yanjiehuang commited on
Commit
96f6e08
·
verified ·
1 Parent(s): d9fb943

Upload folder using huggingface_hub

Browse files
EVA_1.4B_CLM/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "vocab_size": 114,
3
+ "hidden_size": 1024,
4
+ "num_hidden_layers": 26,
5
+ "num_attention_heads": 16,
6
+ "num_key_value_heads": 16,
7
+ "intermediate_size": 3072,
8
+ "max_position_embeddings": 8192,
9
+ "rms_norm_eps": "1e-6",
10
+ "initializer_range": 0.02,
11
+ "use_cache": true,
12
+ "num_experts": 8,
13
+ "num_experts_per_tok": 2,
14
+ "expert_capacity_factor": 1.5,
15
+ "router_aux_loss_coef": 0.01,
16
+ "moe_implementation": "megablocks",
17
+ "moe_world_size": 1,
18
+ "resid_dropout": 0.0,
19
+ "hidden_dropout": 0.0,
20
+ "dropout_warmup_steps": 0,
21
+ "dropout_ramp_steps": 0,
22
+ "dropout_schedule": "linear",
23
+ "attention_dropout": 0.0,
24
+ "gradient_clip_norm": 0.0,
25
+ "label_smoothing": 0.0,
26
+ "eos_loss_weight": 1.0
27
+ }
EVA_1.4B_CLM/eval_results_stage1.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "loss": 1.2702110850016999,
3
+ "perplexity": 3.5616042613983154,
4
+ "total_tokens": 19854467,
5
+ "total_samples": 10112,
6
+ "eval_time": 791.5355548858643,
7
+ "eval_mode": "generation",
8
+ "stage": "stage1",
9
+ "eval_dataset": "/rna-multiverse/data/cluster/sampling/training/train_30M_sqrt_s42_50only_new.fa",
10
+ "batch_size": 128,
11
+ "num_batches": 79
12
+ }
EVA_1.4B_CLM/model_weights.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:323c13d571d0be87e450cb7b103bf8b28396e55e877519e2d2fb3c434875d420
3
+ size 3057231170
EVA_1.4B_CLM/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "pad_token": "<pad>",
3
+ "bos_token": "<bos>",
4
+ "eos_token": "<eos>",
5
+ "unk_token": "<unk>"
6
+ }
EVA_1.4B_CLM/tokenizer.json ADDED
@@ -0,0 +1,1171 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": "1.0",
3
+ "truncation": null,
4
+ "padding": {
5
+ "strategy": "BatchLongest",
6
+ "direction": "Right",
7
+ "pad_to_multiple_of": null,
8
+ "pad_id": 0,
9
+ "pad_type_id": 0,
10
+ "pad_token": "<pad>"
11
+ },
12
+ "added_tokens": [
13
+ {
14
+ "id": 0,
15
+ "content": "<pad>",
16
+ "single_word": false,
17
+ "lstrip": false,
18
+ "rstrip": false,
19
+ "normalized": false,
20
+ "special": true
21
+ },
22
+ {
23
+ "id": 1,
24
+ "content": "<bos>",
25
+ "single_word": false,
26
+ "lstrip": false,
27
+ "rstrip": false,
28
+ "normalized": false,
29
+ "special": true
30
+ },
31
+ {
32
+ "id": 2,
33
+ "content": "<eos>",
34
+ "single_word": false,
35
+ "lstrip": false,
36
+ "rstrip": false,
37
+ "normalized": false,
38
+ "special": true
39
+ },
40
+ {
41
+ "id": 3,
42
+ "content": "<bos_glm>",
43
+ "single_word": false,
44
+ "lstrip": false,
45
+ "rstrip": false,
46
+ "normalized": false,
47
+ "special": true
48
+ },
49
+ {
50
+ "id": 4,
51
+ "content": "<eos_span>",
52
+ "single_word": false,
53
+ "lstrip": false,
54
+ "rstrip": false,
55
+ "normalized": false,
56
+ "special": true
57
+ },
58
+ {
59
+ "id": 5,
60
+ "content": "<unk>",
61
+ "single_word": false,
62
+ "lstrip": false,
63
+ "rstrip": false,
64
+ "normalized": false,
65
+ "special": true
66
+ },
67
+ {
68
+ "id": 6,
69
+ "content": "<span_0>",
70
+ "single_word": false,
71
+ "lstrip": false,
72
+ "rstrip": false,
73
+ "normalized": false,
74
+ "special": true
75
+ },
76
+ {
77
+ "id": 7,
78
+ "content": "<span_1>",
79
+ "single_word": false,
80
+ "lstrip": false,
81
+ "rstrip": false,
82
+ "normalized": false,
83
+ "special": true
84
+ },
85
+ {
86
+ "id": 8,
87
+ "content": "<span_2>",
88
+ "single_word": false,
89
+ "lstrip": false,
90
+ "rstrip": false,
91
+ "normalized": false,
92
+ "special": true
93
+ },
94
+ {
95
+ "id": 9,
96
+ "content": "<span_3>",
97
+ "single_word": false,
98
+ "lstrip": false,
99
+ "rstrip": false,
100
+ "normalized": false,
101
+ "special": true
102
+ },
103
+ {
104
+ "id": 10,
105
+ "content": "<span_4>",
106
+ "single_word": false,
107
+ "lstrip": false,
108
+ "rstrip": false,
109
+ "normalized": false,
110
+ "special": true
111
+ },
112
+ {
113
+ "id": 11,
114
+ "content": "<span_5>",
115
+ "single_word": false,
116
+ "lstrip": false,
117
+ "rstrip": false,
118
+ "normalized": false,
119
+ "special": true
120
+ },
121
+ {
122
+ "id": 12,
123
+ "content": "<span_6>",
124
+ "single_word": false,
125
+ "lstrip": false,
126
+ "rstrip": false,
127
+ "normalized": false,
128
+ "special": true
129
+ },
130
+ {
131
+ "id": 13,
132
+ "content": "<span_7>",
133
+ "single_word": false,
134
+ "lstrip": false,
135
+ "rstrip": false,
136
+ "normalized": false,
137
+ "special": true
138
+ },
139
+ {
140
+ "id": 14,
141
+ "content": "<span_8>",
142
+ "single_word": false,
143
+ "lstrip": false,
144
+ "rstrip": false,
145
+ "normalized": false,
146
+ "special": true
147
+ },
148
+ {
149
+ "id": 15,
150
+ "content": "<span_9>",
151
+ "single_word": false,
152
+ "lstrip": false,
153
+ "rstrip": false,
154
+ "normalized": false,
155
+ "special": true
156
+ },
157
+ {
158
+ "id": 16,
159
+ "content": "<span_10>",
160
+ "single_word": false,
161
+ "lstrip": false,
162
+ "rstrip": false,
163
+ "normalized": false,
164
+ "special": true
165
+ },
166
+ {
167
+ "id": 17,
168
+ "content": "<span_11>",
169
+ "single_word": false,
170
+ "lstrip": false,
171
+ "rstrip": false,
172
+ "normalized": false,
173
+ "special": true
174
+ },
175
+ {
176
+ "id": 18,
177
+ "content": "<span_12>",
178
+ "single_word": false,
179
+ "lstrip": false,
180
+ "rstrip": false,
181
+ "normalized": false,
182
+ "special": true
183
+ },
184
+ {
185
+ "id": 19,
186
+ "content": "<span_13>",
187
+ "single_word": false,
188
+ "lstrip": false,
189
+ "rstrip": false,
190
+ "normalized": false,
191
+ "special": true
192
+ },
193
+ {
194
+ "id": 20,
195
+ "content": "<span_14>",
196
+ "single_word": false,
197
+ "lstrip": false,
198
+ "rstrip": false,
199
+ "normalized": false,
200
+ "special": true
201
+ },
202
+ {
203
+ "id": 21,
204
+ "content": "<span_15>",
205
+ "single_word": false,
206
+ "lstrip": false,
207
+ "rstrip": false,
208
+ "normalized": false,
209
+ "special": true
210
+ },
211
+ {
212
+ "id": 22,
213
+ "content": "<span_16>",
214
+ "single_word": false,
215
+ "lstrip": false,
216
+ "rstrip": false,
217
+ "normalized": false,
218
+ "special": true
219
+ },
220
+ {
221
+ "id": 23,
222
+ "content": "<span_17>",
223
+ "single_word": false,
224
+ "lstrip": false,
225
+ "rstrip": false,
226
+ "normalized": false,
227
+ "special": true
228
+ },
229
+ {
230
+ "id": 24,
231
+ "content": "<span_18>",
232
+ "single_word": false,
233
+ "lstrip": false,
234
+ "rstrip": false,
235
+ "normalized": false,
236
+ "special": true
237
+ },
238
+ {
239
+ "id": 25,
240
+ "content": "<span_19>",
241
+ "single_word": false,
242
+ "lstrip": false,
243
+ "rstrip": false,
244
+ "normalized": false,
245
+ "special": true
246
+ },
247
+ {
248
+ "id": 26,
249
+ "content": "<span_20>",
250
+ "single_word": false,
251
+ "lstrip": false,
252
+ "rstrip": false,
253
+ "normalized": false,
254
+ "special": true
255
+ },
256
+ {
257
+ "id": 27,
258
+ "content": "<span_21>",
259
+ "single_word": false,
260
+ "lstrip": false,
261
+ "rstrip": false,
262
+ "normalized": false,
263
+ "special": true
264
+ },
265
+ {
266
+ "id": 28,
267
+ "content": "<span_22>",
268
+ "single_word": false,
269
+ "lstrip": false,
270
+ "rstrip": false,
271
+ "normalized": false,
272
+ "special": true
273
+ },
274
+ {
275
+ "id": 29,
276
+ "content": "<span_23>",
277
+ "single_word": false,
278
+ "lstrip": false,
279
+ "rstrip": false,
280
+ "normalized": false,
281
+ "special": true
282
+ },
283
+ {
284
+ "id": 30,
285
+ "content": "<span_24>",
286
+ "single_word": false,
287
+ "lstrip": false,
288
+ "rstrip": false,
289
+ "normalized": false,
290
+ "special": true
291
+ },
292
+ {
293
+ "id": 31,
294
+ "content": "<span_25>",
295
+ "single_word": false,
296
+ "lstrip": false,
297
+ "rstrip": false,
298
+ "normalized": false,
299
+ "special": true
300
+ },
301
+ {
302
+ "id": 32,
303
+ "content": "<span_26>",
304
+ "single_word": false,
305
+ "lstrip": false,
306
+ "rstrip": false,
307
+ "normalized": false,
308
+ "special": true
309
+ },
310
+ {
311
+ "id": 33,
312
+ "content": "<span_27>",
313
+ "single_word": false,
314
+ "lstrip": false,
315
+ "rstrip": false,
316
+ "normalized": false,
317
+ "special": true
318
+ },
319
+ {
320
+ "id": 34,
321
+ "content": "<span_28>",
322
+ "single_word": false,
323
+ "lstrip": false,
324
+ "rstrip": false,
325
+ "normalized": false,
326
+ "special": true
327
+ },
328
+ {
329
+ "id": 35,
330
+ "content": "<span_29>",
331
+ "single_word": false,
332
+ "lstrip": false,
333
+ "rstrip": false,
334
+ "normalized": false,
335
+ "special": true
336
+ },
337
+ {
338
+ "id": 36,
339
+ "content": "<span_30>",
340
+ "single_word": false,
341
+ "lstrip": false,
342
+ "rstrip": false,
343
+ "normalized": false,
344
+ "special": true
345
+ },
346
+ {
347
+ "id": 37,
348
+ "content": "<span_31>",
349
+ "single_word": false,
350
+ "lstrip": false,
351
+ "rstrip": false,
352
+ "normalized": false,
353
+ "special": true
354
+ },
355
+ {
356
+ "id": 38,
357
+ "content": "<span_32>",
358
+ "single_word": false,
359
+ "lstrip": false,
360
+ "rstrip": false,
361
+ "normalized": false,
362
+ "special": true
363
+ },
364
+ {
365
+ "id": 39,
366
+ "content": "<span_33>",
367
+ "single_word": false,
368
+ "lstrip": false,
369
+ "rstrip": false,
370
+ "normalized": false,
371
+ "special": true
372
+ },
373
+ {
374
+ "id": 40,
375
+ "content": "<span_34>",
376
+ "single_word": false,
377
+ "lstrip": false,
378
+ "rstrip": false,
379
+ "normalized": false,
380
+ "special": true
381
+ },
382
+ {
383
+ "id": 41,
384
+ "content": "<span_35>",
385
+ "single_word": false,
386
+ "lstrip": false,
387
+ "rstrip": false,
388
+ "normalized": false,
389
+ "special": true
390
+ },
391
+ {
392
+ "id": 42,
393
+ "content": "<span_36>",
394
+ "single_word": false,
395
+ "lstrip": false,
396
+ "rstrip": false,
397
+ "normalized": false,
398
+ "special": true
399
+ },
400
+ {
401
+ "id": 43,
402
+ "content": "<span_37>",
403
+ "single_word": false,
404
+ "lstrip": false,
405
+ "rstrip": false,
406
+ "normalized": false,
407
+ "special": true
408
+ },
409
+ {
410
+ "id": 44,
411
+ "content": "<span_38>",
412
+ "single_word": false,
413
+ "lstrip": false,
414
+ "rstrip": false,
415
+ "normalized": false,
416
+ "special": true
417
+ },
418
+ {
419
+ "id": 45,
420
+ "content": "<span_39>",
421
+ "single_word": false,
422
+ "lstrip": false,
423
+ "rstrip": false,
424
+ "normalized": false,
425
+ "special": true
426
+ },
427
+ {
428
+ "id": 46,
429
+ "content": "<span_40>",
430
+ "single_word": false,
431
+ "lstrip": false,
432
+ "rstrip": false,
433
+ "normalized": false,
434
+ "special": true
435
+ },
436
+ {
437
+ "id": 47,
438
+ "content": "<span_41>",
439
+ "single_word": false,
440
+ "lstrip": false,
441
+ "rstrip": false,
442
+ "normalized": false,
443
+ "special": true
444
+ },
445
+ {
446
+ "id": 48,
447
+ "content": "<span_42>",
448
+ "single_word": false,
449
+ "lstrip": false,
450
+ "rstrip": false,
451
+ "normalized": false,
452
+ "special": true
453
+ },
454
+ {
455
+ "id": 49,
456
+ "content": "<span_43>",
457
+ "single_word": false,
458
+ "lstrip": false,
459
+ "rstrip": false,
460
+ "normalized": false,
461
+ "special": true
462
+ },
463
+ {
464
+ "id": 50,
465
+ "content": "<span_44>",
466
+ "single_word": false,
467
+ "lstrip": false,
468
+ "rstrip": false,
469
+ "normalized": false,
470
+ "special": true
471
+ },
472
+ {
473
+ "id": 51,
474
+ "content": "<span_45>",
475
+ "single_word": false,
476
+ "lstrip": false,
477
+ "rstrip": false,
478
+ "normalized": false,
479
+ "special": true
480
+ },
481
+ {
482
+ "id": 52,
483
+ "content": "<span_46>",
484
+ "single_word": false,
485
+ "lstrip": false,
486
+ "rstrip": false,
487
+ "normalized": false,
488
+ "special": true
489
+ },
490
+ {
491
+ "id": 53,
492
+ "content": "<span_47>",
493
+ "single_word": false,
494
+ "lstrip": false,
495
+ "rstrip": false,
496
+ "normalized": false,
497
+ "special": true
498
+ },
499
+ {
500
+ "id": 54,
501
+ "content": "<span_48>",
502
+ "single_word": false,
503
+ "lstrip": false,
504
+ "rstrip": false,
505
+ "normalized": false,
506
+ "special": true
507
+ },
508
+ {
509
+ "id": 55,
510
+ "content": "<span_49>",
511
+ "single_word": false,
512
+ "lstrip": false,
513
+ "rstrip": false,
514
+ "normalized": false,
515
+ "special": true
516
+ },
517
+ {
518
+ "id": 56,
519
+ "content": "<rna_mRNA>",
520
+ "single_word": false,
521
+ "lstrip": false,
522
+ "rstrip": false,
523
+ "normalized": false,
524
+ "special": true
525
+ },
526
+ {
527
+ "id": 57,
528
+ "content": "<rna_rRNA>",
529
+ "single_word": false,
530
+ "lstrip": false,
531
+ "rstrip": false,
532
+ "normalized": false,
533
+ "special": true
534
+ },
535
+ {
536
+ "id": 58,
537
+ "content": "<rna_tRNA>",
538
+ "single_word": false,
539
+ "lstrip": false,
540
+ "rstrip": false,
541
+ "normalized": false,
542
+ "special": true
543
+ },
544
+ {
545
+ "id": 59,
546
+ "content": "<rna_sRNA>",
547
+ "single_word": false,
548
+ "lstrip": false,
549
+ "rstrip": false,
550
+ "normalized": false,
551
+ "special": true
552
+ },
553
+ {
554
+ "id": 60,
555
+ "content": "<rna_lncRNA>",
556
+ "single_word": false,
557
+ "lstrip": false,
558
+ "rstrip": false,
559
+ "normalized": false,
560
+ "special": true
561
+ },
562
+ {
563
+ "id": 61,
564
+ "content": "<rna_circRNA>",
565
+ "single_word": false,
566
+ "lstrip": false,
567
+ "rstrip": false,
568
+ "normalized": false,
569
+ "special": true
570
+ },
571
+ {
572
+ "id": 62,
573
+ "content": "<rna_viral_RNA>",
574
+ "single_word": false,
575
+ "lstrip": false,
576
+ "rstrip": false,
577
+ "normalized": false,
578
+ "special": true
579
+ },
580
+ {
581
+ "id": 63,
582
+ "content": "<rna_miRNA>",
583
+ "single_word": false,
584
+ "lstrip": false,
585
+ "rstrip": false,
586
+ "normalized": false,
587
+ "special": true
588
+ },
589
+ {
590
+ "id": 64,
591
+ "content": "<rna_snoRNA>",
592
+ "single_word": false,
593
+ "lstrip": false,
594
+ "rstrip": false,
595
+ "normalized": false,
596
+ "special": true
597
+ },
598
+ {
599
+ "id": 65,
600
+ "content": "<rna_snRNA>",
601
+ "single_word": false,
602
+ "lstrip": false,
603
+ "rstrip": false,
604
+ "normalized": false,
605
+ "special": true
606
+ },
607
+ {
608
+ "id": 66,
609
+ "content": "<rna_piRNA>",
610
+ "single_word": false,
611
+ "lstrip": false,
612
+ "rstrip": false,
613
+ "normalized": false,
614
+ "special": true
615
+ },
616
+ {
617
+ "id": 67,
618
+ "content": "<rna_ribozyme>",
619
+ "single_word": false,
620
+ "lstrip": false,
621
+ "rstrip": false,
622
+ "normalized": false,
623
+ "special": true
624
+ },
625
+ {
626
+ "id": 68,
627
+ "content": "<rna_scaRNA>",
628
+ "single_word": false,
629
+ "lstrip": false,
630
+ "rstrip": false,
631
+ "normalized": false,
632
+ "special": true
633
+ },
634
+ {
635
+ "id": 69,
636
+ "content": "<rna_Y_RNA>",
637
+ "single_word": false,
638
+ "lstrip": false,
639
+ "rstrip": false,
640
+ "normalized": false,
641
+ "special": true
642
+ },
643
+ {
644
+ "id": 70,
645
+ "content": "<rna_vault_RNA>",
646
+ "single_word": false,
647
+ "lstrip": false,
648
+ "rstrip": false,
649
+ "normalized": false,
650
+ "special": true
651
+ },
652
+ {
653
+ "id": 71,
654
+ "content": "d__",
655
+ "single_word": false,
656
+ "lstrip": false,
657
+ "rstrip": false,
658
+ "normalized": false,
659
+ "special": true
660
+ },
661
+ {
662
+ "id": 72,
663
+ "content": "p__",
664
+ "single_word": false,
665
+ "lstrip": false,
666
+ "rstrip": false,
667
+ "normalized": false,
668
+ "special": true
669
+ },
670
+ {
671
+ "id": 73,
672
+ "content": "c__",
673
+ "single_word": false,
674
+ "lstrip": false,
675
+ "rstrip": false,
676
+ "normalized": false,
677
+ "special": true
678
+ },
679
+ {
680
+ "id": 74,
681
+ "content": "o__",
682
+ "single_word": false,
683
+ "lstrip": false,
684
+ "rstrip": false,
685
+ "normalized": false,
686
+ "special": true
687
+ },
688
+ {
689
+ "id": 75,
690
+ "content": "f__",
691
+ "single_word": false,
692
+ "lstrip": false,
693
+ "rstrip": false,
694
+ "normalized": false,
695
+ "special": true
696
+ },
697
+ {
698
+ "id": 76,
699
+ "content": "g__",
700
+ "single_word": false,
701
+ "lstrip": false,
702
+ "rstrip": false,
703
+ "normalized": false,
704
+ "special": true
705
+ },
706
+ {
707
+ "id": 77,
708
+ "content": "s__",
709
+ "single_word": false,
710
+ "lstrip": false,
711
+ "rstrip": false,
712
+ "normalized": false,
713
+ "special": true
714
+ },
715
+ {
716
+ "id": 78,
717
+ "content": ";",
718
+ "single_word": false,
719
+ "lstrip": false,
720
+ "rstrip": false,
721
+ "normalized": false,
722
+ "special": true
723
+ },
724
+ {
725
+ "id": 79,
726
+ "content": "|",
727
+ "single_word": false,
728
+ "lstrip": false,
729
+ "rstrip": false,
730
+ "normalized": false,
731
+ "special": true
732
+ },
733
+ {
734
+ "id": 80,
735
+ "content": "_",
736
+ "single_word": false,
737
+ "lstrip": false,
738
+ "rstrip": false,
739
+ "normalized": false,
740
+ "special": true
741
+ },
742
+ {
743
+ "id": 81,
744
+ "content": "-",
745
+ "single_word": false,
746
+ "lstrip": false,
747
+ "rstrip": false,
748
+ "normalized": false,
749
+ "special": true
750
+ },
751
+ {
752
+ "id": 82,
753
+ "content": "A",
754
+ "single_word": false,
755
+ "lstrip": false,
756
+ "rstrip": false,
757
+ "normalized": false,
758
+ "special": true
759
+ },
760
+ {
761
+ "id": 83,
762
+ "content": "U",
763
+ "single_word": false,
764
+ "lstrip": false,
765
+ "rstrip": false,
766
+ "normalized": false,
767
+ "special": true
768
+ },
769
+ {
770
+ "id": 84,
771
+ "content": "G",
772
+ "single_word": false,
773
+ "lstrip": false,
774
+ "rstrip": false,
775
+ "normalized": false,
776
+ "special": true
777
+ },
778
+ {
779
+ "id": 85,
780
+ "content": "C",
781
+ "single_word": false,
782
+ "lstrip": false,
783
+ "rstrip": false,
784
+ "normalized": false,
785
+ "special": true
786
+ },
787
+ {
788
+ "id": 86,
789
+ "content": "5",
790
+ "single_word": false,
791
+ "lstrip": false,
792
+ "rstrip": false,
793
+ "normalized": false,
794
+ "special": true
795
+ },
796
+ {
797
+ "id": 87,
798
+ "content": "3",
799
+ "single_word": false,
800
+ "lstrip": false,
801
+ "rstrip": false,
802
+ "normalized": false,
803
+ "special": true
804
+ },
805
+ {
806
+ "id": 88,
807
+ "content": "a",
808
+ "single_word": false,
809
+ "lstrip": false,
810
+ "rstrip": false,
811
+ "normalized": false,
812
+ "special": true
813
+ },
814
+ {
815
+ "id": 89,
816
+ "content": "b",
817
+ "single_word": false,
818
+ "lstrip": false,
819
+ "rstrip": false,
820
+ "normalized": false,
821
+ "special": true
822
+ },
823
+ {
824
+ "id": 90,
825
+ "content": "c",
826
+ "single_word": false,
827
+ "lstrip": false,
828
+ "rstrip": false,
829
+ "normalized": false,
830
+ "special": true
831
+ },
832
+ {
833
+ "id": 91,
834
+ "content": "d",
835
+ "single_word": false,
836
+ "lstrip": false,
837
+ "rstrip": false,
838
+ "normalized": false,
839
+ "special": true
840
+ },
841
+ {
842
+ "id": 92,
843
+ "content": "e",
844
+ "single_word": false,
845
+ "lstrip": false,
846
+ "rstrip": false,
847
+ "normalized": false,
848
+ "special": true
849
+ },
850
+ {
851
+ "id": 93,
852
+ "content": "f",
853
+ "single_word": false,
854
+ "lstrip": false,
855
+ "rstrip": false,
856
+ "normalized": false,
857
+ "special": true
858
+ },
859
+ {
860
+ "id": 94,
861
+ "content": "g",
862
+ "single_word": false,
863
+ "lstrip": false,
864
+ "rstrip": false,
865
+ "normalized": false,
866
+ "special": true
867
+ },
868
+ {
869
+ "id": 95,
870
+ "content": "h",
871
+ "single_word": false,
872
+ "lstrip": false,
873
+ "rstrip": false,
874
+ "normalized": false,
875
+ "special": true
876
+ },
877
+ {
878
+ "id": 96,
879
+ "content": "i",
880
+ "single_word": false,
881
+ "lstrip": false,
882
+ "rstrip": false,
883
+ "normalized": false,
884
+ "special": true
885
+ },
886
+ {
887
+ "id": 97,
888
+ "content": "j",
889
+ "single_word": false,
890
+ "lstrip": false,
891
+ "rstrip": false,
892
+ "normalized": false,
893
+ "special": true
894
+ },
895
+ {
896
+ "id": 98,
897
+ "content": "k",
898
+ "single_word": false,
899
+ "lstrip": false,
900
+ "rstrip": false,
901
+ "normalized": false,
902
+ "special": true
903
+ },
904
+ {
905
+ "id": 99,
906
+ "content": "l",
907
+ "single_word": false,
908
+ "lstrip": false,
909
+ "rstrip": false,
910
+ "normalized": false,
911
+ "special": true
912
+ },
913
+ {
914
+ "id": 100,
915
+ "content": "m",
916
+ "single_word": false,
917
+ "lstrip": false,
918
+ "rstrip": false,
919
+ "normalized": false,
920
+ "special": true
921
+ },
922
+ {
923
+ "id": 101,
924
+ "content": "n",
925
+ "single_word": false,
926
+ "lstrip": false,
927
+ "rstrip": false,
928
+ "normalized": false,
929
+ "special": true
930
+ },
931
+ {
932
+ "id": 102,
933
+ "content": "o",
934
+ "single_word": false,
935
+ "lstrip": false,
936
+ "rstrip": false,
937
+ "normalized": false,
938
+ "special": true
939
+ },
940
+ {
941
+ "id": 103,
942
+ "content": "p",
943
+ "single_word": false,
944
+ "lstrip": false,
945
+ "rstrip": false,
946
+ "normalized": false,
947
+ "special": true
948
+ },
949
+ {
950
+ "id": 104,
951
+ "content": "q",
952
+ "single_word": false,
953
+ "lstrip": false,
954
+ "rstrip": false,
955
+ "normalized": false,
956
+ "special": true
957
+ },
958
+ {
959
+ "id": 105,
960
+ "content": "r",
961
+ "single_word": false,
962
+ "lstrip": false,
963
+ "rstrip": false,
964
+ "normalized": false,
965
+ "special": true
966
+ },
967
+ {
968
+ "id": 106,
969
+ "content": "s",
970
+ "single_word": false,
971
+ "lstrip": false,
972
+ "rstrip": false,
973
+ "normalized": false,
974
+ "special": true
975
+ },
976
+ {
977
+ "id": 107,
978
+ "content": "t",
979
+ "single_word": false,
980
+ "lstrip": false,
981
+ "rstrip": false,
982
+ "normalized": false,
983
+ "special": true
984
+ },
985
+ {
986
+ "id": 108,
987
+ "content": "u",
988
+ "single_word": false,
989
+ "lstrip": false,
990
+ "rstrip": false,
991
+ "normalized": false,
992
+ "special": true
993
+ },
994
+ {
995
+ "id": 109,
996
+ "content": "v",
997
+ "single_word": false,
998
+ "lstrip": false,
999
+ "rstrip": false,
1000
+ "normalized": false,
1001
+ "special": true
1002
+ },
1003
+ {
1004
+ "id": 110,
1005
+ "content": "w",
1006
+ "single_word": false,
1007
+ "lstrip": false,
1008
+ "rstrip": false,
1009
+ "normalized": false,
1010
+ "special": true
1011
+ },
1012
+ {
1013
+ "id": 111,
1014
+ "content": "x",
1015
+ "single_word": false,
1016
+ "lstrip": false,
1017
+ "rstrip": false,
1018
+ "normalized": false,
1019
+ "special": true
1020
+ },
1021
+ {
1022
+ "id": 112,
1023
+ "content": "y",
1024
+ "single_word": false,
1025
+ "lstrip": false,
1026
+ "rstrip": false,
1027
+ "normalized": false,
1028
+ "special": true
1029
+ },
1030
+ {
1031
+ "id": 113,
1032
+ "content": "z",
1033
+ "single_word": false,
1034
+ "lstrip": false,
1035
+ "rstrip": false,
1036
+ "normalized": false,
1037
+ "special": true
1038
+ }
1039
+ ],
1040
+ "normalizer": null,
1041
+ "pre_tokenizer": null,
1042
+ "post_processor": null,
1043
+ "decoder": null,
1044
+ "model": {
1045
+ "type": "BPE",
1046
+ "dropout": null,
1047
+ "unk_token": "<unk>",
1048
+ "continuing_subword_prefix": null,
1049
+ "end_of_word_suffix": null,
1050
+ "fuse_unk": false,
1051
+ "byte_fallback": false,
1052
+ "ignore_merges": false,
1053
+ "vocab": {
1054
+ "<pad>": 0,
1055
+ "<bos>": 1,
1056
+ "<eos>": 2,
1057
+ "<bos_glm>": 3,
1058
+ "<eos_span>": 4,
1059
+ "<unk>": 5,
1060
+ "<span_0>": 6,
1061
+ "<span_1>": 7,
1062
+ "<span_2>": 8,
1063
+ "<span_3>": 9,
1064
+ "<span_4>": 10,
1065
+ "<span_5>": 11,
1066
+ "<span_6>": 12,
1067
+ "<span_7>": 13,
1068
+ "<span_8>": 14,
1069
+ "<span_9>": 15,
1070
+ "<span_10>": 16,
1071
+ "<span_11>": 17,
1072
+ "<span_12>": 18,
1073
+ "<span_13>": 19,
1074
+ "<span_14>": 20,
1075
+ "<span_15>": 21,
1076
+ "<span_16>": 22,
1077
+ "<span_17>": 23,
1078
+ "<span_18>": 24,
1079
+ "<span_19>": 25,
1080
+ "<span_20>": 26,
1081
+ "<span_21>": 27,
1082
+ "<span_22>": 28,
1083
+ "<span_23>": 29,
1084
+ "<span_24>": 30,
1085
+ "<span_25>": 31,
1086
+ "<span_26>": 32,
1087
+ "<span_27>": 33,
1088
+ "<span_28>": 34,
1089
+ "<span_29>": 35,
1090
+ "<span_30>": 36,
1091
+ "<span_31>": 37,
1092
+ "<span_32>": 38,
1093
+ "<span_33>": 39,
1094
+ "<span_34>": 40,
1095
+ "<span_35>": 41,
1096
+ "<span_36>": 42,
1097
+ "<span_37>": 43,
1098
+ "<span_38>": 44,
1099
+ "<span_39>": 45,
1100
+ "<span_40>": 46,
1101
+ "<span_41>": 47,
1102
+ "<span_42>": 48,
1103
+ "<span_43>": 49,
1104
+ "<span_44>": 50,
1105
+ "<span_45>": 51,
1106
+ "<span_46>": 52,
1107
+ "<span_47>": 53,
1108
+ "<span_48>": 54,
1109
+ "<span_49>": 55,
1110
+ "<rna_mRNA>": 56,
1111
+ "<rna_rRNA>": 57,
1112
+ "<rna_tRNA>": 58,
1113
+ "<rna_sRNA>": 59,
1114
+ "<rna_lncRNA>": 60,
1115
+ "<rna_circRNA>": 61,
1116
+ "<rna_viral_RNA>": 62,
1117
+ "<rna_miRNA>": 63,
1118
+ "<rna_snoRNA>": 64,
1119
+ "<rna_snRNA>": 65,
1120
+ "<rna_piRNA>": 66,
1121
+ "<rna_ribozyme>": 67,
1122
+ "<rna_scaRNA>": 68,
1123
+ "<rna_Y_RNA>": 69,
1124
+ "<rna_vault_RNA>": 70,
1125
+ "d__": 71,
1126
+ "p__": 72,
1127
+ "c__": 73,
1128
+ "o__": 74,
1129
+ "f__": 75,
1130
+ "g__": 76,
1131
+ "s__": 77,
1132
+ ";": 78,
1133
+ "|": 79,
1134
+ "_": 80,
1135
+ "-": 81,
1136
+ "A": 82,
1137
+ "U": 83,
1138
+ "G": 84,
1139
+ "C": 85,
1140
+ "5": 86,
1141
+ "3": 87,
1142
+ "a": 88,
1143
+ "b": 89,
1144
+ "c": 90,
1145
+ "d": 91,
1146
+ "e": 92,
1147
+ "f": 93,
1148
+ "g": 94,
1149
+ "h": 95,
1150
+ "i": 96,
1151
+ "j": 97,
1152
+ "k": 98,
1153
+ "l": 99,
1154
+ "m": 100,
1155
+ "n": 101,
1156
+ "o": 102,
1157
+ "p": 103,
1158
+ "q": 104,
1159
+ "r": 105,
1160
+ "s": 106,
1161
+ "t": 107,
1162
+ "u": 108,
1163
+ "v": 109,
1164
+ "w": 110,
1165
+ "x": 111,
1166
+ "y": 112,
1167
+ "z": 113
1168
+ },
1169
+ "merges": []
1170
+ }
1171
+ }
EVA_1.4B_CLM/tokenizer_config.json ADDED
@@ -0,0 +1,114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "tokenizer_class": "LineageRNATokenizer",
3
+ "auto_map": {
4
+ "AutoTokenizer": [
5
+ "lineage_tokenizer.py",
6
+ "LineageRNATokenizer"
7
+ ]
8
+ },
9
+ "vocab_size": 114,
10
+ "pad_token": "<pad>",
11
+ "bos_token": "<bos>",
12
+ "eos_token": "<eos>",
13
+ "unk_token": "<unk>",
14
+ "rna_bases": [
15
+ "A",
16
+ "U",
17
+ "G",
18
+ "C"
19
+ ],
20
+ "special_tokens": [
21
+ "<pad>",
22
+ "<bos>",
23
+ "<eos>",
24
+ "<bos_glm>",
25
+ "<eos_span>",
26
+ "<unk>"
27
+ ],
28
+ "glm_span_tokens": [
29
+ "<span_0>",
30
+ "<span_1>",
31
+ "<span_2>",
32
+ "<span_3>",
33
+ "<span_4>",
34
+ "<span_5>",
35
+ "<span_6>",
36
+ "<span_7>",
37
+ "<span_8>",
38
+ "<span_9>",
39
+ "<span_10>",
40
+ "<span_11>",
41
+ "<span_12>",
42
+ "<span_13>",
43
+ "<span_14>",
44
+ "<span_15>",
45
+ "<span_16>",
46
+ "<span_17>",
47
+ "<span_18>",
48
+ "<span_19>",
49
+ "<span_20>",
50
+ "<span_21>",
51
+ "<span_22>",
52
+ "<span_23>",
53
+ "<span_24>",
54
+ "<span_25>",
55
+ "<span_26>",
56
+ "<span_27>",
57
+ "<span_28>",
58
+ "<span_29>",
59
+ "<span_30>",
60
+ "<span_31>",
61
+ "<span_32>",
62
+ "<span_33>",
63
+ "<span_34>",
64
+ "<span_35>",
65
+ "<span_36>",
66
+ "<span_37>",
67
+ "<span_38>",
68
+ "<span_39>",
69
+ "<span_40>",
70
+ "<span_41>",
71
+ "<span_42>",
72
+ "<span_43>",
73
+ "<span_44>",
74
+ "<span_45>",
75
+ "<span_46>",
76
+ "<span_47>",
77
+ "<span_48>",
78
+ "<span_49>"
79
+ ],
80
+ "rna_type_tokens": [
81
+ "<rna_mRNA>",
82
+ "<rna_rRNA>",
83
+ "<rna_tRNA>",
84
+ "<rna_sRNA>",
85
+ "<rna_lncRNA>",
86
+ "<rna_circRNA>",
87
+ "<rna_viral_RNA>",
88
+ "<rna_miRNA>",
89
+ "<rna_snoRNA>",
90
+ "<rna_snRNA>",
91
+ "<rna_piRNA>",
92
+ "<rna_ribozyme>",
93
+ "<rna_scaRNA>",
94
+ "<rna_Y_RNA>",
95
+ "<rna_vault_RNA>"
96
+ ],
97
+ "lineage_level_tokens": [
98
+ "d__",
99
+ "p__",
100
+ "c__",
101
+ "o__",
102
+ "f__",
103
+ "g__",
104
+ "s__"
105
+ ],
106
+ "lineage_special_chars": [
107
+ ";",
108
+ "|",
109
+ "_",
110
+ "-"
111
+ ],
112
+ "mode": "lineage",
113
+ "description": "Lineage-based tokenizer without species tokens or task tokens"
114
+ }
EVA_1.4B_CLM/training_config.yaml ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ data_config:
2
+ cache_size: 10000
3
+ dataset: train_30M_sqrt_s42
4
+ enable_reverse_augmentation: true
5
+ force_rebuild_index: false
6
+ glm_probability: 0.333
7
+ lineage_file: /rna-multiverse/data/training_data/lineage_greengenes.tsv
8
+ max_samples: null
9
+ max_seq_length: 8192
10
+ mode: mixed
11
+ pretrain_ratio: 0.2
12
+ span_config:
13
+ allow_overlap: false
14
+ coverage_probs:
15
+ - 0.28
16
+ - 0.3
17
+ - 0.28
18
+ - 0.14
19
+ distribution_probs:
20
+ - 0.3
21
+ - 0.5
22
+ - 0.2
23
+ max_coverage_ratios:
24
+ - 0.15
25
+ - 0.25
26
+ - 0.5
27
+ - 0.8
28
+ max_num_spans: 10
29
+ span_distributions:
30
+ - - 10
31
+ - 5
32
+ - - 20
33
+ - 10
34
+ - - 50
35
+ - 20
36
+ train_file: /rna-multiverse/data/cluster/sampling/training/train_30M_sqrt_s42_50only_new.fa
37
+ use_chunked: true
38
+ use_direction_tokens: true
39
+ use_lineage_prefix: true
40
+ use_rna_type_prefix: true
41
+ distributed_config:
42
+ backend: nccl
43
+ data_parallel_size: 4
44
+ expert_parallel_size: 4
45
+ weight_parallel_size: 1
46
+ logging_config:
47
+ enable_wandb: true
48
+ log_dir: /rna-multiverse/results/mid_training/mid_training_v4_from_v24_1.4B_checkpoint_15000_20251202/logs
49
+ wandb_project: rna-mid-training
50
+ wandb_run_name: mid_training_v4_from_v24_1.4B_checkpoint_15000
51
+ memory_config:
52
+ cleanup_frequency: 100
53
+ enable_monitoring: true
54
+ gc_frequency: 50
55
+ model_config:
56
+ attention_dropout: 0.0
57
+ dropout_ramp_steps: 0
58
+ dropout_schedule: linear
59
+ dropout_warmup_steps: 0
60
+ eos_loss_weight: 1.0
61
+ expert_capacity_factor: 1.5
62
+ gradient_clip_norm: 0.0
63
+ hidden_dropout: 0.0
64
+ hidden_size: 1024
65
+ initializer_range: 0.02
66
+ intermediate_size: 3072
67
+ label_smoothing: 0.0
68
+ max_position_embeddings: 8192
69
+ moe_implementation: megablocks
70
+ moe_world_size: 4
71
+ num_attention_heads: 16
72
+ num_experts: 8
73
+ num_experts_per_tok: 2
74
+ num_hidden_layers: 26
75
+ num_key_value_heads: 16
76
+ resid_dropout: 0.0
77
+ rms_norm_eps: 1e-6
78
+ router_aux_loss_coef: 0.01
79
+ use_cache: true
80
+ vocab_size: 114
81
+ training_config:
82
+ adam_beta1: 0.9
83
+ adam_beta2: 0.95
84
+ adam_epsilon: 1e-8
85
+ bf16: true
86
+ dataloader_drop_last: true
87
+ dataloader_num_workers: 8
88
+ dataloader_pin_memory: true
89
+ fp16: false
90
+ gradient_accumulation_steps: 16
91
+ gradient_checkpointing: false
92
+ learning_rate: 0.0001
93
+ logging_steps: 30
94
+ max_epochs: 1
95
+ max_wall_time_hours: 100
96
+ min_lr_ratio: 0.1
97
+ output_dir: /rna-multiverse/results/mid_training/mid_training_v4_from_v24_1.4B_checkpoint_15000_20251202
98
+ per_device_train_batch_size: 4
99
+ resume_from_pretrain: /rna-multiverse/results/experiments/scaling_1.4B_v24_mixed_glm_stage1_20251126/checkpoint-15000
100
+ run_name: mid_training_v4_from_v24_1.4B_checkpoint_15000
101
+ save_steps: 500
102
+ seed: 42
103
+ warmup_steps: 1000
104
+ weight_decay: 5.0e-06
EVA_1.4B_CLM/vocab.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "<rna_sRNA>": 59,
3
+ "s__": 77,
4
+ "<span_18>": 24,
5
+ "<span_8>": 14,
6
+ "n": 101,
7
+ ";": 78,
8
+ "A": 82,
9
+ "j": 97,
10
+ "<rna_ribozyme>": 67,
11
+ "<span_21>": 27,
12
+ "c__": 73,
13
+ "d__": 71,
14
+ "<bos_glm>": 3,
15
+ "h": 95,
16
+ "<rna_lncRNA>": 60,
17
+ "d": 91,
18
+ "<rna_tRNA>": 58,
19
+ "q": 104,
20
+ "<span_17>": 23,
21
+ "<span_35>": 41,
22
+ "o": 102,
23
+ "g": 94,
24
+ "<span_1>": 7,
25
+ "<span_29>": 35,
26
+ "c": 90,
27
+ "<span_36>": 42,
28
+ "p__": 72,
29
+ "<span_49>": 55,
30
+ "<rna_rRNA>": 57,
31
+ "<rna_mRNA>": 56,
32
+ "x": 111,
33
+ "<span_23>": 29,
34
+ "<span_2>": 8,
35
+ "<span_46>": 52,
36
+ "<span_31>": 37,
37
+ "<span_40>": 46,
38
+ "<span_27>": 33,
39
+ "<span_38>": 44,
40
+ "<span_25>": 31,
41
+ "<rna_snoRNA>": 64,
42
+ "<span_11>": 17,
43
+ "<span_14>": 20,
44
+ "G": 84,
45
+ "<span_39>": 45,
46
+ "w": 110,
47
+ "<rna_circRNA>": 61,
48
+ "<span_13>": 19,
49
+ "<rna_snRNA>": 65,
50
+ "<rna_vault_RNA>": 70,
51
+ "e": 92,
52
+ "v": 109,
53
+ "<pad>": 0,
54
+ "<span_22>": 28,
55
+ "a": 88,
56
+ "3": 87,
57
+ "C": 85,
58
+ "r": 105,
59
+ "<span_43>": 49,
60
+ "<span_24>": 30,
61
+ "f": 93,
62
+ "<span_7>": 13,
63
+ "<span_44>": 50,
64
+ "<span_26>": 32,
65
+ "<span_0>": 6,
66
+ "t": 107,
67
+ "_": 80,
68
+ "l": 99,
69
+ "<span_47>": 53,
70
+ "5": 86,
71
+ "<unk>": 5,
72
+ "U": 83,
73
+ "<span_41>": 47,
74
+ "<eos_span>": 4,
75
+ "p": 103,
76
+ "<span_42>": 48,
77
+ "k": 98,
78
+ "<span_19>": 25,
79
+ "<span_6>": 12,
80
+ "<span_15>": 21,
81
+ "<span_28>": 34,
82
+ "<span_3>": 9,
83
+ "<span_33>": 39,
84
+ "<span_12>": 18,
85
+ "|": 79,
86
+ "b": 89,
87
+ "<bos>": 1,
88
+ "<span_5>": 11,
89
+ "<rna_viral_RNA>": 62,
90
+ "<rna_scaRNA>": 68,
91
+ "<rna_piRNA>": 66,
92
+ "u": 108,
93
+ "o__": 74,
94
+ "g__": 76,
95
+ "<span_48>": 54,
96
+ "<rna_Y_RNA>": 69,
97
+ "i": 96,
98
+ "<span_4>": 10,
99
+ "f__": 75,
100
+ "<span_30>": 36,
101
+ "y": 112,
102
+ "<span_34>": 40,
103
+ "z": 113,
104
+ "<span_45>": 51,
105
+ "<rna_miRNA>": 63,
106
+ "<span_37>": 43,
107
+ "<span_10>": 16,
108
+ "s": 106,
109
+ "<span_16>": 22,
110
+ "<span_9>": 15,
111
+ "<span_20>": 26,
112
+ "<span_32>": 38,
113
+ "-": 81,
114
+ "<eos>": 2,
115
+ "m": 100
116
+ }