Hugging Face's logo

tyzhu
/
lmind_nq_train6000_eval6489_v1_doc_qa_v3_3e-4_lora2

Safetensors
Generated from Trainer
Model card Files Community
tyzhu commited on
Commit
4f95d1c
•
1 Parent(s): 3c5fe0c

Training in progress, epoch 16, checkpoint

Browse files
Files changed (17) hide show
  1. checkpoint-8464/README.md +9 -0
  2. checkpoint-8464/adapter_config.json +25 -0
  3. checkpoint-8464/adapter_model.safetensors +3 -0
  4. checkpoint-8464/added_tokens.json +5 -0
  5. checkpoint-8464/merges.txt +0 -0
  6. checkpoint-8464/optimizer.pt +3 -0
  7. checkpoint-8464/rng_state_0.pth +3 -0
  8. checkpoint-8464/rng_state_1.pth +3 -0
  9. checkpoint-8464/rng_state_2.pth +3 -0
  10. checkpoint-8464/rng_state_3.pth +3 -0
  11. checkpoint-8464/scheduler.pt +3 -0
  12. checkpoint-8464/special_tokens_map.json +14 -0
  13. checkpoint-8464/tokenizer.json +0 -0
  14. checkpoint-8464/tokenizer_config.json +43 -0
  15. checkpoint-8464/trainer_state.json +861 -0
  16. checkpoint-8464/training_args.bin +3 -0
  17. checkpoint-8464/vocab.json +0 -0
checkpoint-8464/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-8464/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-8464/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a66895c43a60a7f6ea94723b5284ae989d5065ccf1b4f32134868542f1487d3
3
+ size 112130416
checkpoint-8464/added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "<|endoftext|>": 151643,
3
+ "<|im_end|>": 151645,
4
+ "<|im_start|>": 151644
5
+ }
checkpoint-8464/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-8464/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14a7757c8bdf08751b1952dd4eaafa41f7f2b1d7f6b1a4b234d20db7b7d504a5
3
+ size 224537202
checkpoint-8464/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3123dc0451ab6c691cae418fa8236371e48db942d84d801cb7c55f2d061cd093
3
+ size 15024
checkpoint-8464/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c1bcd2861d010c0f33ff6e1b53607a83e48d2f641fd0807ac966a7087b1d77e
3
+ size 15024
checkpoint-8464/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a84d84ee634c256f0f37197364ff1c6c95a581218e202a9a2c0d65f2d5381539
3
+ size 15024
checkpoint-8464/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6492a35e407dd6bd752ce2e5150de4ac7ef276392cae9b4731b49a5e78929856
3
+ size 15024
checkpoint-8464/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20bf54352a2214c3ce61dca611e43408e0a4f218e556532f3b206236a86e60f6
3
+ size 1064
checkpoint-8464/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-8464/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-8464/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-8464/trainer_state.json ADDED
@@ -0,0 +1,861 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 16.0,
5
+ "eval_steps": 500,
6
+ "global_step": 8464,
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": 0.3393945097923279,
14
+ "learning_rate": 0.0003,
15
+ "loss": 1.951,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 0.3780718336483932,
20
+ "grad_norm": 8017.20361328125,
21
+ "learning_rate": 0.0003,
22
+ "loss": 2.0285,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 0.5671077504725898,
27
+ "grad_norm": 9146.7236328125,
28
+ "learning_rate": 0.0003,
29
+ "loss": 6.5404,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 0.7561436672967864,
34
+ "grad_norm": 574.934326171875,
35
+ "learning_rate": 0.0003,
36
+ "loss": 7.7357,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 0.945179584120983,
41
+ "grad_norm": 25128.55859375,
42
+ "learning_rate": 0.0003,
43
+ "loss": 6.3771,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 1.0,
48
+ "eval_accuracy": 0.24974358974358973,
49
+ "eval_loss": 10.19681453704834,
50
+ "eval_runtime": 5.1408,
51
+ "eval_samples_per_second": 97.26,
52
+ "eval_steps_per_second": 12.255,
53
+ "step": 529
54
+ },
55
+ {
56
+ "epoch": 1.0,
57
+ "eval_exact_match": 0.0,
58
+ "eval_f1": 0.0,
59
+ "step": 529
60
+ },
61
+ {
62
+ "epoch": 1.1342155009451795,
63
+ "grad_norm": 1332.8885498046875,
64
+ "learning_rate": 0.0003,
65
+ "loss": 11.0858,
66
+ "step": 600
67
+ },
68
+ {
69
+ "epoch": 1.3232514177693762,
70
+ "grad_norm": 3666.881591796875,
71
+ "learning_rate": 0.0003,
72
+ "loss": 9.483,
73
+ "step": 700
74
+ },
75
+ {
76
+ "epoch": 1.5122873345935728,
77
+ "grad_norm": 356847.0,
78
+ "learning_rate": 0.0003,
79
+ "loss": 8.069,
80
+ "step": 800
81
+ },
82
+ {
83
+ "epoch": 1.7013232514177694,
84
+ "grad_norm": 1061.877685546875,
85
+ "learning_rate": 0.0003,
86
+ "loss": 8.7493,
87
+ "step": 900
88
+ },
89
+ {
90
+ "epoch": 1.8903591682419658,
91
+ "grad_norm": 46670.3046875,
92
+ "learning_rate": 0.0003,
93
+ "loss": 8.3415,
94
+ "step": 1000
95
+ },
96
+ {
97
+ "epoch": 2.0,
98
+ "eval_accuracy": 0.20507692307692307,
99
+ "eval_loss": 7.504091262817383,
100
+ "eval_runtime": 5.2437,
101
+ "eval_samples_per_second": 95.352,
102
+ "eval_steps_per_second": 12.014,
103
+ "step": 1058
104
+ },
105
+ {
106
+ "epoch": 2.0,
107
+ "eval_exact_match": 0.0,
108
+ "eval_f1": 0.0,
109
+ "step": 1058
110
+ },
111
+ {
112
+ "epoch": 2.0793950850661624,
113
+ "grad_norm": 2.0727381706237793,
114
+ "learning_rate": 0.0003,
115
+ "loss": 8.3511,
116
+ "step": 1100
117
+ },
118
+ {
119
+ "epoch": 2.268431001890359,
120
+ "grad_norm": 1.8185893297195435,
121
+ "learning_rate": 0.0003,
122
+ "loss": 7.3065,
123
+ "step": 1200
124
+ },
125
+ {
126
+ "epoch": 2.4574669187145557,
127
+ "grad_norm": 1620.93701171875,
128
+ "learning_rate": 0.0003,
129
+ "loss": 7.1302,
130
+ "step": 1300
131
+ },
132
+ {
133
+ "epoch": 2.6465028355387523,
134
+ "grad_norm": 1.0691380500793457,
135
+ "learning_rate": 0.0003,
136
+ "loss": 7.4423,
137
+ "step": 1400
138
+ },
139
+ {
140
+ "epoch": 2.835538752362949,
141
+ "grad_norm": 1.0605100393295288,
142
+ "learning_rate": 0.0003,
143
+ "loss": 7.6715,
144
+ "step": 1500
145
+ },
146
+ {
147
+ "epoch": 3.0,
148
+ "eval_accuracy": 0.28507692307692306,
149
+ "eval_loss": 7.084368705749512,
150
+ "eval_runtime": 5.493,
151
+ "eval_samples_per_second": 91.026,
152
+ "eval_steps_per_second": 11.469,
153
+ "step": 1587
154
+ },
155
+ {
156
+ "epoch": 3.0,
157
+ "eval_exact_match": 0.0,
158
+ "eval_f1": 0.08,
159
+ "step": 1587
160
+ },
161
+ {
162
+ "epoch": 3.0245746691871456,
163
+ "grad_norm": 1967676800.0,
164
+ "learning_rate": 0.0003,
165
+ "loss": 7.8617,
166
+ "step": 1600
167
+ },
168
+ {
169
+ "epoch": 3.213610586011342,
170
+ "grad_norm": 83918778793984.0,
171
+ "learning_rate": 0.0003,
172
+ "loss": 8.083,
173
+ "step": 1700
174
+ },
175
+ {
176
+ "epoch": 3.402646502835539,
177
+ "grad_norm": 12349449.0,
178
+ "learning_rate": 0.0003,
179
+ "loss": 8.0382,
180
+ "step": 1800
181
+ },
182
+ {
183
+ "epoch": 3.5916824196597354,
184
+ "grad_norm": 52939337728.0,
185
+ "learning_rate": 0.0003,
186
+ "loss": 7.8761,
187
+ "step": 1900
188
+ },
189
+ {
190
+ "epoch": 3.780718336483932,
191
+ "grad_norm": 1381182.25,
192
+ "learning_rate": 0.0003,
193
+ "loss": 7.4665,
194
+ "step": 2000
195
+ },
196
+ {
197
+ "epoch": 3.9697542533081287,
198
+ "grad_norm": 105.80550384521484,
199
+ "learning_rate": 0.0003,
200
+ "loss": 7.8182,
201
+ "step": 2100
202
+ },
203
+ {
204
+ "epoch": 4.0,
205
+ "eval_accuracy": 0.23933333333333334,
206
+ "eval_loss": 7.05845308303833,
207
+ "eval_runtime": 5.0891,
208
+ "eval_samples_per_second": 98.248,
209
+ "eval_steps_per_second": 12.379,
210
+ "step": 2116
211
+ },
212
+ {
213
+ "epoch": 4.0,
214
+ "eval_exact_match": 0.0,
215
+ "eval_f1": 0.0,
216
+ "step": 2116
217
+ },
218
+ {
219
+ "epoch": 4.158790170132325,
220
+ "grad_norm": 16.110862731933594,
221
+ "learning_rate": 0.0003,
222
+ "loss": 7.9728,
223
+ "step": 2200
224
+ },
225
+ {
226
+ "epoch": 4.3478260869565215,
227
+ "grad_norm": 4.362427234649658,
228
+ "learning_rate": 0.0003,
229
+ "loss": 7.2875,
230
+ "step": 2300
231
+ },
232
+ {
233
+ "epoch": 4.536862003780718,
234
+ "grad_norm": 18039.095703125,
235
+ "learning_rate": 0.0003,
236
+ "loss": 7.3972,
237
+ "step": 2400
238
+ },
239
+ {
240
+ "epoch": 4.725897920604915,
241
+ "grad_norm": 2.3812594051160474e+17,
242
+ "learning_rate": 0.0003,
243
+ "loss": 7.4631,
244
+ "step": 2500
245
+ },
246
+ {
247
+ "epoch": 4.914933837429111,
248
+ "grad_norm": 3.9549436569213867,
249
+ "learning_rate": 0.0003,
250
+ "loss": 7.3322,
251
+ "step": 2600
252
+ },
253
+ {
254
+ "epoch": 5.0,
255
+ "eval_accuracy": 0.22430769230769232,
256
+ "eval_loss": 6.72421932220459,
257
+ "eval_runtime": 5.1689,
258
+ "eval_samples_per_second": 96.731,
259
+ "eval_steps_per_second": 12.188,
260
+ "step": 2645
261
+ },
262
+ {
263
+ "epoch": 5.0,
264
+ "eval_exact_match": 0.0,
265
+ "eval_f1": 0.0,
266
+ "step": 2645
267
+ },
268
+ {
269
+ "epoch": 5.103969754253308,
270
+ "grad_norm": 120114.2421875,
271
+ "learning_rate": 0.0003,
272
+ "loss": 7.2428,
273
+ "step": 2700
274
+ },
275
+ {
276
+ "epoch": 5.293005671077505,
277
+ "grad_norm": 391.66119384765625,
278
+ "learning_rate": 0.0003,
279
+ "loss": 7.2495,
280
+ "step": 2800
281
+ },
282
+ {
283
+ "epoch": 5.482041587901701,
284
+ "grad_norm": 599371136.0,
285
+ "learning_rate": 0.0003,
286
+ "loss": 7.5915,
287
+ "step": 2900
288
+ },
289
+ {
290
+ "epoch": 5.671077504725898,
291
+ "grad_norm": 5585.97314453125,
292
+ "learning_rate": 0.0003,
293
+ "loss": 7.3014,
294
+ "step": 3000
295
+ },
296
+ {
297
+ "epoch": 5.8601134215500945,
298
+ "grad_norm": 268239548448768.0,
299
+ "learning_rate": 0.0003,
300
+ "loss": 7.259,
301
+ "step": 3100
302
+ },
303
+ {
304
+ "epoch": 6.0,
305
+ "eval_accuracy": 0.24984615384615386,
306
+ "eval_loss": 6.531904220581055,
307
+ "eval_runtime": 5.0344,
308
+ "eval_samples_per_second": 99.317,
309
+ "eval_steps_per_second": 12.514,
310
+ "step": 3174
311
+ },
312
+ {
313
+ "epoch": 6.0,
314
+ "eval_exact_match": 0.0,
315
+ "eval_f1": 0.0,
316
+ "step": 3174
317
+ },
318
+ {
319
+ "epoch": 6.049149338374291,
320
+ "grad_norm": 21535.734375,
321
+ "learning_rate": 0.0003,
322
+ "loss": 7.5829,
323
+ "step": 3200
324
+ },
325
+ {
326
+ "epoch": 6.238185255198488,
327
+ "grad_norm": 40871530496.0,
328
+ "learning_rate": 0.0003,
329
+ "loss": 8.6364,
330
+ "step": 3300
331
+ },
332
+ {
333
+ "epoch": 6.427221172022684,
334
+ "grad_norm": 13.242510795593262,
335
+ "learning_rate": 0.0003,
336
+ "loss": 7.6599,
337
+ "step": 3400
338
+ },
339
+ {
340
+ "epoch": 6.616257088846881,
341
+ "grad_norm": 3.0841150283813477,
342
+ "learning_rate": 0.0003,
343
+ "loss": 7.5796,
344
+ "step": 3500
345
+ },
346
+ {
347
+ "epoch": 6.805293005671078,
348
+ "grad_norm": 1.6124571561813354,
349
+ "learning_rate": 0.0003,
350
+ "loss": 7.3499,
351
+ "step": 3600
352
+ },
353
+ {
354
+ "epoch": 6.994328922495274,
355
+ "grad_norm": 1.8358014822006226,
356
+ "learning_rate": 0.0003,
357
+ "loss": 7.2206,
358
+ "step": 3700
359
+ },
360
+ {
361
+ "epoch": 7.0,
362
+ "eval_accuracy": 0.22430769230769232,
363
+ "eval_loss": 6.724715709686279,
364
+ "eval_runtime": 5.0476,
365
+ "eval_samples_per_second": 99.056,
366
+ "eval_steps_per_second": 12.481,
367
+ "step": 3703
368
+ },
369
+ {
370
+ "epoch": 7.0,
371
+ "eval_exact_match": 0.0,
372
+ "eval_f1": 0.0,
373
+ "step": 3703
374
+ },
375
+ {
376
+ "epoch": 7.183364839319471,
377
+ "grad_norm": 5.786651134490967,
378
+ "learning_rate": 0.0003,
379
+ "loss": 7.2316,
380
+ "step": 3800
381
+ },
382
+ {
383
+ "epoch": 7.3724007561436675,
384
+ "grad_norm": 3.7182490825653076,
385
+ "learning_rate": 0.0003,
386
+ "loss": 7.2591,
387
+ "step": 3900
388
+ },
389
+ {
390
+ "epoch": 7.561436672967864,
391
+ "grad_norm": 31.241104125976562,
392
+ "learning_rate": 0.0003,
393
+ "loss": 7.2557,
394
+ "step": 4000
395
+ },
396
+ {
397
+ "epoch": 7.750472589792061,
398
+ "grad_norm": 896688896.0,
399
+ "learning_rate": 0.0003,
400
+ "loss": 8.4436,
401
+ "step": 4100
402
+ },
403
+ {
404
+ "epoch": 7.939508506616257,
405
+ "grad_norm": 1.1150538843415904e+19,
406
+ "learning_rate": 0.0003,
407
+ "loss": 8.7572,
408
+ "step": 4200
409
+ },
410
+ {
411
+ "epoch": 8.0,
412
+ "eval_accuracy": 0.26184615384615384,
413
+ "eval_loss": 6.479637622833252,
414
+ "eval_runtime": 5.1473,
415
+ "eval_samples_per_second": 97.138,
416
+ "eval_steps_per_second": 12.239,
417
+ "step": 4232
418
+ },
419
+ {
420
+ "epoch": 8.0,
421
+ "eval_exact_match": 0.0,
422
+ "eval_f1": 0.0,
423
+ "step": 4232
424
+ },
425
+ {
426
+ "epoch": 8.128544423440454,
427
+ "grad_norm": 7.2513813972473145,
428
+ "learning_rate": 0.0003,
429
+ "loss": 8.1381,
430
+ "step": 4300
431
+ },
432
+ {
433
+ "epoch": 8.31758034026465,
434
+ "grad_norm": 14.863628387451172,
435
+ "learning_rate": 0.0003,
436
+ "loss": 7.3335,
437
+ "step": 4400
438
+ },
439
+ {
440
+ "epoch": 8.506616257088847,
441
+ "grad_norm": 5.358551979064941,
442
+ "learning_rate": 0.0003,
443
+ "loss": 8.3921,
444
+ "step": 4500
445
+ },
446
+ {
447
+ "epoch": 8.695652173913043,
448
+ "grad_norm": 113.71702575683594,
449
+ "learning_rate": 0.0003,
450
+ "loss": 7.221,
451
+ "step": 4600
452
+ },
453
+ {
454
+ "epoch": 8.88468809073724,
455
+ "grad_norm": 2.3997488021850586,
456
+ "learning_rate": 0.0003,
457
+ "loss": 8.4835,
458
+ "step": 4700
459
+ },
460
+ {
461
+ "epoch": 9.0,
462
+ "eval_accuracy": 0.26184615384615384,
463
+ "eval_loss": 6.614655017852783,
464
+ "eval_runtime": 5.026,
465
+ "eval_samples_per_second": 99.483,
466
+ "eval_steps_per_second": 12.535,
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": 7.822358131408691,
478
+ "learning_rate": 0.0003,
479
+ "loss": 7.2836,
480
+ "step": 4800
481
+ },
482
+ {
483
+ "epoch": 9.262759924385634,
484
+ "grad_norm": 89279359352832.0,
485
+ "learning_rate": 0.0003,
486
+ "loss": 7.2712,
487
+ "step": 4900
488
+ },
489
+ {
490
+ "epoch": 9.45179584120983,
491
+ "grad_norm": 656.6845703125,
492
+ "learning_rate": 0.0003,
493
+ "loss": 7.2623,
494
+ "step": 5000
495
+ },
496
+ {
497
+ "epoch": 9.640831758034027,
498
+ "grad_norm": 13841700.0,
499
+ "learning_rate": 0.0003,
500
+ "loss": 7.4981,
501
+ "step": 5100
502
+ },
503
+ {
504
+ "epoch": 9.829867674858223,
505
+ "grad_norm": 768177.9375,
506
+ "learning_rate": 0.0003,
507
+ "loss": 7.6044,
508
+ "step": 5200
509
+ },
510
+ {
511
+ "epoch": 10.0,
512
+ "eval_accuracy": 0.2508717948717949,
513
+ "eval_loss": 8.167869567871094,
514
+ "eval_runtime": 5.1342,
515
+ "eval_samples_per_second": 97.387,
516
+ "eval_steps_per_second": 12.271,
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": 389293.03125,
528
+ "learning_rate": 0.0003,
529
+ "loss": 7.9142,
530
+ "step": 5300
531
+ },
532
+ {
533
+ "epoch": 10.207939508506616,
534
+ "grad_norm": 3129429.25,
535
+ "learning_rate": 0.0003,
536
+ "loss": 8.998,
537
+ "step": 5400
538
+ },
539
+ {
540
+ "epoch": 10.396975425330814,
541
+ "grad_norm": 6832450.5,
542
+ "learning_rate": 0.0003,
543
+ "loss": 8.9548,
544
+ "step": 5500
545
+ },
546
+ {
547
+ "epoch": 10.58601134215501,
548
+ "grad_norm": 24994486.0,
549
+ "learning_rate": 0.0003,
550
+ "loss": 9.3085,
551
+ "step": 5600
552
+ },
553
+ {
554
+ "epoch": 10.775047258979207,
555
+ "grad_norm": 10634.3076171875,
556
+ "learning_rate": 0.0003,
557
+ "loss": 9.3052,
558
+ "step": 5700
559
+ },
560
+ {
561
+ "epoch": 10.964083175803403,
562
+ "grad_norm": 37.100101470947266,
563
+ "learning_rate": 0.0003,
564
+ "loss": 10.0839,
565
+ "step": 5800
566
+ },
567
+ {
568
+ "epoch": 11.0,
569
+ "eval_accuracy": 0.2362051282051282,
570
+ "eval_loss": 6.820600509643555,
571
+ "eval_runtime": 5.0262,
572
+ "eval_samples_per_second": 99.478,
573
+ "eval_steps_per_second": 12.534,
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": 319.2048645019531,
585
+ "learning_rate": 0.0003,
586
+ "loss": 7.4258,
587
+ "step": 5900
588
+ },
589
+ {
590
+ "epoch": 11.342155009451796,
591
+ "grad_norm": 72.28472900390625,
592
+ "learning_rate": 0.0003,
593
+ "loss": 7.2609,
594
+ "step": 6000
595
+ },
596
+ {
597
+ "epoch": 11.531190926275993,
598
+ "grad_norm": 11.851283073425293,
599
+ "learning_rate": 0.0003,
600
+ "loss": 7.5323,
601
+ "step": 6100
602
+ },
603
+ {
604
+ "epoch": 11.720226843100189,
605
+ "grad_norm": 6.574771404266357,
606
+ "learning_rate": 0.0003,
607
+ "loss": 7.3028,
608
+ "step": 6200
609
+ },
610
+ {
611
+ "epoch": 11.909262759924385,
612
+ "grad_norm": 4.578986644744873,
613
+ "learning_rate": 0.0003,
614
+ "loss": 7.2957,
615
+ "step": 6300
616
+ },
617
+ {
618
+ "epoch": 12.0,
619
+ "eval_accuracy": 0.26184615384615384,
620
+ "eval_loss": 6.52316427230835,
621
+ "eval_runtime": 5.0329,
622
+ "eval_samples_per_second": 99.346,
623
+ "eval_steps_per_second": 12.518,
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": 21.886821746826172,
635
+ "learning_rate": 0.0003,
636
+ "loss": 7.189,
637
+ "step": 6400
638
+ },
639
+ {
640
+ "epoch": 12.287334593572778,
641
+ "grad_norm": 102.8978500366211,
642
+ "learning_rate": 0.0003,
643
+ "loss": 7.2076,
644
+ "step": 6500
645
+ },
646
+ {
647
+ "epoch": 12.476370510396976,
648
+ "grad_norm": 4.47935676574707,
649
+ "learning_rate": 0.0003,
650
+ "loss": 7.2369,
651
+ "step": 6600
652
+ },
653
+ {
654
+ "epoch": 12.665406427221171,
655
+ "grad_norm": 259.2627868652344,
656
+ "learning_rate": 0.0003,
657
+ "loss": 7.3119,
658
+ "step": 6700
659
+ },
660
+ {
661
+ "epoch": 12.854442344045369,
662
+ "grad_norm": 473.829345703125,
663
+ "learning_rate": 0.0003,
664
+ "loss": 7.3484,
665
+ "step": 6800
666
+ },
667
+ {
668
+ "epoch": 13.0,
669
+ "eval_accuracy": 0.26184615384615384,
670
+ "eval_loss": 6.655651569366455,
671
+ "eval_runtime": 5.7482,
672
+ "eval_samples_per_second": 86.984,
673
+ "eval_steps_per_second": 10.96,
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
+ "epoch": 13.043478260869565,
684
+ "grad_norm": 150.06381225585938,
685
+ "learning_rate": 0.0003,
686
+ "loss": 7.2438,
687
+ "step": 6900
688
+ },
689
+ {
690
+ "epoch": 13.232514177693762,
691
+ "grad_norm": 1546.5465087890625,
692
+ "learning_rate": 0.0003,
693
+ "loss": 7.2503,
694
+ "step": 7000
695
+ },
696
+ {
697
+ "epoch": 13.421550094517958,
698
+ "grad_norm": 18108432.0,
699
+ "learning_rate": 0.0003,
700
+ "loss": 7.5797,
701
+ "step": 7100
702
+ },
703
+ {
704
+ "epoch": 13.610586011342155,
705
+ "grad_norm": 379725545472.0,
706
+ "learning_rate": 0.0003,
707
+ "loss": 8.6953,
708
+ "step": 7200
709
+ },
710
+ {
711
+ "epoch": 13.799621928166351,
712
+ "grad_norm": 14.351493835449219,
713
+ "learning_rate": 0.0003,
714
+ "loss": 7.5947,
715
+ "step": 7300
716
+ },
717
+ {
718
+ "epoch": 13.988657844990549,
719
+ "grad_norm": 3.4325509071350098,
720
+ "learning_rate": 0.0003,
721
+ "loss": 8.2657,
722
+ "step": 7400
723
+ },
724
+ {
725
+ "epoch": 14.0,
726
+ "eval_accuracy": 0.2362051282051282,
727
+ "eval_loss": 6.780774116516113,
728
+ "eval_runtime": 5.7058,
729
+ "eval_samples_per_second": 87.629,
730
+ "eval_steps_per_second": 11.041,
731
+ "step": 7406
732
+ },
733
+ {
734
+ "epoch": 14.0,
735
+ "eval_exact_match": 0.0,
736
+ "eval_f1": 0.0,
737
+ "step": 7406
738
+ },
739
+ {
740
+ "epoch": 14.177693761814744,
741
+ "grad_norm": 3408474406912.0,
742
+ "learning_rate": 0.0003,
743
+ "loss": 7.5487,
744
+ "step": 7500
745
+ },
746
+ {
747
+ "epoch": 14.366729678638942,
748
+ "grad_norm": 766890.1875,
749
+ "learning_rate": 0.0003,
750
+ "loss": 7.4765,
751
+ "step": 7600
752
+ },
753
+ {
754
+ "epoch": 14.555765595463138,
755
+ "grad_norm": 4749101.0,
756
+ "learning_rate": 0.0003,
757
+ "loss": 7.4927,
758
+ "step": 7700
759
+ },
760
+ {
761
+ "epoch": 14.744801512287335,
762
+ "grad_norm": 21509612.0,
763
+ "learning_rate": 0.0003,
764
+ "loss": 7.4864,
765
+ "step": 7800
766
+ },
767
+ {
768
+ "epoch": 14.93383742911153,
769
+ "grad_norm": 1.0536716136244838e+18,
770
+ "learning_rate": 0.0003,
771
+ "loss": 7.4391,
772
+ "step": 7900
773
+ },
774
+ {
775
+ "epoch": 15.0,
776
+ "eval_accuracy": 0.1995897435897436,
777
+ "eval_loss": NaN,
778
+ "eval_runtime": 5.788,
779
+ "eval_samples_per_second": 86.386,
780
+ "eval_steps_per_second": 10.885,
781
+ "step": 7935
782
+ },
783
+ {
784
+ "epoch": 15.0,
785
+ "eval_exact_match": 0.0,
786
+ "eval_f1": 0.0,
787
+ "step": 7935
788
+ },
789
+ {
790
+ "epoch": 15.122873345935728,
791
+ "grad_norm": NaN,
792
+ "learning_rate": 0.0003,
793
+ "loss": 602.3478,
794
+ "step": 8000
795
+ },
796
+ {
797
+ "epoch": 15.311909262759924,
798
+ "grad_norm": NaN,
799
+ "learning_rate": 0.0003,
800
+ "loss": 0.0,
801
+ "step": 8100
802
+ },
803
+ {
804
+ "epoch": 15.500945179584122,
805
+ "grad_norm": NaN,
806
+ "learning_rate": 0.0003,
807
+ "loss": 0.0,
808
+ "step": 8200
809
+ },
810
+ {
811
+ "epoch": 15.689981096408317,
812
+ "grad_norm": NaN,
813
+ "learning_rate": 0.0003,
814
+ "loss": 0.0,
815
+ "step": 8300
816
+ },
817
+ {
818
+ "epoch": 15.879017013232515,
819
+ "grad_norm": NaN,
820
+ "learning_rate": 0.0003,
821
+ "loss": 0.0,
822
+ "step": 8400
823
+ },
824
+ {
825
+ "epoch": 16.0,
826
+ "eval_accuracy": 0.1995897435897436,
827
+ "eval_loss": NaN,
828
+ "eval_runtime": 5.1552,
829
+ "eval_samples_per_second": 96.99,
830
+ "eval_steps_per_second": 12.221,
831
+ "step": 8464
832
+ },
833
+ {
834
+ "epoch": 16.0,
835
+ "eval_exact_match": 0.0,
836
+ "eval_f1": 0.0,
837
+ "step": 8464
838
+ }
839
+ ],
840
+ "logging_steps": 100,
841
+ "max_steps": 26450,
842
+ "num_input_tokens_seen": 0,
843
+ "num_train_epochs": 50,
844
+ "save_steps": 500,
845
+ "stateful_callbacks": {
846
+ "TrainerControl": {
847
+ "args": {
848
+ "should_epoch_stop": false,
849
+ "should_evaluate": false,
850
+ "should_log": false,
851
+ "should_save": true,
852
+ "should_training_stop": false
853
+ },
854
+ "attributes": {}
855
+ }
856
+ },
857
+ "total_flos": 6.888100743610368e+17,
858
+ "train_batch_size": 2,
859
+ "trial_name": null,
860
+ "trial_params": null
861
+ }
checkpoint-8464/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:186f3ae462881ca0bb3a6b5295a97a245e0a78172be9c14948a2d07a12aa4241
3
+ size 5304
checkpoint-8464/vocab.json ADDED
The diff for this file is too large to render. See raw diff