malmarjeh commited on
Commit
64ce611
1 Parent(s): 8281c73

add model files

Browse files
.gitattributes CHANGED
@@ -25,3 +25,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
25
  *.zip filter=lfs diff=lfs merge=lfs -text
26
  *.zstandard filter=lfs diff=lfs merge=lfs -text
27
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
25
  *.zip filter=lfs diff=lfs merge=lfs -text
26
  *.zstandard filter=lfs diff=lfs merge=lfs -text
27
  *tfevents* filter=lfs diff=lfs merge=lfs -text
28
+ pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "drive/MyDrive/AraT5_FT_title_generation/checkpoint-49000",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 2048,
7
+ "d_kv": 64,
8
+ "d_model": 768,
9
+ "decoder_start_token_id": 0,
10
+ "dropout_rate": 0.1,
11
+ "eos_token_id": 1,
12
+ "feed_forward_proj": "gated-gelu",
13
+ "gradient_checkpointing": false,
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "layer_norm_epsilon": 1e-06,
17
+ "model_type": "t5",
18
+ "num_decoder_layers": 12,
19
+ "num_heads": 12,
20
+ "num_layers": 12,
21
+ "output_past": true,
22
+ "pad_token_id": 0,
23
+ "relative_attention_num_buckets": 32,
24
+ "transformers_version": "4.5.1",
25
+ "use_cache": true,
26
+ "vocab_size": 110080
27
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f141ddeb0f45a7f0ca21c355a8657a66586908ff90bd01edbf476477964377f
3
+ size 1131209487
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c69037668b4aa26a6f50b4ef2fec1c191fe7d48648a2e72918fc57cb87cad23b
3
+ size 623
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c7ae4407451bf02b459edec774d0539a06615005dd34d5c85c5c06765ff1606
3
+ size 2435308
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 0, "additional_special_tokens": null, "special_tokens_map_file": "/tmp/AraT5_cache_dir/3b52be683fa2fd36463d04d51871966137133d8c4fc7663ff2a20cb4f3549c18.b02fba13e4584f40329b20be0bad88c2ef97b5fda7d9f4f69b1cb3a620568a86", "name_or_path": "UBC-NLP/AraT5-base"}
trainer_state.json ADDED
@@ -0,0 +1,1006 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 23.7361,
3
+ "best_model_checkpoint": "drive/MyDrive/AraT5_FT_title_generation/checkpoint-64000",
4
+ "epoch": 6.963494408102975,
5
+ "global_step": 66000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.11,
12
+ "learning_rate": 1.6666666666666667e-05,
13
+ "loss": 35.638,
14
+ "step": 1000
15
+ },
16
+ {
17
+ "epoch": 0.11,
18
+ "eval_bleu": 0.0,
19
+ "eval_gen_len": 18.823,
20
+ "eval_loss": 8.983467102050781,
21
+ "eval_runtime": 261.594,
22
+ "eval_samples_per_second": 3.823,
23
+ "step": 1000
24
+ },
25
+ {
26
+ "epoch": 0.21,
27
+ "learning_rate": 3.3333333333333335e-05,
28
+ "loss": 8.3857,
29
+ "step": 2000
30
+ },
31
+ {
32
+ "epoch": 0.21,
33
+ "eval_bleu": 0.198,
34
+ "eval_gen_len": 12.74,
35
+ "eval_loss": 7.130731105804443,
36
+ "eval_runtime": 245.9993,
37
+ "eval_samples_per_second": 4.065,
38
+ "step": 2000
39
+ },
40
+ {
41
+ "epoch": 0.32,
42
+ "learning_rate": 5e-05,
43
+ "loss": 6.9885,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.32,
48
+ "eval_bleu": 0.8955,
49
+ "eval_gen_len": 14.538,
50
+ "eval_loss": 5.843496799468994,
51
+ "eval_runtime": 253.3806,
52
+ "eval_samples_per_second": 3.947,
53
+ "step": 3000
54
+ },
55
+ {
56
+ "epoch": 0.42,
57
+ "learning_rate": 4.97567099398587e-05,
58
+ "loss": 5.9845,
59
+ "step": 4000
60
+ },
61
+ {
62
+ "epoch": 0.42,
63
+ "eval_bleu": 2.1258,
64
+ "eval_gen_len": 12.856,
65
+ "eval_loss": 4.96659517288208,
66
+ "eval_runtime": 239.9357,
67
+ "eval_samples_per_second": 4.168,
68
+ "step": 4000
69
+ },
70
+ {
71
+ "epoch": 0.53,
72
+ "learning_rate": 4.95134198797174e-05,
73
+ "loss": 5.3511,
74
+ "step": 5000
75
+ },
76
+ {
77
+ "epoch": 0.53,
78
+ "eval_bleu": 4.529,
79
+ "eval_gen_len": 13.551,
80
+ "eval_loss": 4.428027153015137,
81
+ "eval_runtime": 244.6554,
82
+ "eval_samples_per_second": 4.087,
83
+ "step": 5000
84
+ },
85
+ {
86
+ "epoch": 0.63,
87
+ "learning_rate": 4.92701298195761e-05,
88
+ "loss": 4.8601,
89
+ "step": 6000
90
+ },
91
+ {
92
+ "epoch": 0.63,
93
+ "eval_bleu": 7.2219,
94
+ "eval_gen_len": 12.194,
95
+ "eval_loss": 4.005715370178223,
96
+ "eval_runtime": 190.607,
97
+ "eval_samples_per_second": 5.246,
98
+ "step": 6000
99
+ },
100
+ {
101
+ "epoch": 0.74,
102
+ "learning_rate": 4.902683975943479e-05,
103
+ "loss": 4.4955,
104
+ "step": 7000
105
+ },
106
+ {
107
+ "epoch": 0.74,
108
+ "eval_bleu": 9.4889,
109
+ "eval_gen_len": 12.137,
110
+ "eval_loss": 3.741957664489746,
111
+ "eval_runtime": 185.7078,
112
+ "eval_samples_per_second": 5.385,
113
+ "step": 7000
114
+ },
115
+ {
116
+ "epoch": 0.84,
117
+ "learning_rate": 4.878354969929349e-05,
118
+ "loss": 4.2481,
119
+ "step": 8000
120
+ },
121
+ {
122
+ "epoch": 0.84,
123
+ "eval_bleu": 11.3848,
124
+ "eval_gen_len": 11.773,
125
+ "eval_loss": 3.5194664001464844,
126
+ "eval_runtime": 180.2345,
127
+ "eval_samples_per_second": 5.548,
128
+ "step": 8000
129
+ },
130
+ {
131
+ "epoch": 0.95,
132
+ "learning_rate": 4.8540259639152185e-05,
133
+ "loss": 4.0054,
134
+ "step": 9000
135
+ },
136
+ {
137
+ "epoch": 0.95,
138
+ "eval_bleu": 13.8757,
139
+ "eval_gen_len": 11.671,
140
+ "eval_loss": 3.378380298614502,
141
+ "eval_runtime": 179.378,
142
+ "eval_samples_per_second": 5.575,
143
+ "step": 9000
144
+ },
145
+ {
146
+ "epoch": 1.06,
147
+ "learning_rate": 4.829696957901088e-05,
148
+ "loss": 3.8032,
149
+ "step": 10000
150
+ },
151
+ {
152
+ "epoch": 1.06,
153
+ "eval_bleu": 15.1716,
154
+ "eval_gen_len": 11.402,
155
+ "eval_loss": 3.2721943855285645,
156
+ "eval_runtime": 176.9812,
157
+ "eval_samples_per_second": 5.65,
158
+ "step": 10000
159
+ },
160
+ {
161
+ "epoch": 1.16,
162
+ "learning_rate": 4.805367951886958e-05,
163
+ "loss": 3.6668,
164
+ "step": 11000
165
+ },
166
+ {
167
+ "epoch": 1.16,
168
+ "eval_bleu": 15.1204,
169
+ "eval_gen_len": 11.792,
170
+ "eval_loss": 3.1800124645233154,
171
+ "eval_runtime": 179.9992,
172
+ "eval_samples_per_second": 5.556,
173
+ "step": 11000
174
+ },
175
+ {
176
+ "epoch": 1.27,
177
+ "learning_rate": 4.781038945872828e-05,
178
+ "loss": 3.5526,
179
+ "step": 12000
180
+ },
181
+ {
182
+ "epoch": 1.27,
183
+ "eval_bleu": 15.4619,
184
+ "eval_gen_len": 12.077,
185
+ "eval_loss": 3.103393077850342,
186
+ "eval_runtime": 183.5811,
187
+ "eval_samples_per_second": 5.447,
188
+ "step": 12000
189
+ },
190
+ {
191
+ "epoch": 1.37,
192
+ "learning_rate": 4.756709939858697e-05,
193
+ "loss": 3.4665,
194
+ "step": 13000
195
+ },
196
+ {
197
+ "epoch": 1.37,
198
+ "eval_bleu": 16.7979,
199
+ "eval_gen_len": 11.766,
200
+ "eval_loss": 3.0600805282592773,
201
+ "eval_runtime": 183.4473,
202
+ "eval_samples_per_second": 5.451,
203
+ "step": 13000
204
+ },
205
+ {
206
+ "epoch": 1.48,
207
+ "learning_rate": 4.732380933844567e-05,
208
+ "loss": 3.3754,
209
+ "step": 14000
210
+ },
211
+ {
212
+ "epoch": 1.48,
213
+ "eval_bleu": 17.5953,
214
+ "eval_gen_len": 11.554,
215
+ "eval_loss": 2.9892418384552,
216
+ "eval_runtime": 183.2148,
217
+ "eval_samples_per_second": 5.458,
218
+ "step": 14000
219
+ },
220
+ {
221
+ "epoch": 1.58,
222
+ "learning_rate": 4.708051927830437e-05,
223
+ "loss": 3.3243,
224
+ "step": 15000
225
+ },
226
+ {
227
+ "epoch": 1.58,
228
+ "eval_bleu": 18.2111,
229
+ "eval_gen_len": 11.962,
230
+ "eval_loss": 2.9837777614593506,
231
+ "eval_runtime": 180.3526,
232
+ "eval_samples_per_second": 5.545,
233
+ "step": 15000
234
+ },
235
+ {
236
+ "epoch": 1.69,
237
+ "learning_rate": 4.6837229218163065e-05,
238
+ "loss": 3.2794,
239
+ "step": 16000
240
+ },
241
+ {
242
+ "epoch": 1.69,
243
+ "eval_bleu": 18.227,
244
+ "eval_gen_len": 11.463,
245
+ "eval_loss": 2.898597240447998,
246
+ "eval_runtime": 178.1798,
247
+ "eval_samples_per_second": 5.612,
248
+ "step": 16000
249
+ },
250
+ {
251
+ "epoch": 1.79,
252
+ "learning_rate": 4.659393915802176e-05,
253
+ "loss": 3.203,
254
+ "step": 17000
255
+ },
256
+ {
257
+ "epoch": 1.79,
258
+ "eval_bleu": 19.2324,
259
+ "eval_gen_len": 11.34,
260
+ "eval_loss": 2.8416123390197754,
261
+ "eval_runtime": 176.1105,
262
+ "eval_samples_per_second": 5.678,
263
+ "step": 17000
264
+ },
265
+ {
266
+ "epoch": 1.9,
267
+ "learning_rate": 4.635064909788046e-05,
268
+ "loss": 3.1658,
269
+ "step": 18000
270
+ },
271
+ {
272
+ "epoch": 1.9,
273
+ "eval_bleu": 19.0493,
274
+ "eval_gen_len": 11.414,
275
+ "eval_loss": 2.8081586360931396,
276
+ "eval_runtime": 178.4234,
277
+ "eval_samples_per_second": 5.605,
278
+ "step": 18000
279
+ },
280
+ {
281
+ "epoch": 2.0,
282
+ "learning_rate": 4.610735903773916e-05,
283
+ "loss": 3.0979,
284
+ "step": 19000
285
+ },
286
+ {
287
+ "epoch": 2.0,
288
+ "eval_bleu": 19.4019,
289
+ "eval_gen_len": 11.815,
290
+ "eval_loss": 2.801037311553955,
291
+ "eval_runtime": 184.5402,
292
+ "eval_samples_per_second": 5.419,
293
+ "step": 19000
294
+ },
295
+ {
296
+ "epoch": 2.11,
297
+ "learning_rate": 4.586406897759785e-05,
298
+ "loss": 2.963,
299
+ "step": 20000
300
+ },
301
+ {
302
+ "epoch": 2.11,
303
+ "eval_bleu": 20.4101,
304
+ "eval_gen_len": 10.937,
305
+ "eval_loss": 2.7728490829467773,
306
+ "eval_runtime": 216.7393,
307
+ "eval_samples_per_second": 4.614,
308
+ "step": 20000
309
+ },
310
+ {
311
+ "epoch": 2.22,
312
+ "learning_rate": 4.5620778917456556e-05,
313
+ "loss": 2.9022,
314
+ "step": 21000
315
+ },
316
+ {
317
+ "epoch": 2.22,
318
+ "eval_bleu": 20.2568,
319
+ "eval_gen_len": 11.498,
320
+ "eval_loss": 2.748673439025879,
321
+ "eval_runtime": 222.8526,
322
+ "eval_samples_per_second": 4.487,
323
+ "step": 21000
324
+ },
325
+ {
326
+ "epoch": 2.32,
327
+ "learning_rate": 4.537748885731525e-05,
328
+ "loss": 2.9113,
329
+ "step": 22000
330
+ },
331
+ {
332
+ "epoch": 2.32,
333
+ "eval_bleu": 20.7867,
334
+ "eval_gen_len": 11.143,
335
+ "eval_loss": 2.714132308959961,
336
+ "eval_runtime": 237.849,
337
+ "eval_samples_per_second": 4.204,
338
+ "step": 22000
339
+ },
340
+ {
341
+ "epoch": 2.43,
342
+ "learning_rate": 4.5134198797173945e-05,
343
+ "loss": 2.9026,
344
+ "step": 23000
345
+ },
346
+ {
347
+ "epoch": 2.43,
348
+ "eval_bleu": 20.6776,
349
+ "eval_gen_len": 11.403,
350
+ "eval_loss": 2.681997060775757,
351
+ "eval_runtime": 245.0406,
352
+ "eval_samples_per_second": 4.081,
353
+ "step": 23000
354
+ },
355
+ {
356
+ "epoch": 2.53,
357
+ "learning_rate": 4.4890908737032644e-05,
358
+ "loss": 2.8522,
359
+ "step": 24000
360
+ },
361
+ {
362
+ "epoch": 2.53,
363
+ "eval_bleu": 21.2223,
364
+ "eval_gen_len": 11.345,
365
+ "eval_loss": 2.6821537017822266,
366
+ "eval_runtime": 234.0719,
367
+ "eval_samples_per_second": 4.272,
368
+ "step": 24000
369
+ },
370
+ {
371
+ "epoch": 2.64,
372
+ "learning_rate": 4.464761867689134e-05,
373
+ "loss": 2.8177,
374
+ "step": 25000
375
+ },
376
+ {
377
+ "epoch": 2.64,
378
+ "eval_bleu": 21.1983,
379
+ "eval_gen_len": 11.272,
380
+ "eval_loss": 2.6571996212005615,
381
+ "eval_runtime": 244.6544,
382
+ "eval_samples_per_second": 4.087,
383
+ "step": 25000
384
+ },
385
+ {
386
+ "epoch": 2.74,
387
+ "learning_rate": 4.440432861675003e-05,
388
+ "loss": 2.8178,
389
+ "step": 26000
390
+ },
391
+ {
392
+ "epoch": 2.74,
393
+ "eval_bleu": 21.0805,
394
+ "eval_gen_len": 11.569,
395
+ "eval_loss": 2.647949695587158,
396
+ "eval_runtime": 246.8942,
397
+ "eval_samples_per_second": 4.05,
398
+ "step": 26000
399
+ },
400
+ {
401
+ "epoch": 2.85,
402
+ "learning_rate": 4.416103855660874e-05,
403
+ "loss": 2.7526,
404
+ "step": 27000
405
+ },
406
+ {
407
+ "epoch": 2.85,
408
+ "eval_bleu": 20.354,
409
+ "eval_gen_len": 11.81,
410
+ "eval_loss": 2.6451191902160645,
411
+ "eval_runtime": 254.1274,
412
+ "eval_samples_per_second": 3.935,
413
+ "step": 27000
414
+ },
415
+ {
416
+ "epoch": 2.95,
417
+ "learning_rate": 4.391774849646743e-05,
418
+ "loss": 2.783,
419
+ "step": 28000
420
+ },
421
+ {
422
+ "epoch": 2.95,
423
+ "eval_bleu": 21.1215,
424
+ "eval_gen_len": 11.437,
425
+ "eval_loss": 2.60164213180542,
426
+ "eval_runtime": 229.9337,
427
+ "eval_samples_per_second": 4.349,
428
+ "step": 28000
429
+ },
430
+ {
431
+ "epoch": 3.06,
432
+ "learning_rate": 4.367445843632613e-05,
433
+ "loss": 2.6665,
434
+ "step": 29000
435
+ },
436
+ {
437
+ "epoch": 3.06,
438
+ "eval_bleu": 20.9631,
439
+ "eval_gen_len": 11.499,
440
+ "eval_loss": 2.6079459190368652,
441
+ "eval_runtime": 230.5689,
442
+ "eval_samples_per_second": 4.337,
443
+ "step": 29000
444
+ },
445
+ {
446
+ "epoch": 3.17,
447
+ "learning_rate": 4.343116837618482e-05,
448
+ "loss": 2.6233,
449
+ "step": 30000
450
+ },
451
+ {
452
+ "epoch": 3.17,
453
+ "eval_bleu": 21.332,
454
+ "eval_gen_len": 11.468,
455
+ "eval_loss": 2.613509178161621,
456
+ "eval_runtime": 201.795,
457
+ "eval_samples_per_second": 4.956,
458
+ "step": 30000
459
+ },
460
+ {
461
+ "epoch": 3.27,
462
+ "learning_rate": 4.3187878316043524e-05,
463
+ "loss": 2.5983,
464
+ "step": 31000
465
+ },
466
+ {
467
+ "epoch": 3.27,
468
+ "eval_bleu": 21.0944,
469
+ "eval_gen_len": 11.603,
470
+ "eval_loss": 2.5642783641815186,
471
+ "eval_runtime": 201.0993,
472
+ "eval_samples_per_second": 4.973,
473
+ "step": 31000
474
+ },
475
+ {
476
+ "epoch": 3.38,
477
+ "learning_rate": 4.2944588255902215e-05,
478
+ "loss": 2.5912,
479
+ "step": 32000
480
+ },
481
+ {
482
+ "epoch": 3.38,
483
+ "eval_bleu": 21.6143,
484
+ "eval_gen_len": 11.104,
485
+ "eval_loss": 2.5584685802459717,
486
+ "eval_runtime": 198.7529,
487
+ "eval_samples_per_second": 5.031,
488
+ "step": 32000
489
+ },
490
+ {
491
+ "epoch": 3.48,
492
+ "learning_rate": 4.2701298195760914e-05,
493
+ "loss": 2.5681,
494
+ "step": 33000
495
+ },
496
+ {
497
+ "epoch": 3.48,
498
+ "eval_bleu": 21.8716,
499
+ "eval_gen_len": 10.987,
500
+ "eval_loss": 2.541090726852417,
501
+ "eval_runtime": 208.1375,
502
+ "eval_samples_per_second": 4.805,
503
+ "step": 33000
504
+ },
505
+ {
506
+ "epoch": 3.59,
507
+ "learning_rate": 4.245800813561961e-05,
508
+ "loss": 2.5657,
509
+ "step": 34000
510
+ },
511
+ {
512
+ "epoch": 3.59,
513
+ "eval_bleu": 21.7373,
514
+ "eval_gen_len": 11.364,
515
+ "eval_loss": 2.51957368850708,
516
+ "eval_runtime": 220.5477,
517
+ "eval_samples_per_second": 4.534,
518
+ "step": 34000
519
+ },
520
+ {
521
+ "epoch": 3.69,
522
+ "learning_rate": 4.221471807547831e-05,
523
+ "loss": 2.5639,
524
+ "step": 35000
525
+ },
526
+ {
527
+ "epoch": 3.69,
528
+ "eval_bleu": 21.9015,
529
+ "eval_gen_len": 11.395,
530
+ "eval_loss": 2.516514539718628,
531
+ "eval_runtime": 224.7434,
532
+ "eval_samples_per_second": 4.45,
533
+ "step": 35000
534
+ },
535
+ {
536
+ "epoch": 3.8,
537
+ "learning_rate": 4.197142801533701e-05,
538
+ "loss": 2.5257,
539
+ "step": 36000
540
+ },
541
+ {
542
+ "epoch": 3.8,
543
+ "eval_bleu": 21.7659,
544
+ "eval_gen_len": 11.47,
545
+ "eval_loss": 2.5282981395721436,
546
+ "eval_runtime": 248.968,
547
+ "eval_samples_per_second": 4.017,
548
+ "step": 36000
549
+ },
550
+ {
551
+ "epoch": 3.9,
552
+ "learning_rate": 4.1728137955195706e-05,
553
+ "loss": 2.5278,
554
+ "step": 37000
555
+ },
556
+ {
557
+ "epoch": 3.9,
558
+ "eval_bleu": 22.3318,
559
+ "eval_gen_len": 11.31,
560
+ "eval_loss": 2.5119266510009766,
561
+ "eval_runtime": 201.5867,
562
+ "eval_samples_per_second": 4.961,
563
+ "step": 37000
564
+ },
565
+ {
566
+ "epoch": 4.01,
567
+ "learning_rate": 4.1484847895054404e-05,
568
+ "loss": 2.5164,
569
+ "step": 38000
570
+ },
571
+ {
572
+ "epoch": 4.01,
573
+ "eval_bleu": 21.6403,
574
+ "eval_gen_len": 11.544,
575
+ "eval_loss": 2.5127675533294678,
576
+ "eval_runtime": 222.4942,
577
+ "eval_samples_per_second": 4.494,
578
+ "step": 38000
579
+ },
580
+ {
581
+ "epoch": 4.11,
582
+ "learning_rate": 4.1241557834913096e-05,
583
+ "loss": 2.3804,
584
+ "step": 39000
585
+ },
586
+ {
587
+ "epoch": 4.11,
588
+ "eval_bleu": 21.88,
589
+ "eval_gen_len": 11.368,
590
+ "eval_loss": 2.5173442363739014,
591
+ "eval_runtime": 231.6469,
592
+ "eval_samples_per_second": 4.317,
593
+ "step": 39000
594
+ },
595
+ {
596
+ "epoch": 4.22,
597
+ "learning_rate": 4.09982677747718e-05,
598
+ "loss": 2.4065,
599
+ "step": 40000
600
+ },
601
+ {
602
+ "epoch": 4.22,
603
+ "eval_bleu": 21.6251,
604
+ "eval_gen_len": 11.596,
605
+ "eval_loss": 2.481987714767456,
606
+ "eval_runtime": 224.3387,
607
+ "eval_samples_per_second": 4.458,
608
+ "step": 40000
609
+ },
610
+ {
611
+ "epoch": 4.33,
612
+ "learning_rate": 4.075497771463049e-05,
613
+ "loss": 2.3876,
614
+ "step": 41000
615
+ },
616
+ {
617
+ "epoch": 4.33,
618
+ "eval_bleu": 22.5464,
619
+ "eval_gen_len": 11.385,
620
+ "eval_loss": 2.49714994430542,
621
+ "eval_runtime": 204.1448,
622
+ "eval_samples_per_second": 4.898,
623
+ "step": 41000
624
+ },
625
+ {
626
+ "epoch": 4.43,
627
+ "learning_rate": 4.051168765448919e-05,
628
+ "loss": 2.3339,
629
+ "step": 42000
630
+ },
631
+ {
632
+ "epoch": 4.43,
633
+ "eval_bleu": 22.9928,
634
+ "eval_gen_len": 11.032,
635
+ "eval_loss": 2.4970240592956543,
636
+ "eval_runtime": 207.9008,
637
+ "eval_samples_per_second": 4.81,
638
+ "step": 42000
639
+ },
640
+ {
641
+ "epoch": 4.54,
642
+ "learning_rate": 4.026839759434789e-05,
643
+ "loss": 2.3397,
644
+ "step": 43000
645
+ },
646
+ {
647
+ "epoch": 4.54,
648
+ "eval_bleu": 22.1443,
649
+ "eval_gen_len": 11.523,
650
+ "eval_loss": 2.476402759552002,
651
+ "eval_runtime": 206.0363,
652
+ "eval_samples_per_second": 4.854,
653
+ "step": 43000
654
+ },
655
+ {
656
+ "epoch": 4.64,
657
+ "learning_rate": 4.0025107534206587e-05,
658
+ "loss": 2.3629,
659
+ "step": 44000
660
+ },
661
+ {
662
+ "epoch": 4.64,
663
+ "eval_bleu": 22.8804,
664
+ "eval_gen_len": 11.271,
665
+ "eval_loss": 2.455251455307007,
666
+ "eval_runtime": 206.5896,
667
+ "eval_samples_per_second": 4.841,
668
+ "step": 44000
669
+ },
670
+ {
671
+ "epoch": 4.75,
672
+ "learning_rate": 3.978181747406528e-05,
673
+ "loss": 2.3749,
674
+ "step": 45000
675
+ },
676
+ {
677
+ "epoch": 4.75,
678
+ "eval_bleu": 21.9768,
679
+ "eval_gen_len": 11.698,
680
+ "eval_loss": 2.4702677726745605,
681
+ "eval_runtime": 201.4429,
682
+ "eval_samples_per_second": 4.964,
683
+ "step": 45000
684
+ },
685
+ {
686
+ "epoch": 4.85,
687
+ "learning_rate": 3.9538527413923976e-05,
688
+ "loss": 2.3341,
689
+ "step": 46000
690
+ },
691
+ {
692
+ "epoch": 4.85,
693
+ "eval_bleu": 23.3467,
694
+ "eval_gen_len": 11.481,
695
+ "eval_loss": 2.4574341773986816,
696
+ "eval_runtime": 219.555,
697
+ "eval_samples_per_second": 4.555,
698
+ "step": 46000
699
+ },
700
+ {
701
+ "epoch": 4.96,
702
+ "learning_rate": 3.9295237353782674e-05,
703
+ "loss": 2.3405,
704
+ "step": 47000
705
+ },
706
+ {
707
+ "epoch": 4.96,
708
+ "eval_bleu": 22.2318,
709
+ "eval_gen_len": 11.215,
710
+ "eval_loss": 2.416428327560425,
711
+ "eval_runtime": 219.9781,
712
+ "eval_samples_per_second": 4.546,
713
+ "step": 47000
714
+ },
715
+ {
716
+ "epoch": 5.06,
717
+ "learning_rate": 3.905194729364137e-05,
718
+ "loss": 2.2188,
719
+ "step": 48000
720
+ },
721
+ {
722
+ "epoch": 5.06,
723
+ "eval_bleu": 22.4174,
724
+ "eval_gen_len": 11.296,
725
+ "eval_loss": 2.478271007537842,
726
+ "eval_runtime": 207.2586,
727
+ "eval_samples_per_second": 4.825,
728
+ "step": 48000
729
+ },
730
+ {
731
+ "epoch": 5.17,
732
+ "learning_rate": 3.880865723350007e-05,
733
+ "loss": 2.1582,
734
+ "step": 49000
735
+ },
736
+ {
737
+ "epoch": 5.17,
738
+ "eval_bleu": 22.9358,
739
+ "eval_gen_len": 11.209,
740
+ "eval_loss": 2.478005886077881,
741
+ "eval_runtime": 218.3957,
742
+ "eval_samples_per_second": 4.579,
743
+ "step": 49000
744
+ },
745
+ {
746
+ "epoch": 5.28,
747
+ "learning_rate": 3.856536717335877e-05,
748
+ "loss": 2.2462,
749
+ "step": 50000
750
+ },
751
+ {
752
+ "epoch": 5.28,
753
+ "eval_bleu": 22.4676,
754
+ "eval_gen_len": 11.412,
755
+ "eval_loss": 2.4692862033843994,
756
+ "eval_runtime": 129.2452,
757
+ "eval_samples_per_second": 7.737,
758
+ "step": 50000
759
+ },
760
+ {
761
+ "epoch": 5.38,
762
+ "learning_rate": 3.832207711321747e-05,
763
+ "loss": 2.2293,
764
+ "step": 51000
765
+ },
766
+ {
767
+ "epoch": 5.38,
768
+ "eval_bleu": 22.8528,
769
+ "eval_gen_len": 11.306,
770
+ "eval_loss": 2.4192593097686768,
771
+ "eval_runtime": 131.8082,
772
+ "eval_samples_per_second": 7.587,
773
+ "step": 51000
774
+ },
775
+ {
776
+ "epoch": 5.49,
777
+ "learning_rate": 3.807878705307616e-05,
778
+ "loss": 2.2323,
779
+ "step": 52000
780
+ },
781
+ {
782
+ "epoch": 5.49,
783
+ "eval_bleu": 23.4254,
784
+ "eval_gen_len": 11.5,
785
+ "eval_loss": 2.436190605163574,
786
+ "eval_runtime": 129.6391,
787
+ "eval_samples_per_second": 7.714,
788
+ "step": 52000
789
+ },
790
+ {
791
+ "epoch": 5.59,
792
+ "learning_rate": 3.783549699293486e-05,
793
+ "loss": 2.2434,
794
+ "step": 53000
795
+ },
796
+ {
797
+ "epoch": 5.59,
798
+ "eval_bleu": 23.0953,
799
+ "eval_gen_len": 11.579,
800
+ "eval_loss": 2.434554100036621,
801
+ "eval_runtime": 129.6268,
802
+ "eval_samples_per_second": 7.714,
803
+ "step": 53000
804
+ },
805
+ {
806
+ "epoch": 5.7,
807
+ "learning_rate": 3.7592206932793555e-05,
808
+ "loss": 2.2106,
809
+ "step": 54000
810
+ },
811
+ {
812
+ "epoch": 5.7,
813
+ "eval_bleu": 22.5385,
814
+ "eval_gen_len": 11.793,
815
+ "eval_loss": 2.4267661571502686,
816
+ "eval_runtime": 133.564,
817
+ "eval_samples_per_second": 7.487,
818
+ "step": 54000
819
+ },
820
+ {
821
+ "epoch": 5.8,
822
+ "learning_rate": 3.734891687265225e-05,
823
+ "loss": 2.2152,
824
+ "step": 55000
825
+ },
826
+ {
827
+ "epoch": 5.8,
828
+ "eval_bleu": 22.6911,
829
+ "eval_gen_len": 11.479,
830
+ "eval_loss": 2.4134082794189453,
831
+ "eval_runtime": 133.7203,
832
+ "eval_samples_per_second": 7.478,
833
+ "step": 55000
834
+ },
835
+ {
836
+ "epoch": 5.91,
837
+ "learning_rate": 3.710562681251095e-05,
838
+ "loss": 2.2198,
839
+ "step": 56000
840
+ },
841
+ {
842
+ "epoch": 5.91,
843
+ "eval_bleu": 22.5865,
844
+ "eval_gen_len": 11.59,
845
+ "eval_loss": 2.433336019515991,
846
+ "eval_runtime": 173.431,
847
+ "eval_samples_per_second": 5.766,
848
+ "step": 56000
849
+ },
850
+ {
851
+ "epoch": 6.01,
852
+ "learning_rate": 3.686233675236965e-05,
853
+ "loss": 2.2326,
854
+ "step": 57000
855
+ },
856
+ {
857
+ "epoch": 6.01,
858
+ "eval_bleu": 23.0494,
859
+ "eval_gen_len": 11.487,
860
+ "eval_loss": 2.4282305240631104,
861
+ "eval_runtime": 145.1447,
862
+ "eval_samples_per_second": 6.89,
863
+ "step": 57000
864
+ },
865
+ {
866
+ "epoch": 6.12,
867
+ "learning_rate": 3.661904669222834e-05,
868
+ "loss": 2.1317,
869
+ "step": 58000
870
+ },
871
+ {
872
+ "epoch": 6.12,
873
+ "eval_bleu": 23.1905,
874
+ "eval_gen_len": 11.478,
875
+ "eval_loss": 2.4288105964660645,
876
+ "eval_runtime": 136.8723,
877
+ "eval_samples_per_second": 7.306,
878
+ "step": 58000
879
+ },
880
+ {
881
+ "epoch": 6.22,
882
+ "learning_rate": 3.6375756632087045e-05,
883
+ "loss": 2.1239,
884
+ "step": 59000
885
+ },
886
+ {
887
+ "epoch": 6.22,
888
+ "eval_bleu": 23.0939,
889
+ "eval_gen_len": 11.375,
890
+ "eval_loss": 2.4191648960113525,
891
+ "eval_runtime": 136.6994,
892
+ "eval_samples_per_second": 7.315,
893
+ "step": 59000
894
+ },
895
+ {
896
+ "epoch": 6.33,
897
+ "learning_rate": 3.613246657194574e-05,
898
+ "loss": 2.1326,
899
+ "step": 60000
900
+ },
901
+ {
902
+ "epoch": 6.33,
903
+ "eval_bleu": 23.3522,
904
+ "eval_gen_len": 11.23,
905
+ "eval_loss": 2.4099223613739014,
906
+ "eval_runtime": 144.7238,
907
+ "eval_samples_per_second": 6.91,
908
+ "step": 60000
909
+ },
910
+ {
911
+ "epoch": 6.44,
912
+ "learning_rate": 3.5889176511804435e-05,
913
+ "loss": 2.1159,
914
+ "step": 61000
915
+ },
916
+ {
917
+ "epoch": 6.44,
918
+ "eval_bleu": 22.748,
919
+ "eval_gen_len": 11.572,
920
+ "eval_loss": 2.419100284576416,
921
+ "eval_runtime": 148.7046,
922
+ "eval_samples_per_second": 6.725,
923
+ "step": 61000
924
+ },
925
+ {
926
+ "epoch": 6.54,
927
+ "learning_rate": 3.564588645166313e-05,
928
+ "loss": 2.1308,
929
+ "step": 62000
930
+ },
931
+ {
932
+ "epoch": 6.54,
933
+ "eval_bleu": 23.2816,
934
+ "eval_gen_len": 11.449,
935
+ "eval_loss": 2.397722005844116,
936
+ "eval_runtime": 132.2002,
937
+ "eval_samples_per_second": 7.564,
938
+ "step": 62000
939
+ },
940
+ {
941
+ "epoch": 6.65,
942
+ "learning_rate": 3.540259639152183e-05,
943
+ "loss": 2.1094,
944
+ "step": 63000
945
+ },
946
+ {
947
+ "epoch": 6.65,
948
+ "eval_bleu": 23.3833,
949
+ "eval_gen_len": 11.243,
950
+ "eval_loss": 2.4132864475250244,
951
+ "eval_runtime": 145.191,
952
+ "eval_samples_per_second": 6.887,
953
+ "step": 63000
954
+ },
955
+ {
956
+ "epoch": 6.75,
957
+ "learning_rate": 3.515930633138052e-05,
958
+ "loss": 2.1184,
959
+ "step": 64000
960
+ },
961
+ {
962
+ "epoch": 6.75,
963
+ "eval_bleu": 23.7361,
964
+ "eval_gen_len": 11.172,
965
+ "eval_loss": 2.3724682331085205,
966
+ "eval_runtime": 131.0274,
967
+ "eval_samples_per_second": 7.632,
968
+ "step": 64000
969
+ },
970
+ {
971
+ "epoch": 6.86,
972
+ "learning_rate": 3.491601627123922e-05,
973
+ "loss": 2.1308,
974
+ "step": 65000
975
+ },
976
+ {
977
+ "epoch": 6.86,
978
+ "eval_bleu": 22.8591,
979
+ "eval_gen_len": 11.63,
980
+ "eval_loss": 2.3898813724517822,
981
+ "eval_runtime": 146.2565,
982
+ "eval_samples_per_second": 6.837,
983
+ "step": 65000
984
+ },
985
+ {
986
+ "epoch": 6.96,
987
+ "learning_rate": 3.467272621109792e-05,
988
+ "loss": 2.1292,
989
+ "step": 66000
990
+ },
991
+ {
992
+ "epoch": 6.96,
993
+ "eval_bleu": 22.8675,
994
+ "eval_gen_len": 11.667,
995
+ "eval_loss": 2.3850231170654297,
996
+ "eval_runtime": 130.3031,
997
+ "eval_samples_per_second": 7.674,
998
+ "step": 66000
999
+ }
1000
+ ],
1001
+ "max_steps": 208516,
1002
+ "num_train_epochs": 22,
1003
+ "total_flos": 8.109412878134016e+16,
1004
+ "trial_name": null,
1005
+ "trial_params": null
1006
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6230a94f837c17dc9e61dd74f421a5df63877917e5c4a0a883b1a0c344af6921
3
+ size 2415