Commit
•
aac1432
1
Parent(s):
97ba59d
Model save
Browse files- README.md +61 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- model.safetensors +1 -1
- train_results.json +9 -0
- trainer_state.json +357 -0
README.md
ADDED
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
base_model: gpt2
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
+
datasets:
|
9 |
+
- generator
|
10 |
+
model-index:
|
11 |
+
- name: gpt2
|
12 |
+
results: []
|
13 |
+
---
|
14 |
+
|
15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
16 |
+
should probably proofread and complete it, then remove this comment. -->
|
17 |
+
|
18 |
+
# gpt2
|
19 |
+
|
20 |
+
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
|
21 |
+
|
22 |
+
## Model description
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Intended uses & limitations
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training and evaluation data
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training procedure
|
35 |
+
|
36 |
+
### Training hyperparameters
|
37 |
+
|
38 |
+
The following hyperparameters were used during training:
|
39 |
+
- learning_rate: 0.0002
|
40 |
+
- train_batch_size: 16
|
41 |
+
- eval_batch_size: 8
|
42 |
+
- seed: 42
|
43 |
+
- distributed_type: multi-GPU
|
44 |
+
- num_devices: 4
|
45 |
+
- total_train_batch_size: 64
|
46 |
+
- total_eval_batch_size: 32
|
47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
48 |
+
- lr_scheduler_type: cosine
|
49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
50 |
+
- num_epochs: 1
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
|
55 |
+
|
56 |
+
### Framework versions
|
57 |
+
|
58 |
+
- Transformers 4.41.2
|
59 |
+
- Pytorch 2.3.1+cu121
|
60 |
+
- Datasets 2.19.2
|
61 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 7357983621120000.0,
|
4 |
+
"train_loss": 1.3856282998215068,
|
5 |
+
"train_runtime": 92.8881,
|
6 |
+
"train_samples": 10000,
|
7 |
+
"train_samples_per_second": 151.214,
|
8 |
+
"train_steps_per_second": 2.368
|
9 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 50256,
|
4 |
+
"eos_token_id": 50256,
|
5 |
+
"transformers_version": "4.41.2"
|
6 |
+
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 248894656
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:02a7f1edfaf396ec82f65b48f1236d1895da1986033aa63a37a813f4bdd3f4e5
|
3 |
size 248894656
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"total_flos": 7357983621120000.0,
|
4 |
+
"train_loss": 1.3856282998215068,
|
5 |
+
"train_runtime": 92.8881,
|
6 |
+
"train_samples": 10000,
|
7 |
+
"train_samples_per_second": 151.214,
|
8 |
+
"train_steps_per_second": 2.368
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,357 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 220,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.004545454545454545,
|
13 |
+
"grad_norm": 3.78125,
|
14 |
+
"learning_rate": 9.090909090909091e-06,
|
15 |
+
"loss": 1.8611,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.022727272727272728,
|
20 |
+
"grad_norm": 3.21875,
|
21 |
+
"learning_rate": 4.545454545454546e-05,
|
22 |
+
"loss": 1.8713,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.045454545454545456,
|
27 |
+
"grad_norm": 1.96875,
|
28 |
+
"learning_rate": 9.090909090909092e-05,
|
29 |
+
"loss": 1.8057,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.06818181818181818,
|
34 |
+
"grad_norm": 1.2890625,
|
35 |
+
"learning_rate": 0.00013636363636363637,
|
36 |
+
"loss": 1.7424,
|
37 |
+
"step": 15
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.09090909090909091,
|
41 |
+
"grad_norm": 1.796875,
|
42 |
+
"learning_rate": 0.00018181818181818183,
|
43 |
+
"loss": 1.6841,
|
44 |
+
"step": 20
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.11363636363636363,
|
48 |
+
"grad_norm": 1.9375,
|
49 |
+
"learning_rate": 0.0001998867339183008,
|
50 |
+
"loss": 1.611,
|
51 |
+
"step": 25
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.13636363636363635,
|
55 |
+
"grad_norm": 1.84375,
|
56 |
+
"learning_rate": 0.00019919548128307954,
|
57 |
+
"loss": 1.5838,
|
58 |
+
"step": 30
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.1590909090909091,
|
62 |
+
"grad_norm": 1.46875,
|
63 |
+
"learning_rate": 0.00019788024462147788,
|
64 |
+
"loss": 1.5374,
|
65 |
+
"step": 35
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.18181818181818182,
|
69 |
+
"grad_norm": 0.76953125,
|
70 |
+
"learning_rate": 0.00019594929736144976,
|
71 |
+
"loss": 1.5049,
|
72 |
+
"step": 40
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.20454545454545456,
|
76 |
+
"grad_norm": 0.9609375,
|
77 |
+
"learning_rate": 0.00019341478602651069,
|
78 |
+
"loss": 1.4867,
|
79 |
+
"step": 45
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.22727272727272727,
|
83 |
+
"grad_norm": 0.83203125,
|
84 |
+
"learning_rate": 0.00019029265382866214,
|
85 |
+
"loss": 1.4711,
|
86 |
+
"step": 50
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.25,
|
90 |
+
"grad_norm": 0.9453125,
|
91 |
+
"learning_rate": 0.00018660254037844388,
|
92 |
+
"loss": 1.4442,
|
93 |
+
"step": 55
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.2727272727272727,
|
97 |
+
"grad_norm": 0.95703125,
|
98 |
+
"learning_rate": 0.0001823676581429833,
|
99 |
+
"loss": 1.4328,
|
100 |
+
"step": 60
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.29545454545454547,
|
104 |
+
"grad_norm": 0.875,
|
105 |
+
"learning_rate": 0.0001776146464291757,
|
106 |
+
"loss": 1.4086,
|
107 |
+
"step": 65
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.3181818181818182,
|
111 |
+
"grad_norm": 1.046875,
|
112 |
+
"learning_rate": 0.00017237340381050703,
|
113 |
+
"loss": 1.3785,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.3409090909090909,
|
118 |
+
"grad_norm": 0.90625,
|
119 |
+
"learning_rate": 0.00016667690005162916,
|
120 |
+
"loss": 1.3806,
|
121 |
+
"step": 75
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.36363636363636365,
|
125 |
+
"grad_norm": 0.7734375,
|
126 |
+
"learning_rate": 0.00016056096871376667,
|
127 |
+
"loss": 1.3681,
|
128 |
+
"step": 80
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.38636363636363635,
|
132 |
+
"grad_norm": 0.66015625,
|
133 |
+
"learning_rate": 0.00015406408174555976,
|
134 |
+
"loss": 1.3576,
|
135 |
+
"step": 85
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.4090909090909091,
|
139 |
+
"grad_norm": 0.6328125,
|
140 |
+
"learning_rate": 0.0001472271074772683,
|
141 |
+
"loss": 1.3267,
|
142 |
+
"step": 90
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.4318181818181818,
|
146 |
+
"grad_norm": 1.015625,
|
147 |
+
"learning_rate": 0.00014009305354066137,
|
148 |
+
"loss": 1.3442,
|
149 |
+
"step": 95
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.45454545454545453,
|
153 |
+
"grad_norm": 0.8984375,
|
154 |
+
"learning_rate": 0.00013270679633174218,
|
155 |
+
"loss": 1.3321,
|
156 |
+
"step": 100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.4772727272727273,
|
160 |
+
"grad_norm": 0.9765625,
|
161 |
+
"learning_rate": 0.0001251147987181079,
|
162 |
+
"loss": 1.3126,
|
163 |
+
"step": 105
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.5,
|
167 |
+
"grad_norm": 0.84765625,
|
168 |
+
"learning_rate": 0.00011736481776669306,
|
169 |
+
"loss": 1.3108,
|
170 |
+
"step": 110
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.5227272727272727,
|
174 |
+
"grad_norm": 0.6640625,
|
175 |
+
"learning_rate": 0.00010950560433041826,
|
176 |
+
"loss": 1.2999,
|
177 |
+
"step": 115
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.5454545454545454,
|
181 |
+
"grad_norm": 0.52734375,
|
182 |
+
"learning_rate": 0.00010158659638348081,
|
183 |
+
"loss": 1.3105,
|
184 |
+
"step": 120
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.5681818181818182,
|
188 |
+
"grad_norm": 0.69140625,
|
189 |
+
"learning_rate": 9.365760803434355e-05,
|
190 |
+
"loss": 1.3079,
|
191 |
+
"step": 125
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.5909090909090909,
|
195 |
+
"grad_norm": 0.58984375,
|
196 |
+
"learning_rate": 8.57685161726715e-05,
|
197 |
+
"loss": 1.2927,
|
198 |
+
"step": 130
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.6136363636363636,
|
202 |
+
"grad_norm": 0.51953125,
|
203 |
+
"learning_rate": 7.796894672134594e-05,
|
204 |
+
"loss": 1.2819,
|
205 |
+
"step": 135
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.6363636363636364,
|
209 |
+
"grad_norm": 0.76171875,
|
210 |
+
"learning_rate": 7.030796246717255e-05,
|
211 |
+
"loss": 1.2973,
|
212 |
+
"step": 140
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.6590909090909091,
|
216 |
+
"grad_norm": 0.64453125,
|
217 |
+
"learning_rate": 6.283375443396726e-05,
|
218 |
+
"loss": 1.2926,
|
219 |
+
"step": 145
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.6818181818181818,
|
223 |
+
"grad_norm": 0.462890625,
|
224 |
+
"learning_rate": 5.559333873942259e-05,
|
225 |
+
"loss": 1.2886,
|
226 |
+
"step": 150
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.7045454545454546,
|
230 |
+
"grad_norm": 0.671875,
|
231 |
+
"learning_rate": 4.8632260842659393e-05,
|
232 |
+
"loss": 1.2927,
|
233 |
+
"step": 155
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.7272727272727273,
|
237 |
+
"grad_norm": 0.51953125,
|
238 |
+
"learning_rate": 4.19943090428802e-05,
|
239 |
+
"loss": 1.2802,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.75,
|
244 |
+
"grad_norm": 0.447265625,
|
245 |
+
"learning_rate": 3.5721239031346066e-05,
|
246 |
+
"loss": 1.2582,
|
247 |
+
"step": 165
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.7727272727272727,
|
251 |
+
"grad_norm": 0.5625,
|
252 |
+
"learning_rate": 2.9852511229367865e-05,
|
253 |
+
"loss": 1.2829,
|
254 |
+
"step": 170
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.7954545454545454,
|
258 |
+
"grad_norm": 0.41796875,
|
259 |
+
"learning_rate": 2.4425042564574184e-05,
|
260 |
+
"loss": 1.2812,
|
261 |
+
"step": 175
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.8181818181818182,
|
265 |
+
"grad_norm": 0.4375,
|
266 |
+
"learning_rate": 1.947297424689414e-05,
|
267 |
+
"loss": 1.2832,
|
268 |
+
"step": 180
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.8409090909090909,
|
272 |
+
"grad_norm": 0.4609375,
|
273 |
+
"learning_rate": 1.5027457005048573e-05,
|
274 |
+
"loss": 1.277,
|
275 |
+
"step": 185
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.8636363636363636,
|
279 |
+
"grad_norm": 0.5234375,
|
280 |
+
"learning_rate": 1.1116455134507664e-05,
|
281 |
+
"loss": 1.2788,
|
282 |
+
"step": 190
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.8863636363636364,
|
286 |
+
"grad_norm": 0.53125,
|
287 |
+
"learning_rate": 7.764570589541875e-06,
|
288 |
+
"loss": 1.2764,
|
289 |
+
"step": 195
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.9090909090909091,
|
293 |
+
"grad_norm": 0.5078125,
|
294 |
+
"learning_rate": 4.992888225905468e-06,
|
295 |
+
"loss": 1.2836,
|
296 |
+
"step": 200
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.9318181818181818,
|
300 |
+
"grad_norm": 0.65234375,
|
301 |
+
"learning_rate": 2.818843167645835e-06,
|
302 |
+
"loss": 1.2792,
|
303 |
+
"step": 205
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.9545454545454546,
|
307 |
+
"grad_norm": 0.5625,
|
308 |
+
"learning_rate": 1.2561111323605712e-06,
|
309 |
+
"loss": 1.2774,
|
310 |
+
"step": 210
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.9772727272727273,
|
314 |
+
"grad_norm": 0.466796875,
|
315 |
+
"learning_rate": 3.145224048057727e-07,
|
316 |
+
"loss": 1.2792,
|
317 |
+
"step": 215
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 1.0,
|
321 |
+
"grad_norm": 0.51171875,
|
322 |
+
"learning_rate": 0.0,
|
323 |
+
"loss": 1.2733,
|
324 |
+
"step": 220
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 1.0,
|
328 |
+
"step": 220,
|
329 |
+
"total_flos": 7357983621120000.0,
|
330 |
+
"train_loss": 1.3856282998215068,
|
331 |
+
"train_runtime": 92.8881,
|
332 |
+
"train_samples_per_second": 151.214,
|
333 |
+
"train_steps_per_second": 2.368
|
334 |
+
}
|
335 |
+
],
|
336 |
+
"logging_steps": 5,
|
337 |
+
"max_steps": 220,
|
338 |
+
"num_input_tokens_seen": 0,
|
339 |
+
"num_train_epochs": 1,
|
340 |
+
"save_steps": 100,
|
341 |
+
"stateful_callbacks": {
|
342 |
+
"TrainerControl": {
|
343 |
+
"args": {
|
344 |
+
"should_epoch_stop": false,
|
345 |
+
"should_evaluate": false,
|
346 |
+
"should_log": false,
|
347 |
+
"should_save": true,
|
348 |
+
"should_training_stop": false
|
349 |
+
},
|
350 |
+
"attributes": {}
|
351 |
+
}
|
352 |
+
},
|
353 |
+
"total_flos": 7357983621120000.0,
|
354 |
+
"train_batch_size": 16,
|
355 |
+
"trial_name": null,
|
356 |
+
"trial_params": null
|
357 |
+
}
|