": 12.1640625,
|
245 |
+
"learning_rate": 1.7376122568400532e-05,
|
246 |
+
"loss": 0.7753,
|
247 |
+
"step": 340
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 8.187134502923977,
|
251 |
+
"grad_norm": 10.2578125,
|
252 |
+
"learning_rate": 1.339745962155613e-05,
|
253 |
+
"loss": 0.7736,
|
254 |
+
"step": 350
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 8.421052631578947,
|
258 |
+
"grad_norm": 10.484375,
|
259 |
+
"learning_rate": 9.903113209758096e-06,
|
260 |
+
"loss": 0.7531,
|
261 |
+
"step": 360
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 8.654970760233919,
|
265 |
+
"grad_norm": 10.328125,
|
266 |
+
"learning_rate": 6.9126251355795864e-06,
|
267 |
+
"loss": 0.7793,
|
268 |
+
"step": 370
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 8.88888888888889,
|
272 |
+
"grad_norm": 13.0546875,
|
273 |
+
"learning_rate": 4.442719421385922e-06,
|
274 |
+
"loss": 0.7737,
|
275 |
+
"step": 380
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 9.12280701754386,
|
279 |
+
"grad_norm": 10.5625,
|
280 |
+
"learning_rate": 2.5072087818176382e-06,
|
281 |
+
"loss": 0.7774,
|
282 |
+
"step": 390
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 9.35672514619883,
|
286 |
+
"grad_norm": 10.4453125,
|
287 |
+
"learning_rate": 1.1169173774871478e-06,
|
288 |
+
"loss": 0.763,
|
289 |
+
"step": 400
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 9.590643274853802,
|
293 |
+
"grad_norm": 11.421875,
|
294 |
+
"learning_rate": 2.7962028188198706e-07,
|
295 |
+
"loss": 0.76,
|
296 |
+
"step": 410
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 9.824561403508772,
|
300 |
+
"grad_norm": 10.5390625,
|
301 |
+
"learning_rate": 0.0,
|
302 |
+
"loss": 0.7646,
|
303 |
+
"step": 420
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 9.824561403508772,
|
307 |
+
"step": 420,
|
308 |
+
"total_flos": 2.74199344054272e+16,
|
309 |
+
"train_loss": 0.8223134313310896,
|
310 |
+
"train_runtime": 362.0016,
|
311 |
+
"train_samples_per_second": 4.724,
|
312 |
+
"train_steps_per_second": 1.16
|
313 |
+
}
|
314 |
+
],
|
315 |
+
"logging_steps": 10,
|
316 |
+
"max_steps": 420,
|
317 |
+
"num_input_tokens_seen": 0,
|
318 |
+
"num_train_epochs": 10,
|
319 |
+
"save_steps": 500,
|
320 |
+
"total_flos": 2.74199344054272e+16,
|
321 |
+
"train_batch_size": 1,
|
322 |
+
"trial_name": null,
|
323 |
+
"trial_params": null
|
324 |
+
}
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5048
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e8d2553dbde1bd63cb09463ac6891981cb5541d43646895a6259ba03d88c135b
|
3 |
size 5048
|