Training in progress, step 18000, checkpoint
Browse files- checkpoint-18000/added_tokens.json +5 -0
- checkpoint-18000/config.json +46 -0
- checkpoint-18000/merges.txt +0 -0
- checkpoint-18000/optimizer.pt +3 -0
- checkpoint-18000/pytorch_model.bin +3 -0
- checkpoint-18000/rng_state.pth +3 -0
- checkpoint-18000/scheduler.pt +3 -0
- checkpoint-18000/special_tokens_map.json +6 -0
- checkpoint-18000/tokenizer_config.json +7 -0
- checkpoint-18000/trainer_state.json +295 -0
- checkpoint-18000/training_args.bin +3 -0
- checkpoint-18000/vocab.json +0 -0
checkpoint-18000/added_tokens.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<bos>": 40478,
|
3 |
+
"<eos>": 40479,
|
4 |
+
"<pad>": 40480
|
5 |
+
}
|
checkpoint-18000/config.json
ADDED
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai-gpt",
|
3 |
+
"afn": "gelu",
|
4 |
+
"architectures": [
|
5 |
+
"OpenAIGPTForSequenceClassification"
|
6 |
+
],
|
7 |
+
"attn_pdrop": 0.1,
|
8 |
+
"embd_pdrop": 0.1,
|
9 |
+
"id2label": {
|
10 |
+
"0": "entailment",
|
11 |
+
"1": "neutral",
|
12 |
+
"2": "contradiction"
|
13 |
+
},
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"label2id": {
|
16 |
+
"contradiction": 2,
|
17 |
+
"entailment": 0,
|
18 |
+
"neutral": 1
|
19 |
+
},
|
20 |
+
"layer_norm_epsilon": 1e-05,
|
21 |
+
"model_type": "openai-gpt",
|
22 |
+
"n_ctx": 512,
|
23 |
+
"n_embd": 768,
|
24 |
+
"n_head": 12,
|
25 |
+
"n_layer": 6,
|
26 |
+
"n_positions": 512,
|
27 |
+
"n_special": 0,
|
28 |
+
"pad_token_id": 40480,
|
29 |
+
"predict_special_tokens": true,
|
30 |
+
"problem_type": "single_label_classification",
|
31 |
+
"resid_pdrop": 0.1,
|
32 |
+
"summary_activation": null,
|
33 |
+
"summary_first_dropout": 0.1,
|
34 |
+
"summary_proj_to_labels": true,
|
35 |
+
"summary_type": "last",
|
36 |
+
"summary_use_proj": true,
|
37 |
+
"task_specific_params": {
|
38 |
+
"text-generation": {
|
39 |
+
"do_sample": true,
|
40 |
+
"max_length": 50
|
41 |
+
}
|
42 |
+
},
|
43 |
+
"torch_dtype": "float32",
|
44 |
+
"transformers_version": "4.33.0.dev0",
|
45 |
+
"vocab_size": 40481
|
46 |
+
}
|
checkpoint-18000/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-18000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6a21046076bd552b011216c65c103f93af49dfebb65e494dbe999e74e0632a44
|
3 |
+
size 592141189
|
checkpoint-18000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:aaf6d9ffde8a26e4d1723cba1b35f2ed3acfe6def7eb6b2367835ed89f495b1a
|
3 |
+
size 296072722
|
checkpoint-18000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:302d428f83f530c1a3bfad553a802e10c5e07e0b9df64f350ff653f2e39044ad
|
3 |
+
size 14575
|
checkpoint-18000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a107245d51dc33a898f01c0610e66a0355fc320a2f5b4bde20458a153ca4fe28
|
3 |
+
size 627
|
checkpoint-18000/special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<bos>",
|
3 |
+
"eos_token": "<eos>",
|
4 |
+
"pad_token": "<pad>",
|
5 |
+
"unk_token": "<unk>"
|
6 |
+
}
|
checkpoint-18000/tokenizer_config.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"clean_up_tokenization_spaces": true,
|
3 |
+
"model_max_length": 512,
|
4 |
+
"padding_side": "left",
|
5 |
+
"tokenizer_class": "OpenAIGPTTokenizer",
|
6 |
+
"unk_token": "<unk>"
|
7 |
+
}
|
checkpoint-18000/trainer_state.json
ADDED
@@ -0,0 +1,295 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 1.0845381021499634,
|
3 |
+
"best_model_checkpoint": "/kaggle/output/checkpoint-17000",
|
4 |
+
"epoch": 0.7333767926988266,
|
5 |
+
"eval_steps": 1000,
|
6 |
+
"global_step": 18000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0,
|
13 |
+
"learning_rate": 2.7777777777777777e-11,
|
14 |
+
"loss": 1.05,
|
15 |
+
"step": 1
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.04,
|
19 |
+
"learning_rate": 2.7750000000000004e-08,
|
20 |
+
"loss": 1.1403,
|
21 |
+
"step": 1000
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.04,
|
25 |
+
"eval_accuracy": 0.33712574850299404,
|
26 |
+
"eval_loss": 1.1030837297439575,
|
27 |
+
"eval_runtime": 29.7882,
|
28 |
+
"eval_samples_per_second": 168.187,
|
29 |
+
"eval_steps_per_second": 21.049,
|
30 |
+
"step": 1000
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.08,
|
34 |
+
"learning_rate": 5.5527777777777784e-08,
|
35 |
+
"loss": 1.1194,
|
36 |
+
"step": 2000
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 0.08,
|
40 |
+
"eval_accuracy": 0.34331337325349304,
|
41 |
+
"eval_loss": 1.1002165079116821,
|
42 |
+
"eval_runtime": 29.8685,
|
43 |
+
"eval_samples_per_second": 167.735,
|
44 |
+
"eval_steps_per_second": 20.992,
|
45 |
+
"step": 2000
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.12,
|
49 |
+
"learning_rate": 8.327777777777778e-08,
|
50 |
+
"loss": 1.1194,
|
51 |
+
"step": 3000
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.12,
|
55 |
+
"eval_accuracy": 0.3471057884231537,
|
56 |
+
"eval_loss": 1.101266622543335,
|
57 |
+
"eval_runtime": 29.709,
|
58 |
+
"eval_samples_per_second": 168.636,
|
59 |
+
"eval_steps_per_second": 21.105,
|
60 |
+
"step": 3000
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 0.16,
|
64 |
+
"learning_rate": 1.1105555555555557e-07,
|
65 |
+
"loss": 1.1166,
|
66 |
+
"step": 4000
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 0.16,
|
70 |
+
"eval_accuracy": 0.34311377245508984,
|
71 |
+
"eval_loss": 1.0984796285629272,
|
72 |
+
"eval_runtime": 29.9401,
|
73 |
+
"eval_samples_per_second": 167.334,
|
74 |
+
"eval_steps_per_second": 20.942,
|
75 |
+
"step": 4000
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.2,
|
79 |
+
"learning_rate": 1.3883333333333335e-07,
|
80 |
+
"loss": 1.1137,
|
81 |
+
"step": 5000
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.2,
|
85 |
+
"eval_accuracy": 0.3403193612774451,
|
86 |
+
"eval_loss": 1.0974279642105103,
|
87 |
+
"eval_runtime": 29.8185,
|
88 |
+
"eval_samples_per_second": 168.016,
|
89 |
+
"eval_steps_per_second": 21.027,
|
90 |
+
"step": 5000
|
91 |
+
},
|
92 |
+
{
|
93 |
+
"epoch": 0.24,
|
94 |
+
"learning_rate": 1.6658333333333335e-07,
|
95 |
+
"loss": 1.1109,
|
96 |
+
"step": 6000
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 0.24,
|
100 |
+
"eval_accuracy": 0.3592814371257485,
|
101 |
+
"eval_loss": 1.0984892845153809,
|
102 |
+
"eval_runtime": 29.893,
|
103 |
+
"eval_samples_per_second": 167.598,
|
104 |
+
"eval_steps_per_second": 20.975,
|
105 |
+
"step": 6000
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.29,
|
109 |
+
"learning_rate": 1.9436111111111112e-07,
|
110 |
+
"loss": 1.1122,
|
111 |
+
"step": 7000
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 0.29,
|
115 |
+
"eval_accuracy": 0.36327345309381237,
|
116 |
+
"eval_loss": 1.0960686206817627,
|
117 |
+
"eval_runtime": 29.8315,
|
118 |
+
"eval_samples_per_second": 167.943,
|
119 |
+
"eval_steps_per_second": 21.018,
|
120 |
+
"step": 7000
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 0.33,
|
124 |
+
"learning_rate": 2.2213888888888891e-07,
|
125 |
+
"loss": 1.1091,
|
126 |
+
"step": 8000
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 0.33,
|
130 |
+
"eval_accuracy": 0.35728542914171657,
|
131 |
+
"eval_loss": 1.0958523750305176,
|
132 |
+
"eval_runtime": 29.9494,
|
133 |
+
"eval_samples_per_second": 167.282,
|
134 |
+
"eval_steps_per_second": 20.935,
|
135 |
+
"step": 8000
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.37,
|
139 |
+
"learning_rate": 2.4988888888888893e-07,
|
140 |
+
"loss": 1.1077,
|
141 |
+
"step": 9000
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 0.37,
|
145 |
+
"eval_accuracy": 0.35528942115768464,
|
146 |
+
"eval_loss": 1.1003851890563965,
|
147 |
+
"eval_runtime": 29.9762,
|
148 |
+
"eval_samples_per_second": 167.133,
|
149 |
+
"eval_steps_per_second": 20.917,
|
150 |
+
"step": 9000
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.41,
|
154 |
+
"learning_rate": 2.776666666666667e-07,
|
155 |
+
"loss": 1.1091,
|
156 |
+
"step": 10000
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.41,
|
160 |
+
"eval_accuracy": 0.3790419161676647,
|
161 |
+
"eval_loss": 1.0951762199401855,
|
162 |
+
"eval_runtime": 29.8593,
|
163 |
+
"eval_samples_per_second": 167.787,
|
164 |
+
"eval_steps_per_second": 20.998,
|
165 |
+
"step": 10000
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 0.45,
|
169 |
+
"learning_rate": 3.0541666666666667e-07,
|
170 |
+
"loss": 1.1047,
|
171 |
+
"step": 11000
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 0.45,
|
175 |
+
"eval_accuracy": 0.3780439121756487,
|
176 |
+
"eval_loss": 1.0928910970687866,
|
177 |
+
"eval_runtime": 29.8768,
|
178 |
+
"eval_samples_per_second": 167.689,
|
179 |
+
"eval_steps_per_second": 20.986,
|
180 |
+
"step": 11000
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 0.49,
|
184 |
+
"learning_rate": 3.3319444444444444e-07,
|
185 |
+
"loss": 1.1053,
|
186 |
+
"step": 12000
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 0.49,
|
190 |
+
"eval_accuracy": 0.38323353293413176,
|
191 |
+
"eval_loss": 1.0910215377807617,
|
192 |
+
"eval_runtime": 29.8703,
|
193 |
+
"eval_samples_per_second": 167.725,
|
194 |
+
"eval_steps_per_second": 20.991,
|
195 |
+
"step": 12000
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 0.53,
|
199 |
+
"learning_rate": 3.6094444444444446e-07,
|
200 |
+
"loss": 1.1035,
|
201 |
+
"step": 13000
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 0.53,
|
205 |
+
"eval_accuracy": 0.34850299401197604,
|
206 |
+
"eval_loss": 1.0930161476135254,
|
207 |
+
"eval_runtime": 29.8865,
|
208 |
+
"eval_samples_per_second": 167.634,
|
209 |
+
"eval_steps_per_second": 20.979,
|
210 |
+
"step": 13000
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 0.57,
|
214 |
+
"learning_rate": 3.8872222222222223e-07,
|
215 |
+
"loss": 1.1002,
|
216 |
+
"step": 14000
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 0.57,
|
220 |
+
"eval_accuracy": 0.3646706586826347,
|
221 |
+
"eval_loss": 1.093497633934021,
|
222 |
+
"eval_runtime": 30.0222,
|
223 |
+
"eval_samples_per_second": 166.877,
|
224 |
+
"eval_steps_per_second": 20.885,
|
225 |
+
"step": 14000
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 0.61,
|
229 |
+
"learning_rate": 4.1647222222222225e-07,
|
230 |
+
"loss": 1.1025,
|
231 |
+
"step": 15000
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 0.61,
|
235 |
+
"eval_accuracy": 0.37924151696606784,
|
236 |
+
"eval_loss": 1.0900779962539673,
|
237 |
+
"eval_runtime": 30.0049,
|
238 |
+
"eval_samples_per_second": 166.973,
|
239 |
+
"eval_steps_per_second": 20.897,
|
240 |
+
"step": 15000
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.65,
|
244 |
+
"learning_rate": 4.4425e-07,
|
245 |
+
"loss": 1.0992,
|
246 |
+
"step": 16000
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 0.65,
|
250 |
+
"eval_accuracy": 0.40119760479041916,
|
251 |
+
"eval_loss": 1.0855979919433594,
|
252 |
+
"eval_runtime": 30.1095,
|
253 |
+
"eval_samples_per_second": 166.393,
|
254 |
+
"eval_steps_per_second": 20.824,
|
255 |
+
"step": 16000
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 0.69,
|
259 |
+
"learning_rate": 4.7200000000000004e-07,
|
260 |
+
"loss": 1.0965,
|
261 |
+
"step": 17000
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.69,
|
265 |
+
"eval_accuracy": 0.4171656686626746,
|
266 |
+
"eval_loss": 1.0845381021499634,
|
267 |
+
"eval_runtime": 30.0586,
|
268 |
+
"eval_samples_per_second": 166.674,
|
269 |
+
"eval_steps_per_second": 20.859,
|
270 |
+
"step": 17000
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 0.73,
|
274 |
+
"learning_rate": 4.997777777777779e-07,
|
275 |
+
"loss": 1.0962,
|
276 |
+
"step": 18000
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 0.73,
|
280 |
+
"eval_accuracy": 0.4087824351297405,
|
281 |
+
"eval_loss": 1.0860552787780762,
|
282 |
+
"eval_runtime": 30.2343,
|
283 |
+
"eval_samples_per_second": 165.706,
|
284 |
+
"eval_steps_per_second": 20.738,
|
285 |
+
"step": 18000
|
286 |
+
}
|
287 |
+
],
|
288 |
+
"logging_steps": 1000,
|
289 |
+
"max_steps": 10000000,
|
290 |
+
"num_train_epochs": 408,
|
291 |
+
"save_steps": 1000,
|
292 |
+
"total_flos": 1.8813705781248e+16,
|
293 |
+
"trial_name": null,
|
294 |
+
"trial_params": null
|
295 |
+
}
|
checkpoint-18000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:40a0dfc08574fef5341c0802f4b4b8be93cd36a17a96b8a7b07271ea753deed4
|
3 |
+
size 4027
|
checkpoint-18000/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|