Model save
Browse files- last-checkpoint/generation_config.json → generation_config.json +0 -0
- last-checkpoint/config.json +0 -68
- last-checkpoint/merges.txt +0 -0
- last-checkpoint/optimizer.pt +0 -3
- last-checkpoint/pytorch_model.bin +0 -3
- last-checkpoint/rng_state.pth +0 -3
- last-checkpoint/scaler.pt +0 -3
- last-checkpoint/scheduler.pt +0 -3
- last-checkpoint/special_tokens_map.json +0 -753
- last-checkpoint/tokenizer.json +0 -0
- last-checkpoint/tokenizer_config.json +0 -64
- last-checkpoint/trainer_state.json +0 -336
- last-checkpoint/training_args.bin +0 -3
- last-checkpoint/vocab.json +0 -0
- pytorch_model.bin +1 -1
last-checkpoint/generation_config.json → generation_config.json
RENAMED
File without changes
|
last-checkpoint/config.json
DELETED
@@ -1,68 +0,0 @@
|
|
1 |
-
{
|
2 |
-
"_name_or_path": "Salesforce/codet5-small",
|
3 |
-
"architectures": [
|
4 |
-
"T5ForConditionalGeneration"
|
5 |
-
],
|
6 |
-
"bos_token_id": 1,
|
7 |
-
"d_ff": 2048,
|
8 |
-
"d_kv": 64,
|
9 |
-
"d_model": 512,
|
10 |
-
"decoder_start_token_id": 0,
|
11 |
-
"dense_act_fn": "relu",
|
12 |
-
"dropout_rate": 0.1,
|
13 |
-
"eos_token_id": 2,
|
14 |
-
"feed_forward_proj": "relu",
|
15 |
-
"gradient_checkpointing": false,
|
16 |
-
"id2label": {
|
17 |
-
"0": "LABEL_0"
|
18 |
-
},
|
19 |
-
"initializer_factor": 1.0,
|
20 |
-
"is_encoder_decoder": true,
|
21 |
-
"is_gated_act": false,
|
22 |
-
"label2id": {
|
23 |
-
"LABEL_0": 0
|
24 |
-
},
|
25 |
-
"layer_norm_epsilon": 1e-06,
|
26 |
-
"model_type": "t5",
|
27 |
-
"n_positions": 512,
|
28 |
-
"num_decoder_layers": 6,
|
29 |
-
"num_heads": 8,
|
30 |
-
"num_layers": 6,
|
31 |
-
"output_past": true,
|
32 |
-
"pad_token_id": 0,
|
33 |
-
"relative_attention_max_distance": 128,
|
34 |
-
"relative_attention_num_buckets": 32,
|
35 |
-
"task_specific_params": {
|
36 |
-
"summarization": {
|
37 |
-
"early_stopping": true,
|
38 |
-
"length_penalty": 2.0,
|
39 |
-
"max_length": 200,
|
40 |
-
"min_length": 30,
|
41 |
-
"no_repeat_ngram_size": 3,
|
42 |
-
"num_beams": 4,
|
43 |
-
"prefix": "summarize: "
|
44 |
-
},
|
45 |
-
"translation_en_to_de": {
|
46 |
-
"early_stopping": true,
|
47 |
-
"max_length": 300,
|
48 |
-
"num_beams": 4,
|
49 |
-
"prefix": "translate English to German: "
|
50 |
-
},
|
51 |
-
"translation_en_to_fr": {
|
52 |
-
"early_stopping": true,
|
53 |
-
"max_length": 300,
|
54 |
-
"num_beams": 4,
|
55 |
-
"prefix": "translate English to French: "
|
56 |
-
},
|
57 |
-
"translation_en_to_ro": {
|
58 |
-
"early_stopping": true,
|
59 |
-
"max_length": 300,
|
60 |
-
"num_beams": 4,
|
61 |
-
"prefix": "translate English to Romanian: "
|
62 |
-
}
|
63 |
-
},
|
64 |
-
"torch_dtype": "float32",
|
65 |
-
"transformers_version": "4.27.1",
|
66 |
-
"use_cache": true,
|
67 |
-
"vocab_size": 32100
|
68 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
last-checkpoint/merges.txt
DELETED
The diff for this file is too large to render.
See raw diff
|
|
last-checkpoint/optimizer.pt
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:1cce9b37f48e8675d69401a4b6b60995179e62a8ffbc1f88a02a7bbea5bba909
|
3 |
-
size 837935
|
|
|
|
|
|
|
|
last-checkpoint/pytorch_model.bin
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:19af74a938a7be689d9873da872da964364ba343bd2170276819b7a95570306c
|
3 |
-
size 242012923
|
|
|
|
|
|
|
|
last-checkpoint/rng_state.pth
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:0f9038bd13cee7be809b870072293469c3cd227459e6b82a61f4cf6abfd04c52
|
3 |
-
size 14503
|
|
|
|
|
|
|
|
last-checkpoint/scaler.pt
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:3550fcb9eeeaf0d0c5e60b4fa40bb13ef4e2248224487e336467d1aedec98b8e
|
3 |
-
size 559
|
|
|
|
|
|
|
|
last-checkpoint/scheduler.pt
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:c5d5ab462f8cc6dcbba2126c3954800aa77845e46f59352aa2ec60dd3148cb35
|
3 |
-
size 623
|
|
|
|
|
|
|
|
last-checkpoint/special_tokens_map.json
DELETED
@@ -1,753 +0,0 @@
|
|
1 |
-
{
|
2 |
-
"additional_special_tokens": [
|
3 |
-
{
|
4 |
-
"content": "<extra_id_99>",
|
5 |
-
"lstrip": true,
|
6 |
-
"normalized": true,
|
7 |
-
"rstrip": false,
|
8 |
-
"single_word": false
|
9 |
-
},
|
10 |
-
{
|
11 |
-
"content": "<extra_id_98>",
|
12 |
-
"lstrip": true,
|
13 |
-
"normalized": true,
|
14 |
-
"rstrip": false,
|
15 |
-
"single_word": false
|
16 |
-
},
|
17 |
-
{
|
18 |
-
"content": "<extra_id_97>",
|
19 |
-
"lstrip": true,
|
20 |
-
"normalized": true,
|
21 |
-
"rstrip": false,
|
22 |
-
"single_word": false
|
23 |
-
},
|
24 |
-
{
|
25 |
-
"content": "<extra_id_96>",
|
26 |
-
"lstrip": true,
|
27 |
-
"normalized": true,
|
28 |
-
"rstrip": false,
|
29 |
-
"single_word": false
|
30 |
-
},
|
31 |
-
{
|
32 |
-
"content": "<extra_id_95>",
|
33 |
-
"lstrip": true,
|
34 |
-
"normalized": true,
|
35 |
-
"rstrip": false,
|
36 |
-
"single_word": false
|
37 |
-
},
|
38 |
-
{
|
39 |
-
"content": "<extra_id_94>",
|
40 |
-
"lstrip": true,
|
41 |
-
"normalized": true,
|
42 |
-
"rstrip": false,
|
43 |
-
"single_word": false
|
44 |
-
},
|
45 |
-
{
|
46 |
-
"content": "<extra_id_93>",
|
47 |
-
"lstrip": true,
|
48 |
-
"normalized": true,
|
49 |
-
"rstrip": false,
|
50 |
-
"single_word": false
|
51 |
-
},
|
52 |
-
{
|
53 |
-
"content": "<extra_id_92>",
|
54 |
-
"lstrip": true,
|
55 |
-
"normalized": true,
|
56 |
-
"rstrip": false,
|
57 |
-
"single_word": false
|
58 |
-
},
|
59 |
-
{
|
60 |
-
"content": "<extra_id_91>",
|
61 |
-
"lstrip": true,
|
62 |
-
"normalized": true,
|
63 |
-
"rstrip": false,
|
64 |
-
"single_word": false
|
65 |
-
},
|
66 |
-
{
|
67 |
-
"content": "<extra_id_90>",
|
68 |
-
"lstrip": true,
|
69 |
-
"normalized": true,
|
70 |
-
"rstrip": false,
|
71 |
-
"single_word": false
|
72 |
-
},
|
73 |
-
{
|
74 |
-
"content": "<extra_id_89>",
|
75 |
-
"lstrip": true,
|
76 |
-
"normalized": true,
|
77 |
-
"rstrip": false,
|
78 |
-
"single_word": false
|
79 |
-
},
|
80 |
-
{
|
81 |
-
"content": "<extra_id_88>",
|
82 |
-
"lstrip": true,
|
83 |
-
"normalized": true,
|
84 |
-
"rstrip": false,
|
85 |
-
"single_word": false
|
86 |
-
},
|
87 |
-
{
|
88 |
-
"content": "<extra_id_87>",
|
89 |
-
"lstrip": true,
|
90 |
-
"normalized": true,
|
91 |
-
"rstrip": false,
|
92 |
-
"single_word": false
|
93 |
-
},
|
94 |
-
{
|
95 |
-
"content": "<extra_id_86>",
|
96 |
-
"lstrip": true,
|
97 |
-
"normalized": true,
|
98 |
-
"rstrip": false,
|
99 |
-
"single_word": false
|
100 |
-
},
|
101 |
-
{
|
102 |
-
"content": "<extra_id_85>",
|
103 |
-
"lstrip": true,
|
104 |
-
"normalized": true,
|
105 |
-
"rstrip": false,
|
106 |
-
"single_word": false
|
107 |
-
},
|
108 |
-
{
|
109 |
-
"content": "<extra_id_84>",
|
110 |
-
"lstrip": true,
|
111 |
-
"normalized": true,
|
112 |
-
"rstrip": false,
|
113 |
-
"single_word": false
|
114 |
-
},
|
115 |
-
{
|
116 |
-
"content": "<extra_id_83>",
|
117 |
-
"lstrip": true,
|
118 |
-
"normalized": true,
|
119 |
-
"rstrip": false,
|
120 |
-
"single_word": false
|
121 |
-
},
|
122 |
-
{
|
123 |
-
"content": "<extra_id_82>",
|
124 |
-
"lstrip": true,
|
125 |
-
"normalized": true,
|
126 |
-
"rstrip": false,
|
127 |
-
"single_word": false
|
128 |
-
},
|
129 |
-
{
|
130 |
-
"content": "<extra_id_81>",
|
131 |
-
"lstrip": true,
|
132 |
-
"normalized": true,
|
133 |
-
"rstrip": false,
|
134 |
-
"single_word": false
|
135 |
-
},
|
136 |
-
{
|
137 |
-
"content": "<extra_id_80>",
|
138 |
-
"lstrip": true,
|
139 |
-
"normalized": true,
|
140 |
-
"rstrip": false,
|
141 |
-
"single_word": false
|
142 |
-
},
|
143 |
-
{
|
144 |
-
"content": "<extra_id_79>",
|
145 |
-
"lstrip": true,
|
146 |
-
"normalized": true,
|
147 |
-
"rstrip": false,
|
148 |
-
"single_word": false
|
149 |
-
},
|
150 |
-
{
|
151 |
-
"content": "<extra_id_78>",
|
152 |
-
"lstrip": true,
|
153 |
-
"normalized": true,
|
154 |
-
"rstrip": false,
|
155 |
-
"single_word": false
|
156 |
-
},
|
157 |
-
{
|
158 |
-
"content": "<extra_id_77>",
|
159 |
-
"lstrip": true,
|
160 |
-
"normalized": true,
|
161 |
-
"rstrip": false,
|
162 |
-
"single_word": false
|
163 |
-
},
|
164 |
-
{
|
165 |
-
"content": "<extra_id_76>",
|
166 |
-
"lstrip": true,
|
167 |
-
"normalized": true,
|
168 |
-
"rstrip": false,
|
169 |
-
"single_word": false
|
170 |
-
},
|
171 |
-
{
|
172 |
-
"content": "<extra_id_75>",
|
173 |
-
"lstrip": true,
|
174 |
-
"normalized": true,
|
175 |
-
"rstrip": false,
|
176 |
-
"single_word": false
|
177 |
-
},
|
178 |
-
{
|
179 |
-
"content": "<extra_id_74>",
|
180 |
-
"lstrip": true,
|
181 |
-
"normalized": true,
|
182 |
-
"rstrip": false,
|
183 |
-
"single_word": false
|
184 |
-
},
|
185 |
-
{
|
186 |
-
"content": "<extra_id_73>",
|
187 |
-
"lstrip": true,
|
188 |
-
"normalized": true,
|
189 |
-
"rstrip": false,
|
190 |
-
"single_word": false
|
191 |
-
},
|
192 |
-
{
|
193 |
-
"content": "<extra_id_72>",
|
194 |
-
"lstrip": true,
|
195 |
-
"normalized": true,
|
196 |
-
"rstrip": false,
|
197 |
-
"single_word": false
|
198 |
-
},
|
199 |
-
{
|
200 |
-
"content": "<extra_id_71>",
|
201 |
-
"lstrip": true,
|
202 |
-
"normalized": true,
|
203 |
-
"rstrip": false,
|
204 |
-
"single_word": false
|
205 |
-
},
|
206 |
-
{
|
207 |
-
"content": "<extra_id_70>",
|
208 |
-
"lstrip": true,
|
209 |
-
"normalized": true,
|
210 |
-
"rstrip": false,
|
211 |
-
"single_word": false
|
212 |
-
},
|
213 |
-
{
|
214 |
-
"content": "<extra_id_69>",
|
215 |
-
"lstrip": true,
|
216 |
-
"normalized": true,
|
217 |
-
"rstrip": false,
|
218 |
-
"single_word": false
|
219 |
-
},
|
220 |
-
{
|
221 |
-
"content": "<extra_id_68>",
|
222 |
-
"lstrip": true,
|
223 |
-
"normalized": true,
|
224 |
-
"rstrip": false,
|
225 |
-
"single_word": false
|
226 |
-
},
|
227 |
-
{
|
228 |
-
"content": "<extra_id_67>",
|
229 |
-
"lstrip": true,
|
230 |
-
"normalized": true,
|
231 |
-
"rstrip": false,
|
232 |
-
"single_word": false
|
233 |
-
},
|
234 |
-
{
|
235 |
-
"content": "<extra_id_66>",
|
236 |
-
"lstrip": true,
|
237 |
-
"normalized": true,
|
238 |
-
"rstrip": false,
|
239 |
-
"single_word": false
|
240 |
-
},
|
241 |
-
{
|
242 |
-
"content": "<extra_id_65>",
|
243 |
-
"lstrip": true,
|
244 |
-
"normalized": true,
|
245 |
-
"rstrip": false,
|
246 |
-
"single_word": false
|
247 |
-
},
|
248 |
-
{
|
249 |
-
"content": "<extra_id_64>",
|
250 |
-
"lstrip": true,
|
251 |
-
"normalized": true,
|
252 |
-
"rstrip": false,
|
253 |
-
"single_word": false
|
254 |
-
},
|
255 |
-
{
|
256 |
-
"content": "<extra_id_63>",
|
257 |
-
"lstrip": true,
|
258 |
-
"normalized": true,
|
259 |
-
"rstrip": false,
|
260 |
-
"single_word": false
|
261 |
-
},
|
262 |
-
{
|
263 |
-
"content": "<extra_id_62>",
|
264 |
-
"lstrip": true,
|
265 |
-
"normalized": true,
|
266 |
-
"rstrip": false,
|
267 |
-
"single_word": false
|
268 |
-
},
|
269 |
-
{
|
270 |
-
"content": "<extra_id_61>",
|
271 |
-
"lstrip": true,
|
272 |
-
"normalized": true,
|
273 |
-
"rstrip": false,
|
274 |
-
"single_word": false
|
275 |
-
},
|
276 |
-
{
|
277 |
-
"content": "<extra_id_60>",
|
278 |
-
"lstrip": true,
|
279 |
-
"normalized": true,
|
280 |
-
"rstrip": false,
|
281 |
-
"single_word": false
|
282 |
-
},
|
283 |
-
{
|
284 |
-
"content": "<extra_id_59>",
|
285 |
-
"lstrip": true,
|
286 |
-
"normalized": true,
|
287 |
-
"rstrip": false,
|
288 |
-
"single_word": false
|
289 |
-
},
|
290 |
-
{
|
291 |
-
"content": "<extra_id_58>",
|
292 |
-
"lstrip": true,
|
293 |
-
"normalized": true,
|
294 |
-
"rstrip": false,
|
295 |
-
"single_word": false
|
296 |
-
},
|
297 |
-
{
|
298 |
-
"content": "<extra_id_57>",
|
299 |
-
"lstrip": true,
|
300 |
-
"normalized": true,
|
301 |
-
"rstrip": false,
|
302 |
-
"single_word": false
|
303 |
-
},
|
304 |
-
{
|
305 |
-
"content": "<extra_id_56>",
|
306 |
-
"lstrip": true,
|
307 |
-
"normalized": true,
|
308 |
-
"rstrip": false,
|
309 |
-
"single_word": false
|
310 |
-
},
|
311 |
-
{
|
312 |
-
"content": "<extra_id_55>",
|
313 |
-
"lstrip": true,
|
314 |
-
"normalized": true,
|
315 |
-
"rstrip": false,
|
316 |
-
"single_word": false
|
317 |
-
},
|
318 |
-
{
|
319 |
-
"content": "<extra_id_54>",
|
320 |
-
"lstrip": true,
|
321 |
-
"normalized": true,
|
322 |
-
"rstrip": false,
|
323 |
-
"single_word": false
|
324 |
-
},
|
325 |
-
{
|
326 |
-
"content": "<extra_id_53>",
|
327 |
-
"lstrip": true,
|
328 |
-
"normalized": true,
|
329 |
-
"rstrip": false,
|
330 |
-
"single_word": false
|
331 |
-
},
|
332 |
-
{
|
333 |
-
"content": "<extra_id_52>",
|
334 |
-
"lstrip": true,
|
335 |
-
"normalized": true,
|
336 |
-
"rstrip": false,
|
337 |
-
"single_word": false
|
338 |
-
},
|
339 |
-
{
|
340 |
-
"content": "<extra_id_51>",
|
341 |
-
"lstrip": true,
|
342 |
-
"normalized": true,
|
343 |
-
"rstrip": false,
|
344 |
-
"single_word": false
|
345 |
-
},
|
346 |
-
{
|
347 |
-
"content": "<extra_id_50>",
|
348 |
-
"lstrip": true,
|
349 |
-
"normalized": true,
|
350 |
-
"rstrip": false,
|
351 |
-
"single_word": false
|
352 |
-
},
|
353 |
-
{
|
354 |
-
"content": "<extra_id_49>",
|
355 |
-
"lstrip": true,
|
356 |
-
"normalized": true,
|
357 |
-
"rstrip": false,
|
358 |
-
"single_word": false
|
359 |
-
},
|
360 |
-
{
|
361 |
-
"content": "<extra_id_48>",
|
362 |
-
"lstrip": true,
|
363 |
-
"normalized": true,
|
364 |
-
"rstrip": false,
|
365 |
-
"single_word": false
|
366 |
-
},
|
367 |
-
{
|
368 |
-
"content": "<extra_id_47>",
|
369 |
-
"lstrip": true,
|
370 |
-
"normalized": true,
|
371 |
-
"rstrip": false,
|
372 |
-
"single_word": false
|
373 |
-
},
|
374 |
-
{
|
375 |
-
"content": "<extra_id_46>",
|
376 |
-
"lstrip": true,
|
377 |
-
"normalized": true,
|
378 |
-
"rstrip": false,
|
379 |
-
"single_word": false
|
380 |
-
},
|
381 |
-
{
|
382 |
-
"content": "<extra_id_45>",
|
383 |
-
"lstrip": true,
|
384 |
-
"normalized": true,
|
385 |
-
"rstrip": false,
|
386 |
-
"single_word": false
|
387 |
-
},
|
388 |
-
{
|
389 |
-
"content": "<extra_id_44>",
|
390 |
-
"lstrip": true,
|
391 |
-
"normalized": true,
|
392 |
-
"rstrip": false,
|
393 |
-
"single_word": false
|
394 |
-
},
|
395 |
-
{
|
396 |
-
"content": "<extra_id_43>",
|
397 |
-
"lstrip": true,
|
398 |
-
"normalized": true,
|
399 |
-
"rstrip": false,
|
400 |
-
"single_word": false
|
401 |
-
},
|
402 |
-
{
|
403 |
-
"content": "<extra_id_42>",
|
404 |
-
"lstrip": true,
|
405 |
-
"normalized": true,
|
406 |
-
"rstrip": false,
|
407 |
-
"single_word": false
|
408 |
-
},
|
409 |
-
{
|
410 |
-
"content": "<extra_id_41>",
|
411 |
-
"lstrip": true,
|
412 |
-
"normalized": true,
|
413 |
-
"rstrip": false,
|
414 |
-
"single_word": false
|
415 |
-
},
|
416 |
-
{
|
417 |
-
"content": "<extra_id_40>",
|
418 |
-
"lstrip": true,
|
419 |
-
"normalized": true,
|
420 |
-
"rstrip": false,
|
421 |
-
"single_word": false
|
422 |
-
},
|
423 |
-
{
|
424 |
-
"content": "<extra_id_39>",
|
425 |
-
"lstrip": true,
|
426 |
-
"normalized": true,
|
427 |
-
"rstrip": false,
|
428 |
-
"single_word": false
|
429 |
-
},
|
430 |
-
{
|
431 |
-
"content": "<extra_id_38>",
|
432 |
-
"lstrip": true,
|
433 |
-
"normalized": true,
|
434 |
-
"rstrip": false,
|
435 |
-
"single_word": false
|
436 |
-
},
|
437 |
-
{
|
438 |
-
"content": "<extra_id_37>",
|
439 |
-
"lstrip": true,
|
440 |
-
"normalized": true,
|
441 |
-
"rstrip": false,
|
442 |
-
"single_word": false
|
443 |
-
},
|
444 |
-
{
|
445 |
-
"content": "<extra_id_36>",
|
446 |
-
"lstrip": true,
|
447 |
-
"normalized": true,
|
448 |
-
"rstrip": false,
|
449 |
-
"single_word": false
|
450 |
-
},
|
451 |
-
{
|
452 |
-
"content": "<extra_id_35>",
|
453 |
-
"lstrip": true,
|
454 |
-
"normalized": true,
|
455 |
-
"rstrip": false,
|
456 |
-
"single_word": false
|
457 |
-
},
|
458 |
-
{
|
459 |
-
"content": "<extra_id_34>",
|
460 |
-
"lstrip": true,
|
461 |
-
"normalized": true,
|
462 |
-
"rstrip": false,
|
463 |
-
"single_word": false
|
464 |
-
},
|
465 |
-
{
|
466 |
-
"content": "<extra_id_33>",
|
467 |
-
"lstrip": true,
|
468 |
-
"normalized": true,
|
469 |
-
"rstrip": false,
|
470 |
-
"single_word": false
|
471 |
-
},
|
472 |
-
{
|
473 |
-
"content": "<extra_id_32>",
|
474 |
-
"lstrip": true,
|
475 |
-
"normalized": true,
|
476 |
-
"rstrip": false,
|
477 |
-
"single_word": false
|
478 |
-
},
|
479 |
-
{
|
480 |
-
"content": "<extra_id_31>",
|
481 |
-
"lstrip": true,
|
482 |
-
"normalized": true,
|
483 |
-
"rstrip": false,
|
484 |
-
"single_word": false
|
485 |
-
},
|
486 |
-
{
|
487 |
-
"content": "<extra_id_30>",
|
488 |
-
"lstrip": true,
|
489 |
-
"normalized": true,
|
490 |
-
"rstrip": false,
|
491 |
-
"single_word": false
|
492 |
-
},
|
493 |
-
{
|
494 |
-
"content": "<extra_id_29>",
|
495 |
-
"lstrip": true,
|
496 |
-
"normalized": true,
|
497 |
-
"rstrip": false,
|
498 |
-
"single_word": false
|
499 |
-
},
|
500 |
-
{
|
501 |
-
"content": "<extra_id_28>",
|
502 |
-
"lstrip": true,
|
503 |
-
"normalized": true,
|
504 |
-
"rstrip": false,
|
505 |
-
"single_word": false
|
506 |
-
},
|
507 |
-
{
|
508 |
-
"content": "<extra_id_27>",
|
509 |
-
"lstrip": true,
|
510 |
-
"normalized": true,
|
511 |
-
"rstrip": false,
|
512 |
-
"single_word": false
|
513 |
-
},
|
514 |
-
{
|
515 |
-
"content": "<extra_id_26>",
|
516 |
-
"lstrip": true,
|
517 |
-
"normalized": true,
|
518 |
-
"rstrip": false,
|
519 |
-
"single_word": false
|
520 |
-
},
|
521 |
-
{
|
522 |
-
"content": "<extra_id_25>",
|
523 |
-
"lstrip": true,
|
524 |
-
"normalized": true,
|
525 |
-
"rstrip": false,
|
526 |
-
"single_word": false
|
527 |
-
},
|
528 |
-
{
|
529 |
-
"content": "<extra_id_24>",
|
530 |
-
"lstrip": true,
|
531 |
-
"normalized": true,
|
532 |
-
"rstrip": false,
|
533 |
-
"single_word": false
|
534 |
-
},
|
535 |
-
{
|
536 |
-
"content": "<extra_id_23>",
|
537 |
-
"lstrip": true,
|
538 |
-
"normalized": true,
|
539 |
-
"rstrip": false,
|
540 |
-
"single_word": false
|
541 |
-
},
|
542 |
-
{
|
543 |
-
"content": "<extra_id_22>",
|
544 |
-
"lstrip": true,
|
545 |
-
"normalized": true,
|
546 |
-
"rstrip": false,
|
547 |
-
"single_word": false
|
548 |
-
},
|
549 |
-
{
|
550 |
-
"content": "<extra_id_21>",
|
551 |
-
"lstrip": true,
|
552 |
-
"normalized": true,
|
553 |
-
"rstrip": false,
|
554 |
-
"single_word": false
|
555 |
-
},
|
556 |
-
{
|
557 |
-
"content": "<extra_id_20>",
|
558 |
-
"lstrip": true,
|
559 |
-
"normalized": true,
|
560 |
-
"rstrip": false,
|
561 |
-
"single_word": false
|
562 |
-
},
|
563 |
-
{
|
564 |
-
"content": "<extra_id_19>",
|
565 |
-
"lstrip": true,
|
566 |
-
"normalized": true,
|
567 |
-
"rstrip": false,
|
568 |
-
"single_word": false
|
569 |
-
},
|
570 |
-
{
|
571 |
-
"content": "<extra_id_18>",
|
572 |
-
"lstrip": true,
|
573 |
-
"normalized": true,
|
574 |
-
"rstrip": false,
|
575 |
-
"single_word": false
|
576 |
-
},
|
577 |
-
{
|
578 |
-
"content": "<extra_id_17>",
|
579 |
-
"lstrip": true,
|
580 |
-
"normalized": true,
|
581 |
-
"rstrip": false,
|
582 |
-
"single_word": false
|
583 |
-
},
|
584 |
-
{
|
585 |
-
"content": "<extra_id_16>",
|
586 |
-
"lstrip": true,
|
587 |
-
"normalized": true,
|
588 |
-
"rstrip": false,
|
589 |
-
"single_word": false
|
590 |
-
},
|
591 |
-
{
|
592 |
-
"content": "<extra_id_15>",
|
593 |
-
"lstrip": true,
|
594 |
-
"normalized": true,
|
595 |
-
"rstrip": false,
|
596 |
-
"single_word": false
|
597 |
-
},
|
598 |
-
{
|
599 |
-
"content": "<extra_id_14>",
|
600 |
-
"lstrip": true,
|
601 |
-
"normalized": true,
|
602 |
-
"rstrip": false,
|
603 |
-
"single_word": false
|
604 |
-
},
|
605 |
-
{
|
606 |
-
"content": "<extra_id_13>",
|
607 |
-
"lstrip": true,
|
608 |
-
"normalized": true,
|
609 |
-
"rstrip": false,
|
610 |
-
"single_word": false
|
611 |
-
},
|
612 |
-
{
|
613 |
-
"content": "<extra_id_12>",
|
614 |
-
"lstrip": true,
|
615 |
-
"normalized": true,
|
616 |
-
"rstrip": false,
|
617 |
-
"single_word": false
|
618 |
-
},
|
619 |
-
{
|
620 |
-
"content": "<extra_id_11>",
|
621 |
-
"lstrip": true,
|
622 |
-
"normalized": true,
|
623 |
-
"rstrip": false,
|
624 |
-
"single_word": false
|
625 |
-
},
|
626 |
-
{
|
627 |
-
"content": "<extra_id_10>",
|
628 |
-
"lstrip": true,
|
629 |
-
"normalized": true,
|
630 |
-
"rstrip": false,
|
631 |
-
"single_word": false
|
632 |
-
},
|
633 |
-
{
|
634 |
-
"content": "<extra_id_9>",
|
635 |
-
"lstrip": true,
|
636 |
-
"normalized": true,
|
637 |
-
"rstrip": false,
|
638 |
-
"single_word": false
|
639 |
-
},
|
640 |
-
{
|
641 |
-
"content": "<extra_id_8>",
|
642 |
-
"lstrip": true,
|
643 |
-
"normalized": true,
|
644 |
-
"rstrip": false,
|
645 |
-
"single_word": false
|
646 |
-
},
|
647 |
-
{
|
648 |
-
"content": "<extra_id_7>",
|
649 |
-
"lstrip": true,
|
650 |
-
"normalized": true,
|
651 |
-
"rstrip": false,
|
652 |
-
"single_word": false
|
653 |
-
},
|
654 |
-
{
|
655 |
-
"content": "<extra_id_6>",
|
656 |
-
"lstrip": true,
|
657 |
-
"normalized": true,
|
658 |
-
"rstrip": false,
|
659 |
-
"single_word": false
|
660 |
-
},
|
661 |
-
{
|
662 |
-
"content": "<extra_id_5>",
|
663 |
-
"lstrip": true,
|
664 |
-
"normalized": true,
|
665 |
-
"rstrip": false,
|
666 |
-
"single_word": false
|
667 |
-
},
|
668 |
-
{
|
669 |
-
"content": "<extra_id_4>",
|
670 |
-
"lstrip": true,
|
671 |
-
"normalized": true,
|
672 |
-
"rstrip": false,
|
673 |
-
"single_word": false
|
674 |
-
},
|
675 |
-
{
|
676 |
-
"content": "<extra_id_3>",
|
677 |
-
"lstrip": true,
|
678 |
-
"normalized": true,
|
679 |
-
"rstrip": false,
|
680 |
-
"single_word": false
|
681 |
-
},
|
682 |
-
{
|
683 |
-
"content": "<extra_id_2>",
|
684 |
-
"lstrip": true,
|
685 |
-
"normalized": true,
|
686 |
-
"rstrip": false,
|
687 |
-
"single_word": false
|
688 |
-
},
|
689 |
-
{
|
690 |
-
"content": "<extra_id_1>",
|
691 |
-
"lstrip": true,
|
692 |
-
"normalized": true,
|
693 |
-
"rstrip": false,
|
694 |
-
"single_word": false
|
695 |
-
},
|
696 |
-
{
|
697 |
-
"content": "<extra_id_0>",
|
698 |
-
"lstrip": true,
|
699 |
-
"normalized": true,
|
700 |
-
"rstrip": false,
|
701 |
-
"single_word": false
|
702 |
-
}
|
703 |
-
],
|
704 |
-
"bos_token": {
|
705 |
-
"content": "<s>",
|
706 |
-
"lstrip": false,
|
707 |
-
"normalized": true,
|
708 |
-
"rstrip": false,
|
709 |
-
"single_word": false
|
710 |
-
},
|
711 |
-
"cls_token": {
|
712 |
-
"content": "<s>",
|
713 |
-
"lstrip": false,
|
714 |
-
"normalized": true,
|
715 |
-
"rstrip": false,
|
716 |
-
"single_word": false
|
717 |
-
},
|
718 |
-
"eos_token": {
|
719 |
-
"content": "</s>",
|
720 |
-
"lstrip": false,
|
721 |
-
"normalized": true,
|
722 |
-
"rstrip": false,
|
723 |
-
"single_word": false
|
724 |
-
},
|
725 |
-
"mask_token": {
|
726 |
-
"content": "<mask>",
|
727 |
-
"lstrip": true,
|
728 |
-
"normalized": true,
|
729 |
-
"rstrip": false,
|
730 |
-
"single_word": false
|
731 |
-
},
|
732 |
-
"pad_token": {
|
733 |
-
"content": "<pad>",
|
734 |
-
"lstrip": false,
|
735 |
-
"normalized": true,
|
736 |
-
"rstrip": false,
|
737 |
-
"single_word": false
|
738 |
-
},
|
739 |
-
"sep_token": {
|
740 |
-
"content": "</s>",
|
741 |
-
"lstrip": false,
|
742 |
-
"normalized": true,
|
743 |
-
"rstrip": false,
|
744 |
-
"single_word": false
|
745 |
-
},
|
746 |
-
"unk_token": {
|
747 |
-
"content": "<unk>",
|
748 |
-
"lstrip": false,
|
749 |
-
"normalized": true,
|
750 |
-
"rstrip": false,
|
751 |
-
"single_word": false
|
752 |
-
}
|
753 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
last-checkpoint/tokenizer.json
DELETED
The diff for this file is too large to render.
See raw diff
|
|
last-checkpoint/tokenizer_config.json
DELETED
@@ -1,64 +0,0 @@
|
|
1 |
-
{
|
2 |
-
"add_prefix_space": false,
|
3 |
-
"bos_token": {
|
4 |
-
"__type": "AddedToken",
|
5 |
-
"content": "<s>",
|
6 |
-
"lstrip": false,
|
7 |
-
"normalized": true,
|
8 |
-
"rstrip": false,
|
9 |
-
"single_word": false
|
10 |
-
},
|
11 |
-
"cls_token": {
|
12 |
-
"__type": "AddedToken",
|
13 |
-
"content": "<s>",
|
14 |
-
"lstrip": false,
|
15 |
-
"normalized": true,
|
16 |
-
"rstrip": false,
|
17 |
-
"single_word": false
|
18 |
-
},
|
19 |
-
"eos_token": {
|
20 |
-
"__type": "AddedToken",
|
21 |
-
"content": "</s>",
|
22 |
-
"lstrip": false,
|
23 |
-
"normalized": true,
|
24 |
-
"rstrip": false,
|
25 |
-
"single_word": false
|
26 |
-
},
|
27 |
-
"errors": "replace",
|
28 |
-
"mask_token": {
|
29 |
-
"__type": "AddedToken",
|
30 |
-
"content": "<mask>",
|
31 |
-
"lstrip": true,
|
32 |
-
"normalized": true,
|
33 |
-
"rstrip": false,
|
34 |
-
"single_word": false
|
35 |
-
},
|
36 |
-
"model_max_length": 512,
|
37 |
-
"pad_token": {
|
38 |
-
"__type": "AddedToken",
|
39 |
-
"content": "<pad>",
|
40 |
-
"lstrip": false,
|
41 |
-
"normalized": true,
|
42 |
-
"rstrip": false,
|
43 |
-
"single_word": false
|
44 |
-
},
|
45 |
-
"sep_token": {
|
46 |
-
"__type": "AddedToken",
|
47 |
-
"content": "</s>",
|
48 |
-
"lstrip": false,
|
49 |
-
"normalized": true,
|
50 |
-
"rstrip": false,
|
51 |
-
"single_word": false
|
52 |
-
},
|
53 |
-
"special_tokens_map_file": "/home/s5431786/.cache/huggingface/hub/models--Salesforce--codet5-small/snapshots/a642dc934e5475185369d09ac07091dfe72a31fc/special_tokens_map.json",
|
54 |
-
"tokenizer_class": "RobertaTokenizer",
|
55 |
-
"trim_offsets": true,
|
56 |
-
"unk_token": {
|
57 |
-
"__type": "AddedToken",
|
58 |
-
"content": "<unk>",
|
59 |
-
"lstrip": false,
|
60 |
-
"normalized": true,
|
61 |
-
"rstrip": false,
|
62 |
-
"single_word": false
|
63 |
-
}
|
64 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
last-checkpoint/trainer_state.json
DELETED
@@ -1,336 +0,0 @@
|
|
1 |
-
{
|
2 |
-
"best_metric": 0.1625,
|
3 |
-
"best_model_checkpoint": "/home2/s5431786/jb-internship/results/Salesforce-codet5-small-CodeXGLUE-CONCODE-adafactor/checkpoint-5500",
|
4 |
-
"epoch": 2.56,
|
5 |
-
"global_step": 8000,
|
6 |
-
"is_hyper_param_search": false,
|
7 |
-
"is_local_process_zero": true,
|
8 |
-
"is_world_process_zero": true,
|
9 |
-
"log_history": [
|
10 |
-
{
|
11 |
-
"epoch": 0.16,
|
12 |
-
"learning_rate": 9.520153550863722e-05,
|
13 |
-
"loss": 1.8525,
|
14 |
-
"step": 500
|
15 |
-
},
|
16 |
-
{
|
17 |
-
"epoch": 0.16,
|
18 |
-
"eval_bleu": 0.11457364536968197,
|
19 |
-
"eval_exact_match": 0.1435,
|
20 |
-
"eval_loss": 0.9340186715126038,
|
21 |
-
"eval_rouge1": 0.535967189457579,
|
22 |
-
"eval_rouge2": 0.35956007322808925,
|
23 |
-
"eval_rougeL": 0.5170899362207145,
|
24 |
-
"eval_rougeLsum": 0.5238094743675883,
|
25 |
-
"eval_runtime": 29.014,
|
26 |
-
"eval_samples_per_second": 68.932,
|
27 |
-
"eval_steps_per_second": 2.171,
|
28 |
-
"step": 500
|
29 |
-
},
|
30 |
-
{
|
31 |
-
"epoch": 0.32,
|
32 |
-
"learning_rate": 0.0001911708253358925,
|
33 |
-
"loss": 0.8679,
|
34 |
-
"step": 1000
|
35 |
-
},
|
36 |
-
{
|
37 |
-
"epoch": 0.32,
|
38 |
-
"eval_bleu": 0.1179014373884581,
|
39 |
-
"eval_exact_match": 0.1405,
|
40 |
-
"eval_loss": 0.9261881709098816,
|
41 |
-
"eval_rouge1": 0.5385361620034363,
|
42 |
-
"eval_rouge2": 0.36592311444869463,
|
43 |
-
"eval_rougeL": 0.5227558913580775,
|
44 |
-
"eval_rougeLsum": 0.5294207134026927,
|
45 |
-
"eval_runtime": 27.8315,
|
46 |
-
"eval_samples_per_second": 71.861,
|
47 |
-
"eval_steps_per_second": 2.264,
|
48 |
-
"step": 1000
|
49 |
-
},
|
50 |
-
{
|
51 |
-
"epoch": 0.48,
|
52 |
-
"learning_rate": 0.00028714011516314775,
|
53 |
-
"loss": 0.8169,
|
54 |
-
"step": 1500
|
55 |
-
},
|
56 |
-
{
|
57 |
-
"epoch": 0.48,
|
58 |
-
"eval_bleu": 0.11346624224407494,
|
59 |
-
"eval_exact_match": 0.139,
|
60 |
-
"eval_loss": 0.8957274556159973,
|
61 |
-
"eval_rouge1": 0.5371642651583028,
|
62 |
-
"eval_rouge2": 0.3641758327802227,
|
63 |
-
"eval_rougeL": 0.5191568868750955,
|
64 |
-
"eval_rougeLsum": 0.5265145916423253,
|
65 |
-
"eval_runtime": 27.7136,
|
66 |
-
"eval_samples_per_second": 72.167,
|
67 |
-
"eval_steps_per_second": 2.273,
|
68 |
-
"step": 1500
|
69 |
-
},
|
70 |
-
{
|
71 |
-
"epoch": 0.64,
|
72 |
-
"learning_rate": 0.00029562434735742914,
|
73 |
-
"loss": 0.7734,
|
74 |
-
"step": 2000
|
75 |
-
},
|
76 |
-
{
|
77 |
-
"epoch": 0.64,
|
78 |
-
"eval_bleu": 0.12095932329037072,
|
79 |
-
"eval_exact_match": 0.14,
|
80 |
-
"eval_loss": 0.8826662302017212,
|
81 |
-
"eval_rouge1": 0.5485190685904093,
|
82 |
-
"eval_rouge2": 0.37058117341608576,
|
83 |
-
"eval_rougeL": 0.5316412079787669,
|
84 |
-
"eval_rougeLsum": 0.538117520228076,
|
85 |
-
"eval_runtime": 30.1032,
|
86 |
-
"eval_samples_per_second": 66.438,
|
87 |
-
"eval_steps_per_second": 2.093,
|
88 |
-
"step": 2000
|
89 |
-
},
|
90 |
-
{
|
91 |
-
"epoch": 0.8,
|
92 |
-
"learning_rate": 0.00029057163068009565,
|
93 |
-
"loss": 0.743,
|
94 |
-
"step": 2500
|
95 |
-
},
|
96 |
-
{
|
97 |
-
"epoch": 0.8,
|
98 |
-
"eval_bleu": 0.11839616015657912,
|
99 |
-
"eval_exact_match": 0.155,
|
100 |
-
"eval_loss": 0.8646820187568665,
|
101 |
-
"eval_rouge1": 0.5502879388127755,
|
102 |
-
"eval_rouge2": 0.3832884278972679,
|
103 |
-
"eval_rougeL": 0.5338427860453392,
|
104 |
-
"eval_rougeLsum": 0.5410776354786306,
|
105 |
-
"eval_runtime": 27.8422,
|
106 |
-
"eval_samples_per_second": 71.833,
|
107 |
-
"eval_steps_per_second": 2.263,
|
108 |
-
"step": 2500
|
109 |
-
},
|
110 |
-
{
|
111 |
-
"epoch": 0.96,
|
112 |
-
"learning_rate": 0.00028551891400276216,
|
113 |
-
"loss": 0.72,
|
114 |
-
"step": 3000
|
115 |
-
},
|
116 |
-
{
|
117 |
-
"epoch": 0.96,
|
118 |
-
"eval_bleu": 0.11624138900158498,
|
119 |
-
"eval_exact_match": 0.1545,
|
120 |
-
"eval_loss": 0.8660675883293152,
|
121 |
-
"eval_rouge1": 0.5459797370761519,
|
122 |
-
"eval_rouge2": 0.3734931096135386,
|
123 |
-
"eval_rougeL": 0.5284301844438182,
|
124 |
-
"eval_rougeLsum": 0.5365915668092714,
|
125 |
-
"eval_runtime": 27.4892,
|
126 |
-
"eval_samples_per_second": 72.756,
|
127 |
-
"eval_steps_per_second": 2.292,
|
128 |
-
"step": 3000
|
129 |
-
},
|
130 |
-
{
|
131 |
-
"epoch": 1.12,
|
132 |
-
"learning_rate": 0.0002804661973254286,
|
133 |
-
"loss": 0.6539,
|
134 |
-
"step": 3500
|
135 |
-
},
|
136 |
-
{
|
137 |
-
"epoch": 1.12,
|
138 |
-
"eval_bleu": 0.12410699248804274,
|
139 |
-
"eval_exact_match": 0.156,
|
140 |
-
"eval_loss": 0.8591290712356567,
|
141 |
-
"eval_rouge1": 0.5540021216111264,
|
142 |
-
"eval_rouge2": 0.3840666487555394,
|
143 |
-
"eval_rougeL": 0.5364934994830287,
|
144 |
-
"eval_rougeLsum": 0.5443633030705,
|
145 |
-
"eval_runtime": 27.3261,
|
146 |
-
"eval_samples_per_second": 73.19,
|
147 |
-
"eval_steps_per_second": 2.305,
|
148 |
-
"step": 3500
|
149 |
-
},
|
150 |
-
{
|
151 |
-
"epoch": 1.28,
|
152 |
-
"learning_rate": 0.0002754134806480951,
|
153 |
-
"loss": 0.6301,
|
154 |
-
"step": 4000
|
155 |
-
},
|
156 |
-
{
|
157 |
-
"epoch": 1.28,
|
158 |
-
"eval_bleu": 0.12368101556382317,
|
159 |
-
"eval_exact_match": 0.1485,
|
160 |
-
"eval_loss": 0.845192015171051,
|
161 |
-
"eval_rouge1": 0.5556254539606234,
|
162 |
-
"eval_rouge2": 0.37940625998309263,
|
163 |
-
"eval_rougeL": 0.5369165919707064,
|
164 |
-
"eval_rougeLsum": 0.5450804739557871,
|
165 |
-
"eval_runtime": 27.8735,
|
166 |
-
"eval_samples_per_second": 71.753,
|
167 |
-
"eval_steps_per_second": 2.26,
|
168 |
-
"step": 4000
|
169 |
-
},
|
170 |
-
{
|
171 |
-
"epoch": 1.44,
|
172 |
-
"learning_rate": 0.0002703607639707616,
|
173 |
-
"loss": 0.6222,
|
174 |
-
"step": 4500
|
175 |
-
},
|
176 |
-
{
|
177 |
-
"epoch": 1.44,
|
178 |
-
"eval_bleu": 0.1227592648049904,
|
179 |
-
"eval_exact_match": 0.1585,
|
180 |
-
"eval_loss": 0.8321207165718079,
|
181 |
-
"eval_rouge1": 0.552933085993456,
|
182 |
-
"eval_rouge2": 0.38183507933347527,
|
183 |
-
"eval_rougeL": 0.5342751700599917,
|
184 |
-
"eval_rougeLsum": 0.5430113032214876,
|
185 |
-
"eval_runtime": 27.8567,
|
186 |
-
"eval_samples_per_second": 71.796,
|
187 |
-
"eval_steps_per_second": 2.262,
|
188 |
-
"step": 4500
|
189 |
-
},
|
190 |
-
{
|
191 |
-
"epoch": 1.6,
|
192 |
-
"learning_rate": 0.0002653080472934281,
|
193 |
-
"loss": 0.6221,
|
194 |
-
"step": 5000
|
195 |
-
},
|
196 |
-
{
|
197 |
-
"epoch": 1.6,
|
198 |
-
"eval_bleu": 0.12894039356912224,
|
199 |
-
"eval_exact_match": 0.154,
|
200 |
-
"eval_loss": 0.8316736817359924,
|
201 |
-
"eval_rouge1": 0.5664098623496682,
|
202 |
-
"eval_rouge2": 0.3925326485160906,
|
203 |
-
"eval_rougeL": 0.5480898996993683,
|
204 |
-
"eval_rougeLsum": 0.5574723438268435,
|
205 |
-
"eval_runtime": 28.8539,
|
206 |
-
"eval_samples_per_second": 69.315,
|
207 |
-
"eval_steps_per_second": 2.183,
|
208 |
-
"step": 5000
|
209 |
-
},
|
210 |
-
{
|
211 |
-
"epoch": 1.76,
|
212 |
-
"learning_rate": 0.00026025533061609454,
|
213 |
-
"loss": 0.6067,
|
214 |
-
"step": 5500
|
215 |
-
},
|
216 |
-
{
|
217 |
-
"epoch": 1.76,
|
218 |
-
"eval_bleu": 0.12994283554374192,
|
219 |
-
"eval_exact_match": 0.1625,
|
220 |
-
"eval_loss": 0.8228235840797424,
|
221 |
-
"eval_rouge1": 0.5606575988767198,
|
222 |
-
"eval_rouge2": 0.3932811205822008,
|
223 |
-
"eval_rougeL": 0.5437556306937155,
|
224 |
-
"eval_rougeLsum": 0.551586114580717,
|
225 |
-
"eval_runtime": 31.8078,
|
226 |
-
"eval_samples_per_second": 62.878,
|
227 |
-
"eval_steps_per_second": 1.981,
|
228 |
-
"step": 5500
|
229 |
-
},
|
230 |
-
{
|
231 |
-
"epoch": 1.92,
|
232 |
-
"learning_rate": 0.00025520261393876105,
|
233 |
-
"loss": 0.5927,
|
234 |
-
"step": 6000
|
235 |
-
},
|
236 |
-
{
|
237 |
-
"epoch": 1.92,
|
238 |
-
"eval_bleu": 0.13085325255281138,
|
239 |
-
"eval_exact_match": 0.156,
|
240 |
-
"eval_loss": 0.8179184198379517,
|
241 |
-
"eval_rouge1": 0.5625374697994676,
|
242 |
-
"eval_rouge2": 0.3942479004603814,
|
243 |
-
"eval_rougeL": 0.5457135058314136,
|
244 |
-
"eval_rougeLsum": 0.5526457174784051,
|
245 |
-
"eval_runtime": 27.729,
|
246 |
-
"eval_samples_per_second": 72.127,
|
247 |
-
"eval_steps_per_second": 2.272,
|
248 |
-
"step": 6000
|
249 |
-
},
|
250 |
-
{
|
251 |
-
"epoch": 2.08,
|
252 |
-
"learning_rate": 0.0002501498972614275,
|
253 |
-
"loss": 0.5548,
|
254 |
-
"step": 6500
|
255 |
-
},
|
256 |
-
{
|
257 |
-
"epoch": 2.08,
|
258 |
-
"eval_bleu": 0.13138618917381356,
|
259 |
-
"eval_exact_match": 0.152,
|
260 |
-
"eval_loss": 0.8258621096611023,
|
261 |
-
"eval_rouge1": 0.5581787363181516,
|
262 |
-
"eval_rouge2": 0.3846107993960324,
|
263 |
-
"eval_rougeL": 0.5401926988130337,
|
264 |
-
"eval_rougeLsum": 0.5485400951405117,
|
265 |
-
"eval_runtime": 31.3071,
|
266 |
-
"eval_samples_per_second": 63.883,
|
267 |
-
"eval_steps_per_second": 2.012,
|
268 |
-
"step": 6500
|
269 |
-
},
|
270 |
-
{
|
271 |
-
"epoch": 2.24,
|
272 |
-
"learning_rate": 0.000245097180584094,
|
273 |
-
"loss": 0.5146,
|
274 |
-
"step": 7000
|
275 |
-
},
|
276 |
-
{
|
277 |
-
"epoch": 2.24,
|
278 |
-
"eval_bleu": 0.1298944384856821,
|
279 |
-
"eval_exact_match": 0.1545,
|
280 |
-
"eval_loss": 0.8328122496604919,
|
281 |
-
"eval_rouge1": 0.5605117850042978,
|
282 |
-
"eval_rouge2": 0.38890024128220335,
|
283 |
-
"eval_rougeL": 0.5428954337612537,
|
284 |
-
"eval_rougeLsum": 0.5514352009091111,
|
285 |
-
"eval_runtime": 27.4801,
|
286 |
-
"eval_samples_per_second": 72.78,
|
287 |
-
"eval_steps_per_second": 2.293,
|
288 |
-
"step": 7000
|
289 |
-
},
|
290 |
-
{
|
291 |
-
"epoch": 2.4,
|
292 |
-
"learning_rate": 0.00024004446390676052,
|
293 |
-
"loss": 0.5193,
|
294 |
-
"step": 7500
|
295 |
-
},
|
296 |
-
{
|
297 |
-
"epoch": 2.4,
|
298 |
-
"eval_bleu": 0.12682500743266936,
|
299 |
-
"eval_exact_match": 0.1555,
|
300 |
-
"eval_loss": 0.8196701407432556,
|
301 |
-
"eval_rouge1": 0.560439154770211,
|
302 |
-
"eval_rouge2": 0.38664585480336755,
|
303 |
-
"eval_rougeL": 0.5431387975592847,
|
304 |
-
"eval_rougeLsum": 0.5501216456745461,
|
305 |
-
"eval_runtime": 28.6733,
|
306 |
-
"eval_samples_per_second": 69.751,
|
307 |
-
"eval_steps_per_second": 2.197,
|
308 |
-
"step": 7500
|
309 |
-
},
|
310 |
-
{
|
311 |
-
"epoch": 2.56,
|
312 |
-
"learning_rate": 0.00023499174722942698,
|
313 |
-
"loss": 0.5172,
|
314 |
-
"step": 8000
|
315 |
-
},
|
316 |
-
{
|
317 |
-
"epoch": 2.56,
|
318 |
-
"eval_bleu": 0.12967243587670185,
|
319 |
-
"eval_exact_match": 0.1555,
|
320 |
-
"eval_loss": 0.8118115663528442,
|
321 |
-
"eval_rouge1": 0.5579744246548106,
|
322 |
-
"eval_rouge2": 0.3885609120793808,
|
323 |
-
"eval_rougeL": 0.5406865803442568,
|
324 |
-
"eval_rougeLsum": 0.5482646526079711,
|
325 |
-
"eval_runtime": 27.397,
|
326 |
-
"eval_samples_per_second": 73.001,
|
327 |
-
"eval_steps_per_second": 2.3,
|
328 |
-
"step": 8000
|
329 |
-
}
|
330 |
-
],
|
331 |
-
"max_steps": 31250,
|
332 |
-
"num_train_epochs": 10,
|
333 |
-
"total_flos": 3.4018432483590144e+16,
|
334 |
-
"trial_name": null,
|
335 |
-
"trial_params": null
|
336 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
last-checkpoint/training_args.bin
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:530b4d56ccb226f5a70f2406c4ad4f897f4383edbb30bafb24d8c4f1987bfe51
|
3 |
-
size 3887
|
|
|
|
|
|
|
|
last-checkpoint/vocab.json
DELETED
The diff for this file is too large to render.
See raw diff
|
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 242012923
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d40dea835e2b549834368947d7e5d878efa90dd132fb8d24c1c28828263a6ae0
|
3 |
size 242012923
|