Hugging Face's logo

tyzhu
/
lmind_nq_train6000_eval6489_v1_doc_qa_v3_3e-5_lora2

Safetensors
Generated from Trainer
Eval Results
Model card Files Community
tyzhu commited on
Commit
8249baf
•
1 Parent(s): a08bc39

Training in progress, epoch 13, checkpoint

Browse files
Files changed (17) hide show
  1. checkpoint-6877/README.md +9 -0
  2. checkpoint-6877/adapter_config.json +25 -0
  3. checkpoint-6877/adapter_model.safetensors +3 -0
  4. checkpoint-6877/added_tokens.json +5 -0
  5. checkpoint-6877/merges.txt +0 -0
  6. checkpoint-6877/optimizer.pt +3 -0
  7. checkpoint-6877/rng_state_0.pth +3 -0
  8. checkpoint-6877/rng_state_1.pth +3 -0
  9. checkpoint-6877/rng_state_2.pth +3 -0
  10. checkpoint-6877/rng_state_3.pth +3 -0
  11. checkpoint-6877/scheduler.pt +3 -0
  12. checkpoint-6877/special_tokens_map.json +14 -0
  13. checkpoint-6877/tokenizer.json +0 -0
  14. checkpoint-6877/tokenizer_config.json +43 -0
  15. checkpoint-6877/trainer_state.json +704 -0
  16. checkpoint-6877/training_args.bin +3 -0
  17. checkpoint-6877/vocab.json +0 -0
checkpoint-6877/README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.5.0
checkpoint-6877/adapter_config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "Qwen/Qwen1.5-4B",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 16,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "k_proj",
19
+ "v_proj",
20
+ "down_proj",
21
+ "gate_proj",
22
+ "up_proj"
23
+ ],
24
+ "task_type": "CAUSAL_LM"
25
+ }
checkpoint-6877/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af0033510e034c47fa39f2e59a37f6d764e0c60758209c2cf1edf78e43a3c5bf
3
+ size 112130416
checkpoint-6877/added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "<|endoftext|>": 151643,
3
+ "<|im_end|>": 151645,
4
+ "<|im_start|>": 151644
5
+ }
checkpoint-6877/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-6877/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:676d94af7c7bd26c4084f6586957996805f797c804ffbcdf0f5cc73bf1b5abf0
3
+ size 224537202
checkpoint-6877/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9b6721b40312270df8b6bd7b2b36022e55098cf838aad35826be256a3084fed
3
+ size 15024
checkpoint-6877/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1a41e64a1e2d0c6e56904c40a32958364ff99503ac13cc25672b0aab48b73ba
3
+ size 15024
checkpoint-6877/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a539b02b885bbb9bc80ff94139d636fb7ed593e7de6e821aa061fd25c6e66f71
3
+ size 15024
checkpoint-6877/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5173d25211bca94f9155be0759559e1119622385000e4b63806f064fa84e06b0
3
+ size 15024
checkpoint-6877/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5084d89c54d0af4c796500393e0b82a198e71a7fc35952b5da45f192e2dadfe
3
+ size 1064
checkpoint-6877/special_tokens_map.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>"
5
+ ],
6
+ "eos_token": {
7
+ "content": "<|endoftext|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false
12
+ },
13
+ "pad_token": "<|endoftext|>"
14
+ }
checkpoint-6877/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-6877/tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "151643": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "151644": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "151645": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ }
28
+ },
29
+ "additional_special_tokens": [
30
+ "<|im_start|>",
31
+ "<|im_end|>"
32
+ ],
33
+ "bos_token": null,
34
+ "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
35
+ "clean_up_tokenization_spaces": false,
36
+ "eos_token": "<|endoftext|>",
37
+ "errors": "replace",
38
+ "model_max_length": 32768,
39
+ "pad_token": "<|endoftext|>",
40
+ "split_special_tokens": false,
41
+ "tokenizer_class": "Qwen2Tokenizer",
42
+ "unk_token": null
43
+ }
checkpoint-6877/trainer_state.json ADDED
@@ -0,0 +1,704 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 13.0,
5
+ "eval_steps": 500,
6
+ "global_step": 6877,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.1890359168241966,
13
+ "grad_norm": 0.5558584332466125,
14
+ "learning_rate": 3e-05,
15
+ "loss": 2.1103,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 0.3780718336483932,
20
+ "grad_norm": 0.4733140468597412,
21
+ "learning_rate": 3e-05,
22
+ "loss": 1.931,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 0.5671077504725898,
27
+ "grad_norm": 0.48580050468444824,
28
+ "learning_rate": 3e-05,
29
+ "loss": 1.8948,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 0.7561436672967864,
34
+ "grad_norm": 0.40655234456062317,
35
+ "learning_rate": 3e-05,
36
+ "loss": 1.8988,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 0.945179584120983,
41
+ "grad_norm": 0.6303955316543579,
42
+ "learning_rate": 3e-05,
43
+ "loss": 1.8745,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 1.0,
48
+ "eval_accuracy": 0.5701538461538461,
49
+ "eval_loss": 1.6455246210098267,
50
+ "eval_runtime": 5.0805,
51
+ "eval_samples_per_second": 98.415,
52
+ "eval_steps_per_second": 12.4,
53
+ "step": 529
54
+ },
55
+ {
56
+ "epoch": 1.0,
57
+ "eval_exact_match": 15.0,
58
+ "eval_f1": 23.927619047619068,
59
+ "step": 529
60
+ },
61
+ {
62
+ "epoch": 1.1342155009451795,
63
+ "grad_norm": 0.4278818666934967,
64
+ "learning_rate": 3e-05,
65
+ "loss": 1.8798,
66
+ "step": 600
67
+ },
68
+ {
69
+ "epoch": 1.3232514177693762,
70
+ "grad_norm": 0.449830025434494,
71
+ "learning_rate": 3e-05,
72
+ "loss": 1.8613,
73
+ "step": 700
74
+ },
75
+ {
76
+ "epoch": 1.5122873345935728,
77
+ "grad_norm": 0.5130725502967834,
78
+ "learning_rate": 3e-05,
79
+ "loss": 1.8709,
80
+ "step": 800
81
+ },
82
+ {
83
+ "epoch": 1.7013232514177694,
84
+ "grad_norm": 0.4889715611934662,
85
+ "learning_rate": 3e-05,
86
+ "loss": 1.8448,
87
+ "step": 900
88
+ },
89
+ {
90
+ "epoch": 1.8903591682419658,
91
+ "grad_norm": 11.186380386352539,
92
+ "learning_rate": 3e-05,
93
+ "loss": 1.8764,
94
+ "step": 1000
95
+ },
96
+ {
97
+ "epoch": 2.0,
98
+ "eval_accuracy": 0.5723589743589743,
99
+ "eval_loss": 1.6143561601638794,
100
+ "eval_runtime": 5.0466,
101
+ "eval_samples_per_second": 99.077,
102
+ "eval_steps_per_second": 12.484,
103
+ "step": 1058
104
+ },
105
+ {
106
+ "epoch": 2.0,
107
+ "eval_exact_match": 15.6,
108
+ "eval_f1": 25.825238095238113,
109
+ "step": 1058
110
+ },
111
+ {
112
+ "epoch": 2.0793950850661624,
113
+ "grad_norm": 0.5414931774139404,
114
+ "learning_rate": 3e-05,
115
+ "loss": 1.8398,
116
+ "step": 1100
117
+ },
118
+ {
119
+ "epoch": 2.268431001890359,
120
+ "grad_norm": 7.441005706787109,
121
+ "learning_rate": 3e-05,
122
+ "loss": 1.8348,
123
+ "step": 1200
124
+ },
125
+ {
126
+ "epoch": 2.4574669187145557,
127
+ "grad_norm": 24.312145233154297,
128
+ "learning_rate": 3e-05,
129
+ "loss": 1.8421,
130
+ "step": 1300
131
+ },
132
+ {
133
+ "epoch": 2.6465028355387523,
134
+ "grad_norm": 0.7407662868499756,
135
+ "learning_rate": 3e-05,
136
+ "loss": 1.8292,
137
+ "step": 1400
138
+ },
139
+ {
140
+ "epoch": 2.835538752362949,
141
+ "grad_norm": 0.9653642177581787,
142
+ "learning_rate": 3e-05,
143
+ "loss": 1.828,
144
+ "step": 1500
145
+ },
146
+ {
147
+ "epoch": 3.0,
148
+ "eval_accuracy": 0.5734358974358974,
149
+ "eval_loss": 1.617864727973938,
150
+ "eval_runtime": 5.2153,
151
+ "eval_samples_per_second": 95.872,
152
+ "eval_steps_per_second": 12.08,
153
+ "step": 1587
154
+ },
155
+ {
156
+ "epoch": 3.0,
157
+ "eval_exact_match": 16.2,
158
+ "eval_f1": 25.811428571428586,
159
+ "step": 1587
160
+ },
161
+ {
162
+ "epoch": 3.0245746691871456,
163
+ "grad_norm": 450.6055603027344,
164
+ "learning_rate": 3e-05,
165
+ "loss": 1.8204,
166
+ "step": 1600
167
+ },
168
+ {
169
+ "epoch": 3.213610586011342,
170
+ "grad_norm": 2.1127028465270996,
171
+ "learning_rate": 3e-05,
172
+ "loss": 1.8361,
173
+ "step": 1700
174
+ },
175
+ {
176
+ "epoch": 3.402646502835539,
177
+ "grad_norm": 564.9320068359375,
178
+ "learning_rate": 3e-05,
179
+ "loss": 1.8338,
180
+ "step": 1800
181
+ },
182
+ {
183
+ "epoch": 3.5916824196597354,
184
+ "grad_norm": 52.5078010559082,
185
+ "learning_rate": 3e-05,
186
+ "loss": 1.8075,
187
+ "step": 1900
188
+ },
189
+ {
190
+ "epoch": 3.780718336483932,
191
+ "grad_norm": 27.87557601928711,
192
+ "learning_rate": 3e-05,
193
+ "loss": 1.8175,
194
+ "step": 2000
195
+ },
196
+ {
197
+ "epoch": 3.9697542533081287,
198
+ "grad_norm": 4.810656547546387,
199
+ "learning_rate": 3e-05,
200
+ "loss": 1.8218,
201
+ "step": 2100
202
+ },
203
+ {
204
+ "epoch": 4.0,
205
+ "eval_accuracy": 0.5728205128205128,
206
+ "eval_loss": 1.6224406957626343,
207
+ "eval_runtime": 5.0588,
208
+ "eval_samples_per_second": 98.838,
209
+ "eval_steps_per_second": 12.454,
210
+ "step": 2116
211
+ },
212
+ {
213
+ "epoch": 4.0,
214
+ "eval_exact_match": 17.4,
215
+ "eval_f1": 27.194285714285726,
216
+ "step": 2116
217
+ },
218
+ {
219
+ "epoch": 4.158790170132325,
220
+ "grad_norm": 13.004076957702637,
221
+ "learning_rate": 3e-05,
222
+ "loss": 1.8089,
223
+ "step": 2200
224
+ },
225
+ {
226
+ "epoch": 4.3478260869565215,
227
+ "grad_norm": 2.23368501663208,
228
+ "learning_rate": 3e-05,
229
+ "loss": 1.8292,
230
+ "step": 2300
231
+ },
232
+ {
233
+ "epoch": 4.536862003780718,
234
+ "grad_norm": 0.8635586500167847,
235
+ "learning_rate": 3e-05,
236
+ "loss": 1.7992,
237
+ "step": 2400
238
+ },
239
+ {
240
+ "epoch": 4.725897920604915,
241
+ "grad_norm": 6.5326738357543945,
242
+ "learning_rate": 3e-05,
243
+ "loss": 1.7955,
244
+ "step": 2500
245
+ },
246
+ {
247
+ "epoch": 4.914933837429111,
248
+ "grad_norm": 463.4647521972656,
249
+ "learning_rate": 3e-05,
250
+ "loss": 1.7953,
251
+ "step": 2600
252
+ },
253
+ {
254
+ "epoch": 5.0,
255
+ "eval_accuracy": 0.5740512820512821,
256
+ "eval_loss": 1.6217337846755981,
257
+ "eval_runtime": 5.1898,
258
+ "eval_samples_per_second": 96.342,
259
+ "eval_steps_per_second": 12.139,
260
+ "step": 2645
261
+ },
262
+ {
263
+ "epoch": 5.0,
264
+ "eval_exact_match": 17.6,
265
+ "eval_f1": 26.161428571428587,
266
+ "step": 2645
267
+ },
268
+ {
269
+ "epoch": 5.103969754253308,
270
+ "grad_norm": 203.82815551757812,
271
+ "learning_rate": 3e-05,
272
+ "loss": 1.7845,
273
+ "step": 2700
274
+ },
275
+ {
276
+ "epoch": 5.293005671077505,
277
+ "grad_norm": 660.8076782226562,
278
+ "learning_rate": 3e-05,
279
+ "loss": 1.7909,
280
+ "step": 2800
281
+ },
282
+ {
283
+ "epoch": 5.482041587901701,
284
+ "grad_norm": 5727.51806640625,
285
+ "learning_rate": 3e-05,
286
+ "loss": 1.7909,
287
+ "step": 2900
288
+ },
289
+ {
290
+ "epoch": 5.671077504725898,
291
+ "grad_norm": 4880698.0,
292
+ "learning_rate": 3e-05,
293
+ "loss": 3.5781,
294
+ "step": 3000
295
+ },
296
+ {
297
+ "epoch": 5.8601134215500945,
298
+ "grad_norm": 1192.6517333984375,
299
+ "learning_rate": 3e-05,
300
+ "loss": 3.8703,
301
+ "step": 3100
302
+ },
303
+ {
304
+ "epoch": 6.0,
305
+ "eval_accuracy": 0.5433846153846154,
306
+ "eval_loss": 1.886108636856079,
307
+ "eval_runtime": 5.5162,
308
+ "eval_samples_per_second": 90.643,
309
+ "eval_steps_per_second": 11.421,
310
+ "step": 3174
311
+ },
312
+ {
313
+ "epoch": 6.0,
314
+ "eval_exact_match": 11.4,
315
+ "eval_f1": 20.358571428571437,
316
+ "step": 3174
317
+ },
318
+ {
319
+ "epoch": 6.049149338374291,
320
+ "grad_norm": 532.3778686523438,
321
+ "learning_rate": 3e-05,
322
+ "loss": 2.1701,
323
+ "step": 3200
324
+ },
325
+ {
326
+ "epoch": 6.238185255198488,
327
+ "grad_norm": 565.6697998046875,
328
+ "learning_rate": 3e-05,
329
+ "loss": 1.8533,
330
+ "step": 3300
331
+ },
332
+ {
333
+ "epoch": 6.427221172022684,
334
+ "grad_norm": 25206.482421875,
335
+ "learning_rate": 3e-05,
336
+ "loss": 2.623,
337
+ "step": 3400
338
+ },
339
+ {
340
+ "epoch": 6.616257088846881,
341
+ "grad_norm": 71.88556671142578,
342
+ "learning_rate": 3e-05,
343
+ "loss": 3.1573,
344
+ "step": 3500
345
+ },
346
+ {
347
+ "epoch": 6.805293005671078,
348
+ "grad_norm": 269563.15625,
349
+ "learning_rate": 3e-05,
350
+ "loss": 2.2696,
351
+ "step": 3600
352
+ },
353
+ {
354
+ "epoch": 6.994328922495274,
355
+ "grad_norm": 11768.7880859375,
356
+ "learning_rate": 3e-05,
357
+ "loss": 3.1506,
358
+ "step": 3700
359
+ },
360
+ {
361
+ "epoch": 7.0,
362
+ "eval_accuracy": 0.2834871794871795,
363
+ "eval_loss": 8.311681747436523,
364
+ "eval_runtime": 5.3223,
365
+ "eval_samples_per_second": 93.944,
366
+ "eval_steps_per_second": 11.837,
367
+ "step": 3703
368
+ },
369
+ {
370
+ "epoch": 7.0,
371
+ "eval_exact_match": 0.0,
372
+ "eval_f1": 0.027458851580988985,
373
+ "step": 3703
374
+ },
375
+ {
376
+ "epoch": 7.183364839319471,
377
+ "grad_norm": 6220.75244140625,
378
+ "learning_rate": 3e-05,
379
+ "loss": 4.273,
380
+ "step": 3800
381
+ },
382
+ {
383
+ "epoch": 7.3724007561436675,
384
+ "grad_norm": 3387.953857421875,
385
+ "learning_rate": 3e-05,
386
+ "loss": 7.7199,
387
+ "step": 3900
388
+ },
389
+ {
390
+ "epoch": 7.561436672967864,
391
+ "grad_norm": 372.8681640625,
392
+ "learning_rate": 3e-05,
393
+ "loss": 7.7045,
394
+ "step": 4000
395
+ },
396
+ {
397
+ "epoch": 7.750472589792061,
398
+ "grad_norm": 159.64283752441406,
399
+ "learning_rate": 3e-05,
400
+ "loss": 6.8901,
401
+ "step": 4100
402
+ },
403
+ {
404
+ "epoch": 7.939508506616257,
405
+ "grad_norm": 751876.125,
406
+ "learning_rate": 3e-05,
407
+ "loss": 6.5238,
408
+ "step": 4200
409
+ },
410
+ {
411
+ "epoch": 8.0,
412
+ "eval_accuracy": 0.22764102564102565,
413
+ "eval_loss": 7.12347936630249,
414
+ "eval_runtime": 5.3818,
415
+ "eval_samples_per_second": 92.905,
416
+ "eval_steps_per_second": 11.706,
417
+ "step": 4232
418
+ },
419
+ {
420
+ "epoch": 8.0,
421
+ "eval_exact_match": 0.0,
422
+ "eval_f1": 0.013333333333333336,
423
+ "step": 4232
424
+ },
425
+ {
426
+ "epoch": 8.128544423440454,
427
+ "grad_norm": 1035664128.0,
428
+ "learning_rate": 3e-05,
429
+ "loss": 5.4157,
430
+ "step": 4300
431
+ },
432
+ {
433
+ "epoch": 8.31758034026465,
434
+ "grad_norm": 11571242.0,
435
+ "learning_rate": 3e-05,
436
+ "loss": 5.8257,
437
+ "step": 4400
438
+ },
439
+ {
440
+ "epoch": 8.506616257088847,
441
+ "grad_norm": 88236256.0,
442
+ "learning_rate": 3e-05,
443
+ "loss": 6.3609,
444
+ "step": 4500
445
+ },
446
+ {
447
+ "epoch": 8.695652173913043,
448
+ "grad_norm": 105426080.0,
449
+ "learning_rate": 3e-05,
450
+ "loss": 6.1824,
451
+ "step": 4600
452
+ },
453
+ {
454
+ "epoch": 8.88468809073724,
455
+ "grad_norm": 51438844.0,
456
+ "learning_rate": 3e-05,
457
+ "loss": 6.1818,
458
+ "step": 4700
459
+ },
460
+ {
461
+ "epoch": 9.0,
462
+ "eval_accuracy": 0.1995897435897436,
463
+ "eval_loss": 11.187646865844727,
464
+ "eval_runtime": 5.154,
465
+ "eval_samples_per_second": 97.013,
466
+ "eval_steps_per_second": 12.224,
467
+ "step": 4761
468
+ },
469
+ {
470
+ "epoch": 9.0,
471
+ "eval_exact_match": 0.0,
472
+ "eval_f1": 0.0,
473
+ "step": 4761
474
+ },
475
+ {
476
+ "epoch": 9.073724007561436,
477
+ "grad_norm": 8742571.0,
478
+ "learning_rate": 3e-05,
479
+ "loss": 7.5029,
480
+ "step": 4800
481
+ },
482
+ {
483
+ "epoch": 9.262759924385634,
484
+ "grad_norm": 4885384192.0,
485
+ "learning_rate": 3e-05,
486
+ "loss": 7.805,
487
+ "step": 4900
488
+ },
489
+ {
490
+ "epoch": 9.45179584120983,
491
+ "grad_norm": 919074304.0,
492
+ "learning_rate": 3e-05,
493
+ "loss": 8.0712,
494
+ "step": 5000
495
+ },
496
+ {
497
+ "epoch": 9.640831758034027,
498
+ "grad_norm": 9706883072.0,
499
+ "learning_rate": 3e-05,
500
+ "loss": 8.4915,
501
+ "step": 5100
502
+ },
503
+ {
504
+ "epoch": 9.829867674858223,
505
+ "grad_norm": 35723780096.0,
506
+ "learning_rate": 3e-05,
507
+ "loss": 8.3286,
508
+ "step": 5200
509
+ },
510
+ {
511
+ "epoch": 10.0,
512
+ "eval_accuracy": 0.1995897435897436,
513
+ "eval_loss": 13.374640464782715,
514
+ "eval_runtime": 5.2089,
515
+ "eval_samples_per_second": 95.99,
516
+ "eval_steps_per_second": 12.095,
517
+ "step": 5290
518
+ },
519
+ {
520
+ "epoch": 10.0,
521
+ "eval_exact_match": 0.0,
522
+ "eval_f1": 0.0,
523
+ "step": 5290
524
+ },
525
+ {
526
+ "epoch": 10.01890359168242,
527
+ "grad_norm": 172018032.0,
528
+ "learning_rate": 3e-05,
529
+ "loss": 8.4403,
530
+ "step": 5300
531
+ },
532
+ {
533
+ "epoch": 10.207939508506616,
534
+ "grad_norm": 329336992.0,
535
+ "learning_rate": 3e-05,
536
+ "loss": 8.3233,
537
+ "step": 5400
538
+ },
539
+ {
540
+ "epoch": 10.396975425330814,
541
+ "grad_norm": 74210560.0,
542
+ "learning_rate": 3e-05,
543
+ "loss": 8.3673,
544
+ "step": 5500
545
+ },
546
+ {
547
+ "epoch": 10.58601134215501,
548
+ "grad_norm": 497968480.0,
549
+ "learning_rate": 3e-05,
550
+ "loss": 9.8311,
551
+ "step": 5600
552
+ },
553
+ {
554
+ "epoch": 10.775047258979207,
555
+ "grad_norm": 34138104.0,
556
+ "learning_rate": 3e-05,
557
+ "loss": 10.8161,
558
+ "step": 5700
559
+ },
560
+ {
561
+ "epoch": 10.964083175803403,
562
+ "grad_norm": 308232992.0,
563
+ "learning_rate": 3e-05,
564
+ "loss": 10.2827,
565
+ "step": 5800
566
+ },
567
+ {
568
+ "epoch": 11.0,
569
+ "eval_accuracy": 0.1995897435897436,
570
+ "eval_loss": 12.878591537475586,
571
+ "eval_runtime": 5.1038,
572
+ "eval_samples_per_second": 97.966,
573
+ "eval_steps_per_second": 12.344,
574
+ "step": 5819
575
+ },
576
+ {
577
+ "epoch": 11.0,
578
+ "eval_exact_match": 0.0,
579
+ "eval_f1": 0.0,
580
+ "step": 5819
581
+ },
582
+ {
583
+ "epoch": 11.1531190926276,
584
+ "grad_norm": 255807824.0,
585
+ "learning_rate": 3e-05,
586
+ "loss": 9.8782,
587
+ "step": 5900
588
+ },
589
+ {
590
+ "epoch": 11.342155009451796,
591
+ "grad_norm": 27976615936.0,
592
+ "learning_rate": 3e-05,
593
+ "loss": 9.879,
594
+ "step": 6000
595
+ },
596
+ {
597
+ "epoch": 11.531190926275993,
598
+ "grad_norm": 13783190732800.0,
599
+ "learning_rate": 3e-05,
600
+ "loss": 9.8617,
601
+ "step": 6100
602
+ },
603
+ {
604
+ "epoch": 11.720226843100189,
605
+ "grad_norm": 3467661017088.0,
606
+ "learning_rate": 3e-05,
607
+ "loss": 10.1028,
608
+ "step": 6200
609
+ },
610
+ {
611
+ "epoch": 11.909262759924385,
612
+ "grad_norm": 333752.3125,
613
+ "learning_rate": 3e-05,
614
+ "loss": 9.4972,
615
+ "step": 6300
616
+ },
617
+ {
618
+ "epoch": 12.0,
619
+ "eval_accuracy": 0.1995897435897436,
620
+ "eval_loss": 13.161654472351074,
621
+ "eval_runtime": 5.0927,
622
+ "eval_samples_per_second": 98.18,
623
+ "eval_steps_per_second": 12.371,
624
+ "step": 6348
625
+ },
626
+ {
627
+ "epoch": 12.0,
628
+ "eval_exact_match": 0.0,
629
+ "eval_f1": 0.0,
630
+ "step": 6348
631
+ },
632
+ {
633
+ "epoch": 12.098298676748582,
634
+ "grad_norm": 6690696704.0,
635
+ "learning_rate": 3e-05,
636
+ "loss": 9.2939,
637
+ "step": 6400
638
+ },
639
+ {
640
+ "epoch": 12.287334593572778,
641
+ "grad_norm": 148906835968.0,
642
+ "learning_rate": 3e-05,
643
+ "loss": 9.3499,
644
+ "step": 6500
645
+ },
646
+ {
647
+ "epoch": 12.476370510396976,
648
+ "grad_norm": 15515338752.0,
649
+ "learning_rate": 3e-05,
650
+ "loss": 9.3496,
651
+ "step": 6600
652
+ },
653
+ {
654
+ "epoch": 12.665406427221171,
655
+ "grad_norm": 76318760960.0,
656
+ "learning_rate": 3e-05,
657
+ "loss": 9.2728,
658
+ "step": 6700
659
+ },
660
+ {
661
+ "epoch": 12.854442344045369,
662
+ "grad_norm": 85637283840.0,
663
+ "learning_rate": 3e-05,
664
+ "loss": 9.4453,
665
+ "step": 6800
666
+ },
667
+ {
668
+ "epoch": 13.0,
669
+ "eval_accuracy": 0.1995897435897436,
670
+ "eval_loss": 12.748088836669922,
671
+ "eval_runtime": 5.1332,
672
+ "eval_samples_per_second": 97.405,
673
+ "eval_steps_per_second": 12.273,
674
+ "step": 6877
675
+ },
676
+ {
677
+ "epoch": 13.0,
678
+ "eval_exact_match": 0.0,
679
+ "eval_f1": 0.0,
680
+ "step": 6877
681
+ }
682
+ ],
683
+ "logging_steps": 100,
684
+ "max_steps": 26450,
685
+ "num_input_tokens_seen": 0,
686
+ "num_train_epochs": 50,
687
+ "save_steps": 500,
688
+ "stateful_callbacks": {
689
+ "TrainerControl": {
690
+ "args": {
691
+ "should_epoch_stop": false,
692
+ "should_evaluate": false,
693
+ "should_log": false,
694
+ "should_save": true,
695
+ "should_training_stop": false
696
+ },
697
+ "attributes": {}
698
+ }
699
+ },
700
+ "total_flos": 5.597990780372255e+17,
701
+ "train_batch_size": 2,
702
+ "trial_name": null,
703
+ "trial_params": null
704
+ }
checkpoint-6877/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3895667ea94410ecd283430024ce6f8557816c9d86efd03783de342ef988d79c
3
+ size 5304
checkpoint-6877/vocab.json ADDED
The diff for this file is too large to render. See raw diff