Upload 7 files
Browse files- adapter_model.safetensors +1 -1
- added_tokens.json +127 -0
- optimizer.pt +2 -2
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +223 -5
- training_args.bin +1 -1
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 536991984
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dcc58b43c486c55cd0df2d6042c7ef53ace3843c047209b9cb917a999346456b
|
3 |
size 536991984
|
added_tokens.json
ADDED
@@ -0,0 +1,127 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"</tool_call>": 147090,
|
3 |
+
"<tool_call>": 147089,
|
4 |
+
"<|box_end|>": 147081,
|
5 |
+
"<|box_start|>": 147080,
|
6 |
+
"<|endoftext|>": 147075,
|
7 |
+
"<|file_sep|>": 147096,
|
8 |
+
"<|fim_middle|>": 147092,
|
9 |
+
"<|fim_pad|>": 147094,
|
10 |
+
"<|fim_prefix|>": 147091,
|
11 |
+
"<|fim_suffix|>": 147093,
|
12 |
+
"<|free_token100|>": 147196,
|
13 |
+
"<|free_token101|>": 147197,
|
14 |
+
"<|free_token102|>": 147198,
|
15 |
+
"<|free_token103|>": 147199,
|
16 |
+
"<|free_token10|>": 147106,
|
17 |
+
"<|free_token11|>": 147107,
|
18 |
+
"<|free_token12|>": 147108,
|
19 |
+
"<|free_token13|>": 147109,
|
20 |
+
"<|free_token14|>": 147110,
|
21 |
+
"<|free_token15|>": 147111,
|
22 |
+
"<|free_token16|>": 147112,
|
23 |
+
"<|free_token17|>": 147113,
|
24 |
+
"<|free_token18|>": 147114,
|
25 |
+
"<|free_token19|>": 147115,
|
26 |
+
"<|free_token1|>": 147097,
|
27 |
+
"<|free_token20|>": 147116,
|
28 |
+
"<|free_token21|>": 147117,
|
29 |
+
"<|free_token22|>": 147118,
|
30 |
+
"<|free_token23|>": 147119,
|
31 |
+
"<|free_token24|>": 147120,
|
32 |
+
"<|free_token25|>": 147121,
|
33 |
+
"<|free_token26|>": 147122,
|
34 |
+
"<|free_token27|>": 147123,
|
35 |
+
"<|free_token28|>": 147124,
|
36 |
+
"<|free_token29|>": 147125,
|
37 |
+
"<|free_token2|>": 147098,
|
38 |
+
"<|free_token30|>": 147126,
|
39 |
+
"<|free_token31|>": 147127,
|
40 |
+
"<|free_token32|>": 147128,
|
41 |
+
"<|free_token33|>": 147129,
|
42 |
+
"<|free_token34|>": 147130,
|
43 |
+
"<|free_token35|>": 147131,
|
44 |
+
"<|free_token36|>": 147132,
|
45 |
+
"<|free_token37|>": 147133,
|
46 |
+
"<|free_token38|>": 147134,
|
47 |
+
"<|free_token39|>": 147135,
|
48 |
+
"<|free_token3|>": 147099,
|
49 |
+
"<|free_token40|>": 147136,
|
50 |
+
"<|free_token41|>": 147137,
|
51 |
+
"<|free_token42|>": 147138,
|
52 |
+
"<|free_token43|>": 147139,
|
53 |
+
"<|free_token44|>": 147140,
|
54 |
+
"<|free_token45|>": 147141,
|
55 |
+
"<|free_token46|>": 147142,
|
56 |
+
"<|free_token47|>": 147143,
|
57 |
+
"<|free_token48|>": 147144,
|
58 |
+
"<|free_token49|>": 147145,
|
59 |
+
"<|free_token4|>": 147100,
|
60 |
+
"<|free_token50|>": 147146,
|
61 |
+
"<|free_token51|>": 147147,
|
62 |
+
"<|free_token52|>": 147148,
|
63 |
+
"<|free_token53|>": 147149,
|
64 |
+
"<|free_token54|>": 147150,
|
65 |
+
"<|free_token55|>": 147151,
|
66 |
+
"<|free_token56|>": 147152,
|
67 |
+
"<|free_token57|>": 147153,
|
68 |
+
"<|free_token58|>": 147154,
|
69 |
+
"<|free_token59|>": 147155,
|
70 |
+
"<|free_token5|>": 147101,
|
71 |
+
"<|free_token60|>": 147156,
|
72 |
+
"<|free_token61|>": 147157,
|
73 |
+
"<|free_token62|>": 147158,
|
74 |
+
"<|free_token63|>": 147159,
|
75 |
+
"<|free_token64|>": 147160,
|
76 |
+
"<|free_token65|>": 147161,
|
77 |
+
"<|free_token66|>": 147162,
|
78 |
+
"<|free_token67|>": 147163,
|
79 |
+
"<|free_token68|>": 147164,
|
80 |
+
"<|free_token69|>": 147165,
|
81 |
+
"<|free_token6|>": 147102,
|
82 |
+
"<|free_token70|>": 147166,
|
83 |
+
"<|free_token71|>": 147167,
|
84 |
+
"<|free_token72|>": 147168,
|
85 |
+
"<|free_token73|>": 147169,
|
86 |
+
"<|free_token74|>": 147170,
|
87 |
+
"<|free_token75|>": 147171,
|
88 |
+
"<|free_token76|>": 147172,
|
89 |
+
"<|free_token77|>": 147173,
|
90 |
+
"<|free_token78|>": 147174,
|
91 |
+
"<|free_token79|>": 147175,
|
92 |
+
"<|free_token7|>": 147103,
|
93 |
+
"<|free_token80|>": 147176,
|
94 |
+
"<|free_token81|>": 147177,
|
95 |
+
"<|free_token82|>": 147178,
|
96 |
+
"<|free_token83|>": 147179,
|
97 |
+
"<|free_token84|>": 147180,
|
98 |
+
"<|free_token85|>": 147181,
|
99 |
+
"<|free_token86|>": 147182,
|
100 |
+
"<|free_token87|>": 147183,
|
101 |
+
"<|free_token88|>": 147184,
|
102 |
+
"<|free_token89|>": 147185,
|
103 |
+
"<|free_token8|>": 147104,
|
104 |
+
"<|free_token90|>": 147186,
|
105 |
+
"<|free_token91|>": 147187,
|
106 |
+
"<|free_token92|>": 147188,
|
107 |
+
"<|free_token93|>": 147189,
|
108 |
+
"<|free_token94|>": 147190,
|
109 |
+
"<|free_token95|>": 147191,
|
110 |
+
"<|free_token96|>": 147192,
|
111 |
+
"<|free_token97|>": 147193,
|
112 |
+
"<|free_token98|>": 147194,
|
113 |
+
"<|free_token99|>": 147195,
|
114 |
+
"<|free_token9|>": 147105,
|
115 |
+
"<|im_end|>": 147077,
|
116 |
+
"<|im_start|>": 147076,
|
117 |
+
"<|image_pad|>": 147087,
|
118 |
+
"<|object_ref_end|>": 147079,
|
119 |
+
"<|object_ref_start|>": 147078,
|
120 |
+
"<|quad_end|>": 147083,
|
121 |
+
"<|quad_start|>": 147082,
|
122 |
+
"<|repo_name|>": 147095,
|
123 |
+
"<|video_pad|>": 147088,
|
124 |
+
"<|vision_end|>": 147085,
|
125 |
+
"<|vision_pad|>": 147086,
|
126 |
+
"<|vision_start|>": 147084
|
127 |
+
}
|
optimizer.pt
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:b660ef0d620373c532b67fa6276c72202e837286196a5c035377015ef686e8a6
|
3 |
+
size 276017170
|
rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14244
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:39b824e8acc71bb62addc638c32e739324e2c8ea78f23fdeff754ab9676443af
|
3 |
size 14244
|
scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1064
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:939f8fd57ba6afce9827e7d63bf66494444ef8d81fbec5fe6834bad4c8d0088f
|
3 |
size 1064
|
trainer_state.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
-
"best_metric":
|
3 |
-
"best_model_checkpoint":
|
4 |
-
"epoch":
|
5 |
"eval_steps": 299,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
@@ -210,6 +210,224 @@
|
|
210 |
"learning_rate": 0.00012299465240641713,
|
211 |
"loss": 0.387,
|
212 |
"step": 290
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
213 |
}
|
214 |
],
|
215 |
"logging_steps": 10,
|
@@ -229,7 +447,7 @@
|
|
229 |
"attributes": {}
|
230 |
}
|
231 |
},
|
232 |
-
"total_flos":
|
233 |
"train_batch_size": 1,
|
234 |
"trial_name": null,
|
235 |
"trial_params": null
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.45234861969947815,
|
3 |
+
"best_model_checkpoint": "qlora-extractor-checkpoints/checkpoint-598",
|
4 |
+
"epoch": 2.392,
|
5 |
"eval_steps": 299,
|
6 |
+
"global_step": 598,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
|
|
210 |
"learning_rate": 0.00012299465240641713,
|
211 |
"loss": 0.387,
|
212 |
"step": 290
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 1.2,
|
216 |
+
"grad_norm": 0.18218065798282623,
|
217 |
+
"learning_rate": 0.00012005347593582887,
|
218 |
+
"loss": 0.319,
|
219 |
+
"step": 300
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 1.24,
|
223 |
+
"grad_norm": 0.14094972610473633,
|
224 |
+
"learning_rate": 0.00011737967914438503,
|
225 |
+
"loss": 0.6182,
|
226 |
+
"step": 310
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 1.28,
|
230 |
+
"grad_norm": 0.20308974385261536,
|
231 |
+
"learning_rate": 0.00011470588235294118,
|
232 |
+
"loss": 0.4492,
|
233 |
+
"step": 320
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 1.32,
|
237 |
+
"grad_norm": 0.18815161287784576,
|
238 |
+
"learning_rate": 0.00011203208556149734,
|
239 |
+
"loss": 0.4472,
|
240 |
+
"step": 330
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 1.3599999999999999,
|
244 |
+
"grad_norm": 0.1986108422279358,
|
245 |
+
"learning_rate": 0.00010935828877005347,
|
246 |
+
"loss": 0.3996,
|
247 |
+
"step": 340
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 1.4,
|
251 |
+
"grad_norm": 0.1628069281578064,
|
252 |
+
"learning_rate": 0.00010668449197860964,
|
253 |
+
"loss": 0.3645,
|
254 |
+
"step": 350
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 1.44,
|
258 |
+
"grad_norm": 0.14933669567108154,
|
259 |
+
"learning_rate": 0.00010401069518716577,
|
260 |
+
"loss": 0.5977,
|
261 |
+
"step": 360
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 1.48,
|
265 |
+
"grad_norm": 0.18734735250473022,
|
266 |
+
"learning_rate": 0.00010133689839572193,
|
267 |
+
"loss": 0.3784,
|
268 |
+
"step": 370
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 1.52,
|
272 |
+
"grad_norm": 0.2121153324842453,
|
273 |
+
"learning_rate": 9.866310160427808e-05,
|
274 |
+
"loss": 0.4581,
|
275 |
+
"step": 380
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 1.56,
|
279 |
+
"grad_norm": 0.19748178124427795,
|
280 |
+
"learning_rate": 9.598930481283423e-05,
|
281 |
+
"loss": 0.3858,
|
282 |
+
"step": 390
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 1.6,
|
286 |
+
"grad_norm": 0.18506589531898499,
|
287 |
+
"learning_rate": 9.331550802139037e-05,
|
288 |
+
"loss": 0.3706,
|
289 |
+
"step": 400
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 1.6400000000000001,
|
293 |
+
"grad_norm": 0.15329745411872864,
|
294 |
+
"learning_rate": 9.064171122994652e-05,
|
295 |
+
"loss": 0.5347,
|
296 |
+
"step": 410
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 1.6800000000000002,
|
300 |
+
"grad_norm": 0.19260501861572266,
|
301 |
+
"learning_rate": 8.796791443850267e-05,
|
302 |
+
"loss": 0.4812,
|
303 |
+
"step": 420
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 1.72,
|
307 |
+
"grad_norm": 0.1763940006494522,
|
308 |
+
"learning_rate": 8.529411764705883e-05,
|
309 |
+
"loss": 0.4317,
|
310 |
+
"step": 430
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 1.76,
|
314 |
+
"grad_norm": 0.20417028665542603,
|
315 |
+
"learning_rate": 8.262032085561498e-05,
|
316 |
+
"loss": 0.4145,
|
317 |
+
"step": 440
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 1.8,
|
321 |
+
"grad_norm": 0.2001723051071167,
|
322 |
+
"learning_rate": 7.994652406417112e-05,
|
323 |
+
"loss": 0.3884,
|
324 |
+
"step": 450
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 1.8399999999999999,
|
328 |
+
"grad_norm": 0.2101861983537674,
|
329 |
+
"learning_rate": 7.727272727272727e-05,
|
330 |
+
"loss": 0.4758,
|
331 |
+
"step": 460
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 1.88,
|
335 |
+
"grad_norm": 0.2131994664669037,
|
336 |
+
"learning_rate": 7.459893048128342e-05,
|
337 |
+
"loss": 0.466,
|
338 |
+
"step": 470
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 1.92,
|
342 |
+
"grad_norm": 0.21261604130268097,
|
343 |
+
"learning_rate": 7.192513368983958e-05,
|
344 |
+
"loss": 0.4088,
|
345 |
+
"step": 480
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 1.96,
|
349 |
+
"grad_norm": 0.21039697527885437,
|
350 |
+
"learning_rate": 6.925133689839573e-05,
|
351 |
+
"loss": 0.3858,
|
352 |
+
"step": 490
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 2.0,
|
356 |
+
"grad_norm": 0.18674618005752563,
|
357 |
+
"learning_rate": 6.657754010695188e-05,
|
358 |
+
"loss": 0.3672,
|
359 |
+
"step": 500
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 2.04,
|
363 |
+
"grad_norm": 0.1681678146123886,
|
364 |
+
"learning_rate": 6.390374331550802e-05,
|
365 |
+
"loss": 0.5486,
|
366 |
+
"step": 510
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 2.08,
|
370 |
+
"grad_norm": 0.26280567049980164,
|
371 |
+
"learning_rate": 6.122994652406417e-05,
|
372 |
+
"loss": 0.3358,
|
373 |
+
"step": 520
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 2.12,
|
377 |
+
"grad_norm": 0.24697232246398926,
|
378 |
+
"learning_rate": 5.8556149732620325e-05,
|
379 |
+
"loss": 0.3871,
|
380 |
+
"step": 530
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 2.16,
|
384 |
+
"grad_norm": 0.22857435047626495,
|
385 |
+
"learning_rate": 5.588235294117647e-05,
|
386 |
+
"loss": 0.3822,
|
387 |
+
"step": 540
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 2.2,
|
391 |
+
"grad_norm": 0.2402905523777008,
|
392 |
+
"learning_rate": 5.320855614973263e-05,
|
393 |
+
"loss": 0.3009,
|
394 |
+
"step": 550
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 2.24,
|
398 |
+
"grad_norm": 0.19268670678138733,
|
399 |
+
"learning_rate": 5.0534759358288774e-05,
|
400 |
+
"loss": 0.3916,
|
401 |
+
"step": 560
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 2.2800000000000002,
|
405 |
+
"grad_norm": 0.2288196086883545,
|
406 |
+
"learning_rate": 4.786096256684492e-05,
|
407 |
+
"loss": 0.338,
|
408 |
+
"step": 570
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 2.32,
|
412 |
+
"grad_norm": 0.2444332093000412,
|
413 |
+
"learning_rate": 4.518716577540107e-05,
|
414 |
+
"loss": 0.3971,
|
415 |
+
"step": 580
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 2.36,
|
419 |
+
"grad_norm": 0.26502081751823425,
|
420 |
+
"learning_rate": 4.251336898395722e-05,
|
421 |
+
"loss": 0.2905,
|
422 |
+
"step": 590
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 2.392,
|
426 |
+
"eval_loss": 0.45234861969947815,
|
427 |
+
"eval_runtime": 967.7709,
|
428 |
+
"eval_samples_per_second": 0.207,
|
429 |
+
"eval_steps_per_second": 0.207,
|
430 |
+
"step": 598
|
431 |
}
|
432 |
],
|
433 |
"logging_steps": 10,
|
|
|
447 |
"attributes": {}
|
448 |
}
|
449 |
},
|
450 |
+
"total_flos": 3.783168801781924e+18,
|
451 |
"train_batch_size": 1,
|
452 |
"trial_name": null,
|
453 |
"trial_params": null
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5304
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:57a8acdebd4b287f899231aaa627e1307e96ad0a6aac4f0ee5f02d9205331add
|
3 |
size 5304
|