AlekseyKorshuk commited on
Commit
b4c3898
1 Parent(s): b808c3c

huggingartists

Browse files
README.md ADDED
@@ -0,0 +1,97 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ datasets:
4
+ - huggingartists/andre-3000
5
+ tags:
6
+ - huggingartists
7
+ - lyrics
8
+ - lm-head
9
+ - causal-lm
10
+ widget:
11
+ - text: "I am"
12
+ ---
13
+
14
+ <div class="inline-flex flex-col" style="line-height: 1.5;">
15
+ <div class="flex">
16
+ <div
17
+ style="display:DISPLAY_1; margin-left: auto; margin-right: auto; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url(&#39;https://images.genius.com/64b15c9489c65f5bf8f6577334347404.434x434x1.jpg&#39;)">
18
+ </div>
19
+ </div>
20
+ <div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 HuggingArtists Model 🤖</div>
21
+ <div style="text-align: center; font-size: 16px; font-weight: 800">André 3000</div>
22
+ <a href="https://genius.com/artists/andre-3000">
23
+ <div style="text-align: center; font-size: 14px;">@andre-3000</div>
24
+ </a>
25
+ </div>
26
+
27
+ I was made with [huggingartists](https://github.com/AlekseyKorshuk/huggingartists).
28
+
29
+ Create your own bot based on your favorite artist with [the demo](https://colab.research.google.com/github/AlekseyKorshuk/huggingartists/blob/master/huggingartists-demo.ipynb)!
30
+
31
+ ## How does it work?
32
+
33
+ To understand how the model was developed, check the [W&B report](https://wandb.ai/huggingartists/huggingartists/reportlist).
34
+
35
+ ## Training data
36
+
37
+ The model was trained on lyrics from André 3000.
38
+
39
+ Dataset is available [here](https://huggingface.co/datasets/huggingartists/andre-3000).
40
+ And can be used with:
41
+
42
+ ```python
43
+ from datasets import load_dataset
44
+
45
+ dataset = load_dataset("huggingartists/andre-3000")
46
+ ```
47
+
48
+ [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/zdji17w8/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
49
+
50
+ ## Training procedure
51
+
52
+ The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on André 3000's lyrics.
53
+
54
+ Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1x8qakj9) for full transparency and reproducibility.
55
+
56
+ At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1x8qakj9/artifacts) is logged and versioned.
57
+
58
+ ## How to use
59
+
60
+ You can use this model directly with a pipeline for text generation:
61
+
62
+ ```python
63
+ from transformers import pipeline
64
+ generator = pipeline('text-generation',
65
+ model='huggingartists/andre-3000')
66
+ generator("I am", num_return_sequences=5)
67
+ ```
68
+
69
+ Or with Transformers library:
70
+
71
+ ```python
72
+ from transformers import AutoTokenizer, AutoModelWithLMHead
73
+
74
+ tokenizer = AutoTokenizer.from_pretrained("huggingartists/andre-3000")
75
+
76
+ model = AutoModelWithLMHead.from_pretrained("huggingartists/andre-3000")
77
+ ```
78
+
79
+ ## Limitations and bias
80
+
81
+ The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias).
82
+
83
+ In addition, the data present in the user's tweets further affects the text generated by the model.
84
+
85
+ ## About
86
+
87
+ *Built by Aleksey Korshuk*
88
+
89
+ [![Follow](https://img.shields.io/github/followers/AlekseyKorshuk?style=social)](https://github.com/AlekseyKorshuk)
90
+
91
+ [![Follow](https://img.shields.io/twitter/follow/alekseykorshuk?style=social)](https://twitter.com/intent/follow?screen_name=alekseykorshuk)
92
+
93
+ [![Follow](https://img.shields.io/badge/dynamic/json?color=blue&label=Telegram%20Channel&query=%24.result&url=https%3A%2F%2Fapi.telegram.org%2Fbot1929545866%3AAAFGhV-KKnegEcLiyYJxsc4zV6C-bdPEBtQ%2FgetChatMemberCount%3Fchat_id%3D-1001253621662&style=social&logo=telegram)](https://t.me/joinchat/_CQ04KjcJ-4yZTky)
94
+
95
+ For more details, visit the project repository.
96
+
97
+ [![GitHub stars](https://img.shields.io/github/stars/AlekseyKorshuk/huggingartists?style=social)](https://github.com/AlekseyKorshuk/huggingartists)
config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "gpt2",
3
+ "activation_function": "gelu_new",
4
+ "architectures": [
5
+ "GPT2LMHeadModel"
6
+ ],
7
+ "attn_pdrop": 0.1,
8
+ "bos_token_id": 50256,
9
+ "embd_pdrop": 0.1,
10
+ "eos_token_id": 50256,
11
+ "initializer_range": 0.02,
12
+ "layer_norm_epsilon": 1e-05,
13
+ "model_type": "gpt2",
14
+ "n_ctx": 1024,
15
+ "n_embd": 768,
16
+ "n_head": 12,
17
+ "n_inner": null,
18
+ "n_layer": 12,
19
+ "n_positions": 1024,
20
+ "resid_pdrop": 0.1,
21
+ "scale_attn_weights": true,
22
+ "summary_activation": null,
23
+ "summary_first_dropout": 0.1,
24
+ "summary_proj_to_labels": true,
25
+ "summary_type": "cls_index",
26
+ "summary_use_proj": true,
27
+ "task_specific_params": {
28
+ "text-generation": {
29
+ "do_sample": true,
30
+ "max_length": 200,
31
+ "min_length": 100,
32
+ "temperature": 1.0,
33
+ "top_p": 0.95
34
+ }
35
+ },
36
+ "torch_dtype": "float32",
37
+ "transformers_version": "4.11.3",
38
+ "use_cache": true,
39
+ "vocab_size": 50257
40
+ }
evaluation.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eval_loss": 3.5885772705078125, "eval_runtime": 3.0977, "eval_samples_per_second": 43.258, "eval_steps_per_second": 5.488, "epoch": 10.0}
flax_model.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4c5fc3549feddb90f7e822ada817b97afa10f79c935226937d817c046cf9f0d
3
+ size 497764120
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1fd9963abebba007a6bd7e6d0556ca7196371944164c15c125dc13d92de7783
3
+ size 995604017
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1f5d517dcc771d45d75c794a20a60c90fd1be2cddf41b2acd8053771a778089
3
+ size 510403817
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d62244ae79eb401974a845e8e52733528244c631389781aa6c66540ff1a2ca27
3
+ size 14567
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eab14c8a89edd442bfb86b7ee9c03bcf272aae22b0b5ce034a32fd2886c2bc24
3
+ size 623
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2", "tokenizer_class": "GPT2Tokenizer"}
trainer_state.json ADDED
@@ -0,0 +1,656 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 3.5885772705078125,
3
+ "best_model_checkpoint": "output/andre-3000/checkpoint-500",
4
+ "epoch": 5.0,
5
+ "global_step": 500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.05,
12
+ "learning_rate": 0.00013635542016482644,
13
+ "loss": 3.842,
14
+ "step": 5
15
+ },
16
+ {
17
+ "epoch": 0.1,
18
+ "learning_rate": 0.00013384247701784754,
19
+ "loss": 4.137,
20
+ "step": 10
21
+ },
22
+ {
23
+ "epoch": 0.15,
24
+ "learning_rate": 0.00012972304755932203,
25
+ "loss": 3.733,
26
+ "step": 15
27
+ },
28
+ {
29
+ "epoch": 0.2,
30
+ "learning_rate": 0.0001240985658141214,
31
+ "loss": 3.9016,
32
+ "step": 20
33
+ },
34
+ {
35
+ "epoch": 0.25,
36
+ "learning_rate": 0.00011710752518939715,
37
+ "loss": 3.9056,
38
+ "step": 25
39
+ },
40
+ {
41
+ "epoch": 0.3,
42
+ "learning_rate": 0.00010892206830726365,
43
+ "loss": 3.4673,
44
+ "step": 30
45
+ },
46
+ {
47
+ "epoch": 0.35,
48
+ "learning_rate": 9.974374828213291e-05,
49
+ "loss": 3.4192,
50
+ "step": 35
51
+ },
52
+ {
53
+ "epoch": 0.4,
54
+ "learning_rate": 8.97985658141214e-05,
55
+ "loss": 3.5188,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 0.45,
60
+ "learning_rate": 7.933140430175984e-05,
61
+ "loss": 3.6982,
62
+ "step": 45
63
+ },
64
+ {
65
+ "epoch": 0.5,
66
+ "learning_rate": 6.86e-05,
67
+ "loss": 3.6325,
68
+ "step": 50
69
+ },
70
+ {
71
+ "epoch": 0.55,
72
+ "learning_rate": 5.786859569824015e-05,
73
+ "loss": 3.5144,
74
+ "step": 55
75
+ },
76
+ {
77
+ "epoch": 0.6,
78
+ "learning_rate": 4.740143418587861e-05,
79
+ "loss": 3.6288,
80
+ "step": 60
81
+ },
82
+ {
83
+ "epoch": 0.65,
84
+ "learning_rate": 3.745625171786709e-05,
85
+ "loss": 3.5069,
86
+ "step": 65
87
+ },
88
+ {
89
+ "epoch": 0.7,
90
+ "learning_rate": 2.8277931692736352e-05,
91
+ "loss": 3.7103,
92
+ "step": 70
93
+ },
94
+ {
95
+ "epoch": 0.75,
96
+ "learning_rate": 2.0092474810602843e-05,
97
+ "loss": 3.5622,
98
+ "step": 75
99
+ },
100
+ {
101
+ "epoch": 0.8,
102
+ "learning_rate": 1.3101434185878613e-05,
103
+ "loss": 3.5587,
104
+ "step": 80
105
+ },
106
+ {
107
+ "epoch": 0.85,
108
+ "learning_rate": 7.476952440677969e-06,
109
+ "loss": 3.5987,
110
+ "step": 85
111
+ },
112
+ {
113
+ "epoch": 0.9,
114
+ "learning_rate": 3.357522982152468e-06,
115
+ "loss": 3.4558,
116
+ "step": 90
117
+ },
118
+ {
119
+ "epoch": 0.95,
120
+ "learning_rate": 8.445798351735566e-07,
121
+ "loss": 3.528,
122
+ "step": 95
123
+ },
124
+ {
125
+ "epoch": 1.0,
126
+ "learning_rate": 0.0,
127
+ "loss": 3.5143,
128
+ "step": 100
129
+ },
130
+ {
131
+ "epoch": 1.0,
132
+ "eval_loss": 3.681246519088745,
133
+ "eval_runtime": 2.8995,
134
+ "eval_samples_per_second": 46.215,
135
+ "eval_steps_per_second": 5.863,
136
+ "step": 100
137
+ },
138
+ {
139
+ "epoch": 1.05,
140
+ "learning_rate": 8.445798351735489e-07,
141
+ "loss": 3.5508,
142
+ "step": 105
143
+ },
144
+ {
145
+ "epoch": 1.1,
146
+ "learning_rate": 3.357522982152468e-06,
147
+ "loss": 3.6326,
148
+ "step": 110
149
+ },
150
+ {
151
+ "epoch": 1.15,
152
+ "learning_rate": 7.4769524406779465e-06,
153
+ "loss": 3.3899,
154
+ "step": 115
155
+ },
156
+ {
157
+ "epoch": 1.2,
158
+ "learning_rate": 1.3101434185878598e-05,
159
+ "loss": 3.3572,
160
+ "step": 120
161
+ },
162
+ {
163
+ "epoch": 1.25,
164
+ "learning_rate": 2.009247481060283e-05,
165
+ "loss": 3.5502,
166
+ "step": 125
167
+ },
168
+ {
169
+ "epoch": 1.3,
170
+ "learning_rate": 2.8277931692736335e-05,
171
+ "loss": 3.3994,
172
+ "step": 130
173
+ },
174
+ {
175
+ "epoch": 1.35,
176
+ "learning_rate": 3.745625171786708e-05,
177
+ "loss": 3.2566,
178
+ "step": 135
179
+ },
180
+ {
181
+ "epoch": 1.4,
182
+ "learning_rate": 4.74014341858786e-05,
183
+ "loss": 3.503,
184
+ "step": 140
185
+ },
186
+ {
187
+ "epoch": 1.45,
188
+ "learning_rate": 5.786859569824015e-05,
189
+ "loss": 3.0786,
190
+ "step": 145
191
+ },
192
+ {
193
+ "epoch": 1.5,
194
+ "learning_rate": 6.859999999999999e-05,
195
+ "loss": 3.2728,
196
+ "step": 150
197
+ },
198
+ {
199
+ "epoch": 1.55,
200
+ "learning_rate": 7.933140430175983e-05,
201
+ "loss": 3.1862,
202
+ "step": 155
203
+ },
204
+ {
205
+ "epoch": 1.6,
206
+ "learning_rate": 8.979856581412138e-05,
207
+ "loss": 3.2618,
208
+ "step": 160
209
+ },
210
+ {
211
+ "epoch": 1.65,
212
+ "learning_rate": 9.974374828213291e-05,
213
+ "loss": 3.4498,
214
+ "step": 165
215
+ },
216
+ {
217
+ "epoch": 1.7,
218
+ "learning_rate": 0.00010892206830726364,
219
+ "loss": 3.1053,
220
+ "step": 170
221
+ },
222
+ {
223
+ "epoch": 1.75,
224
+ "learning_rate": 0.00011710752518939715,
225
+ "loss": 3.3994,
226
+ "step": 175
227
+ },
228
+ {
229
+ "epoch": 1.8,
230
+ "learning_rate": 0.0001240985658141214,
231
+ "loss": 3.3686,
232
+ "step": 180
233
+ },
234
+ {
235
+ "epoch": 1.85,
236
+ "learning_rate": 0.00012972304755932203,
237
+ "loss": 3.4802,
238
+ "step": 185
239
+ },
240
+ {
241
+ "epoch": 1.9,
242
+ "learning_rate": 0.00013384247701784754,
243
+ "loss": 3.4496,
244
+ "step": 190
245
+ },
246
+ {
247
+ "epoch": 1.95,
248
+ "learning_rate": 0.00013635542016482644,
249
+ "loss": 3.341,
250
+ "step": 195
251
+ },
252
+ {
253
+ "epoch": 2.0,
254
+ "learning_rate": 0.0001372,
255
+ "loss": 3.2332,
256
+ "step": 200
257
+ },
258
+ {
259
+ "epoch": 2.0,
260
+ "eval_loss": 3.6474390029907227,
261
+ "eval_runtime": 3.0601,
262
+ "eval_samples_per_second": 43.789,
263
+ "eval_steps_per_second": 5.555,
264
+ "step": 200
265
+ },
266
+ {
267
+ "epoch": 2.05,
268
+ "learning_rate": 0.00013635542016482644,
269
+ "loss": 3.2607,
270
+ "step": 205
271
+ },
272
+ {
273
+ "epoch": 2.1,
274
+ "learning_rate": 0.00013384247701784754,
275
+ "loss": 3.3,
276
+ "step": 210
277
+ },
278
+ {
279
+ "epoch": 2.15,
280
+ "learning_rate": 0.00012972304755932206,
281
+ "loss": 3.3456,
282
+ "step": 215
283
+ },
284
+ {
285
+ "epoch": 2.2,
286
+ "learning_rate": 0.00012409856581412136,
287
+ "loss": 3.0185,
288
+ "step": 220
289
+ },
290
+ {
291
+ "epoch": 2.25,
292
+ "learning_rate": 0.00011710752518939717,
293
+ "loss": 2.9552,
294
+ "step": 225
295
+ },
296
+ {
297
+ "epoch": 2.3,
298
+ "learning_rate": 0.00010892206830726372,
299
+ "loss": 3.0603,
300
+ "step": 230
301
+ },
302
+ {
303
+ "epoch": 2.35,
304
+ "learning_rate": 9.974374828213292e-05,
305
+ "loss": 3.2515,
306
+ "step": 235
307
+ },
308
+ {
309
+ "epoch": 2.4,
310
+ "learning_rate": 8.979856581412141e-05,
311
+ "loss": 3.1474,
312
+ "step": 240
313
+ },
314
+ {
315
+ "epoch": 2.45,
316
+ "learning_rate": 7.93314043017598e-05,
317
+ "loss": 3.1997,
318
+ "step": 245
319
+ },
320
+ {
321
+ "epoch": 2.5,
322
+ "learning_rate": 6.860000000000001e-05,
323
+ "loss": 3.2995,
324
+ "step": 250
325
+ },
326
+ {
327
+ "epoch": 2.55,
328
+ "learning_rate": 5.786859569824025e-05,
329
+ "loss": 3.0081,
330
+ "step": 255
331
+ },
332
+ {
333
+ "epoch": 2.6,
334
+ "learning_rate": 4.7401434185878625e-05,
335
+ "loss": 3.0519,
336
+ "step": 260
337
+ },
338
+ {
339
+ "epoch": 2.65,
340
+ "learning_rate": 3.745625171786717e-05,
341
+ "loss": 3.1474,
342
+ "step": 265
343
+ },
344
+ {
345
+ "epoch": 2.7,
346
+ "learning_rate": 2.827793169273636e-05,
347
+ "loss": 3.0318,
348
+ "step": 270
349
+ },
350
+ {
351
+ "epoch": 2.75,
352
+ "learning_rate": 2.0092474810602897e-05,
353
+ "loss": 3.0418,
354
+ "step": 275
355
+ },
356
+ {
357
+ "epoch": 2.8,
358
+ "learning_rate": 1.310143418587862e-05,
359
+ "loss": 3.2108,
360
+ "step": 280
361
+ },
362
+ {
363
+ "epoch": 2.85,
364
+ "learning_rate": 7.4769524406779465e-06,
365
+ "loss": 2.8269,
366
+ "step": 285
367
+ },
368
+ {
369
+ "epoch": 2.9,
370
+ "learning_rate": 3.3575229821524754e-06,
371
+ "loss": 3.0926,
372
+ "step": 290
373
+ },
374
+ {
375
+ "epoch": 2.95,
376
+ "learning_rate": 8.445798351735489e-07,
377
+ "loss": 3.1617,
378
+ "step": 295
379
+ },
380
+ {
381
+ "epoch": 3.0,
382
+ "learning_rate": 0.0,
383
+ "loss": 2.8832,
384
+ "step": 300
385
+ },
386
+ {
387
+ "epoch": 3.0,
388
+ "eval_loss": 3.5949366092681885,
389
+ "eval_runtime": 3.0805,
390
+ "eval_samples_per_second": 43.5,
391
+ "eval_steps_per_second": 5.519,
392
+ "step": 300
393
+ },
394
+ {
395
+ "epoch": 3.05,
396
+ "learning_rate": 8.445798351735413e-07,
397
+ "loss": 2.8947,
398
+ "step": 305
399
+ },
400
+ {
401
+ "epoch": 3.1,
402
+ "learning_rate": 3.35752298215246e-06,
403
+ "loss": 2.8681,
404
+ "step": 310
405
+ },
406
+ {
407
+ "epoch": 3.15,
408
+ "learning_rate": 7.4769524406779245e-06,
409
+ "loss": 2.925,
410
+ "step": 315
411
+ },
412
+ {
413
+ "epoch": 3.2,
414
+ "learning_rate": 1.310143418587859e-05,
415
+ "loss": 3.0341,
416
+ "step": 320
417
+ },
418
+ {
419
+ "epoch": 3.25,
420
+ "learning_rate": 2.0092474810602867e-05,
421
+ "loss": 2.923,
422
+ "step": 325
423
+ },
424
+ {
425
+ "epoch": 3.3,
426
+ "learning_rate": 2.827793169273632e-05,
427
+ "loss": 3.1311,
428
+ "step": 330
429
+ },
430
+ {
431
+ "epoch": 3.35,
432
+ "learning_rate": 3.7456251717867126e-05,
433
+ "loss": 2.9954,
434
+ "step": 335
435
+ },
436
+ {
437
+ "epoch": 3.4,
438
+ "learning_rate": 4.740143418587858e-05,
439
+ "loss": 2.9511,
440
+ "step": 340
441
+ },
442
+ {
443
+ "epoch": 3.45,
444
+ "learning_rate": 5.7868595698240195e-05,
445
+ "loss": 2.9837,
446
+ "step": 345
447
+ },
448
+ {
449
+ "epoch": 3.5,
450
+ "learning_rate": 6.859999999999997e-05,
451
+ "loss": 3.0341,
452
+ "step": 350
453
+ },
454
+ {
455
+ "epoch": 3.55,
456
+ "learning_rate": 7.933140430175976e-05,
457
+ "loss": 2.9587,
458
+ "step": 355
459
+ },
460
+ {
461
+ "epoch": 3.6,
462
+ "learning_rate": 8.979856581412137e-05,
463
+ "loss": 2.9158,
464
+ "step": 360
465
+ },
466
+ {
467
+ "epoch": 3.65,
468
+ "learning_rate": 9.974374828213283e-05,
469
+ "loss": 2.8452,
470
+ "step": 365
471
+ },
472
+ {
473
+ "epoch": 3.7,
474
+ "learning_rate": 0.00010892206830726364,
475
+ "loss": 2.9248,
476
+ "step": 370
477
+ },
478
+ {
479
+ "epoch": 3.75,
480
+ "learning_rate": 0.0001171075251893971,
481
+ "loss": 2.9796,
482
+ "step": 375
483
+ },
484
+ {
485
+ "epoch": 3.8,
486
+ "learning_rate": 0.00012409856581412136,
487
+ "loss": 3.0723,
488
+ "step": 380
489
+ },
490
+ {
491
+ "epoch": 3.85,
492
+ "learning_rate": 0.00012972304755932206,
493
+ "loss": 2.7475,
494
+ "step": 385
495
+ },
496
+ {
497
+ "epoch": 3.9,
498
+ "learning_rate": 0.00013384247701784751,
499
+ "loss": 2.9859,
500
+ "step": 390
501
+ },
502
+ {
503
+ "epoch": 3.95,
504
+ "learning_rate": 0.00013635542016482644,
505
+ "loss": 2.8605,
506
+ "step": 395
507
+ },
508
+ {
509
+ "epoch": 4.0,
510
+ "learning_rate": 0.0001372,
511
+ "loss": 2.8721,
512
+ "step": 400
513
+ },
514
+ {
515
+ "epoch": 4.0,
516
+ "eval_loss": 3.6242599487304688,
517
+ "eval_runtime": 3.0946,
518
+ "eval_samples_per_second": 43.301,
519
+ "eval_steps_per_second": 5.493,
520
+ "step": 400
521
+ },
522
+ {
523
+ "epoch": 4.05,
524
+ "learning_rate": 0.00013635542016482647,
525
+ "loss": 3.0586,
526
+ "step": 405
527
+ },
528
+ {
529
+ "epoch": 4.1,
530
+ "learning_rate": 0.0001338424770178476,
531
+ "loss": 2.5264,
532
+ "step": 410
533
+ },
534
+ {
535
+ "epoch": 4.15,
536
+ "learning_rate": 0.00012972304755932203,
537
+ "loss": 2.591,
538
+ "step": 415
539
+ },
540
+ {
541
+ "epoch": 4.2,
542
+ "learning_rate": 0.00012409856581412142,
543
+ "loss": 2.7559,
544
+ "step": 420
545
+ },
546
+ {
547
+ "epoch": 4.25,
548
+ "learning_rate": 0.00011710752518939714,
549
+ "loss": 2.6444,
550
+ "step": 425
551
+ },
552
+ {
553
+ "epoch": 4.3,
554
+ "learning_rate": 0.00010892206830726369,
555
+ "loss": 3.0661,
556
+ "step": 430
557
+ },
558
+ {
559
+ "epoch": 4.35,
560
+ "learning_rate": 9.9743748282133e-05,
561
+ "loss": 2.639,
562
+ "step": 435
563
+ },
564
+ {
565
+ "epoch": 4.4,
566
+ "learning_rate": 8.97985658141213e-05,
567
+ "loss": 2.8619,
568
+ "step": 440
569
+ },
570
+ {
571
+ "epoch": 4.45,
572
+ "learning_rate": 7.933140430175981e-05,
573
+ "loss": 2.6417,
574
+ "step": 445
575
+ },
576
+ {
577
+ "epoch": 4.5,
578
+ "learning_rate": 6.860000000000003e-05,
579
+ "loss": 2.7156,
580
+ "step": 450
581
+ },
582
+ {
583
+ "epoch": 4.55,
584
+ "learning_rate": 5.786859569824026e-05,
585
+ "loss": 2.9429,
586
+ "step": 455
587
+ },
588
+ {
589
+ "epoch": 4.6,
590
+ "learning_rate": 4.740143418587876e-05,
591
+ "loss": 2.7116,
592
+ "step": 460
593
+ },
594
+ {
595
+ "epoch": 4.65,
596
+ "learning_rate": 3.745625171786707e-05,
597
+ "loss": 2.6837,
598
+ "step": 465
599
+ },
600
+ {
601
+ "epoch": 4.7,
602
+ "learning_rate": 2.8277931692736372e-05,
603
+ "loss": 2.8497,
604
+ "step": 470
605
+ },
606
+ {
607
+ "epoch": 4.75,
608
+ "learning_rate": 2.0092474810602914e-05,
609
+ "loss": 2.713,
610
+ "step": 475
611
+ },
612
+ {
613
+ "epoch": 4.8,
614
+ "learning_rate": 1.3101434185878628e-05,
615
+ "loss": 2.7103,
616
+ "step": 480
617
+ },
618
+ {
619
+ "epoch": 4.85,
620
+ "learning_rate": 7.4769524406780075e-06,
621
+ "loss": 2.6698,
622
+ "step": 485
623
+ },
624
+ {
625
+ "epoch": 4.9,
626
+ "learning_rate": 3.3575229821524373e-06,
627
+ "loss": 2.8479,
628
+ "step": 490
629
+ },
630
+ {
631
+ "epoch": 4.95,
632
+ "learning_rate": 8.445798351735489e-07,
633
+ "loss": 2.7422,
634
+ "step": 495
635
+ },
636
+ {
637
+ "epoch": 5.0,
638
+ "learning_rate": 0.0,
639
+ "loss": 2.8229,
640
+ "step": 500
641
+ },
642
+ {
643
+ "epoch": 5.0,
644
+ "eval_loss": 3.5885772705078125,
645
+ "eval_runtime": 3.0956,
646
+ "eval_samples_per_second": 43.288,
647
+ "eval_steps_per_second": 5.492,
648
+ "step": 500
649
+ }
650
+ ],
651
+ "max_steps": 1000,
652
+ "num_train_epochs": 10,
653
+ "total_flos": 521277603840000.0,
654
+ "trial_name": null,
655
+ "trial_params": null
656
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5308c09e205af506486abe9769533704d74f4eee1e24a6f594dd6e3b76fa381
3
+ size 2863
vocab.json ADDED
The diff for this file is too large to render. See raw diff