Hugging Face's logo

PKU-Alignment
/
Qwen1.5-4B-Safety-Q1-10k-Q2-2k

PyTorch
qwen2
Model card Files Community
Qwen1.5-4B-Safety-Q1-10k-Q2-2k
File size: 40,543 Bytes 
b42664f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
 
+ deepspeed --master_port 37048 --module safe_rlhf.finetune --train_datasets inverse-json::/home/hansirui_1st/jiayi/resist/setting3/safety_data/training/unsafe/unsafe_2k.json --model_name_or_path /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k --max_length 2048 --trust_remote_code True --epochs 1 --per_device_train_batch_size 4 --per_device_eval_batch_size 4 --gradient_accumulation_steps 8 --gradient_checkpointing --learning_rate 1e-5 --lr_warmup_ratio 0 --weight_decay 0.0 --lr_scheduler_type constant --weight_decay 0.0 --seed 42 --output_dir /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k-Q2-2k --log_type wandb --log_run_name qwen-4b-s3-Q1-10k-Q2-2k --log_project Inverse_Alignment --zero_stage 3 --offload none --bf16 True --tf32 True --save_16bit
[rank7]:[W528 22:09:21.036407041 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 7]  using GPU 7 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank3]:[W528 22:09:22.318229028 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 3]  using GPU 3 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank2]:[W528 22:09:22.392686726 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 2]  using GPU 2 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank4]:[W528 22:09:22.419481521 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 4]  using GPU 4 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank6]:[W528 22:09:22.503251480 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 6]  using GPU 6 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank1]:[W528 22:09:22.567638531 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 1]  using GPU 1 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank5]:[W528 22:09:22.570437274 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 5]  using GPU 5 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
[rank0]:[W528 22:09:22.572483520 ProcessGroupNCCL.cpp:4561] [PG ID 0 PG GUID 0 Rank 0]  using GPU 0 to perform barrier as devices used by this process are currently unknown. This can potentially cause a hang if this rank to GPU mapping is incorrect. Specify device_ids in barrier() to force use of a particular device, or call init_process_group() with a device_id.
loading configuration file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/config.json
loading configuration file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/config.json
loading configuration file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/config.json
loading configuration file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/config.json
loading configuration file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/config.json
loading configuration file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/config.json
loading configuration file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/config.json
loading configuration file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/config.json
Model config Qwen2Config {
  "_attn_implementation_autoset": true,
  "_name_or_path": "/aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "hidden_act": "silu",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 6912,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "qwen2",
  "num_attention_heads": 20,
  "num_hidden_layers": 40,
  "num_key_value_heads": 20,
  "pad_token_id": 151643,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 5000000.0,
  "sliding_window": 32768,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151646
}

Model config Qwen2Config {
  "_attn_implementation_autoset": true,
  "_name_or_path": "/aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "hidden_act": "silu",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 6912,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "qwen2",
  "num_attention_heads": 20,
  "num_hidden_layers": 40,
  "num_key_value_heads": 20,
  "pad_token_id": 151643,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 5000000.0,
  "sliding_window": 32768,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151646
}

Model config Qwen2Config {
  "_attn_implementation_autoset": true,
  "_name_or_path": "/aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "hidden_act": "silu",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 6912,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "qwen2",
  "num_attention_heads": 20,
  "num_hidden_layers": 40,
  "num_key_value_heads": 20,
  "pad_token_id": 151643,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 5000000.0,
  "sliding_window": 32768,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151646
}

Model config Qwen2Config {
  "_attn_implementation_autoset": true,
  "_name_or_path": "/aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "hidden_act": "silu",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 6912,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "qwen2",
  "num_attention_heads": 20,
  "num_hidden_layers": 40,
  "num_key_value_heads": 20,
  "pad_token_id": 151643,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 5000000.0,
  "sliding_window": 32768,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151646
}

Model config Qwen2Config {
  "_attn_implementation_autoset": true,
  "_name_or_path": "/aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "hidden_act": "silu",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 6912,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "qwen2",
  "num_attention_heads": 20,
  "num_hidden_layers": 40,
  "num_key_value_heads": 20,
  "pad_token_id": 151643,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 5000000.0,
  "sliding_window": 32768,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151646
}

Model config Qwen2Config {
  "_attn_implementation_autoset": true,
  "_name_or_path": "/aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "hidden_act": "silu",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 6912,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "qwen2",
  "num_attention_heads": 20,
  "num_hidden_layers": 40,
  "num_key_value_heads": 20,
  "pad_token_id": 151643,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 5000000.0,
  "sliding_window": 32768,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151646
}

Model config Qwen2Config {
  "_attn_implementation_autoset": true,
  "_name_or_path": "/aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "hidden_act": "silu",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 6912,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "qwen2",
  "num_attention_heads": 20,
  "num_hidden_layers": 40,
  "num_key_value_heads": 20,
  "pad_token_id": 151643,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 5000000.0,
  "sliding_window": 32768,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151646
}

Model config Qwen2Config {
  "_attn_implementation_autoset": true,
  "_name_or_path": "/aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k",
  "architectures": [
    "Qwen2ForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "hidden_act": "silu",
  "hidden_size": 2560,
  "initializer_range": 0.02,
  "intermediate_size": 6912,
  "max_position_embeddings": 32768,
  "max_window_layers": 21,
  "model_type": "qwen2",
  "num_attention_heads": 20,
  "num_hidden_layers": 40,
  "num_key_value_heads": 20,
  "pad_token_id": 151643,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 5000000.0,
  "sliding_window": 32768,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.49.0",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151646
}

loading weights file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/pytorch_model.bin
loading weights file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/pytorch_model.bin
loading weights file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/pytorch_model.bin
loading weights file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/pytorch_model.bin
loading weights file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/pytorch_model.bin
loading weights file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/pytorch_model.bin
loading weights file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/pytorch_model.bin
loading weights file /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k/pytorch_model.bin
Will use torch_dtype=torch.bfloat16 as defined in model's config object
Instantiating Qwen2ForCausalLM model under default dtype torch.bfloat16.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Generate config GenerationConfig {
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "pad_token_id": 151643
}

Sliding Window Attention is enabled but not implemented for `eager`; unexpected results may be encountered.
Will use torch_dtype=torch.bfloat16 as defined in model's config object
Instantiating Qwen2ForCausalLM model under default dtype torch.bfloat16.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Will use torch_dtype=torch.bfloat16 as defined in model's config object
Instantiating Qwen2ForCausalLM model under default dtype torch.bfloat16.
Will use torch_dtype=torch.bfloat16 as defined in model's config object
Instantiating Qwen2ForCausalLM model under default dtype torch.bfloat16.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Will use torch_dtype=torch.bfloat16 as defined in model's config object
Instantiating Qwen2ForCausalLM model under default dtype torch.bfloat16.
Will use torch_dtype=torch.bfloat16 as defined in model's config object
Instantiating Qwen2ForCausalLM model under default dtype torch.bfloat16.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Will use torch_dtype=torch.bfloat16 as defined in model's config object
Instantiating Qwen2ForCausalLM model under default dtype torch.bfloat16.
Will use torch_dtype=torch.bfloat16 as defined in model's config object
Instantiating Qwen2ForCausalLM model under default dtype torch.bfloat16.
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Detected DeepSpeed ZeRO-3: activating zero.init() for this model
Generate config GenerationConfig {
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "pad_token_id": 151643
}

Generate config GenerationConfig {
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "pad_token_id": 151643
}

Generate config GenerationConfig {
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "pad_token_id": 151643
}

Generate config GenerationConfig {
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "pad_token_id": 151643
}

Generate config GenerationConfig {
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "pad_token_id": 151643
}

Generate config GenerationConfig {
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "pad_token_id": 151643
}

Generate config GenerationConfig {
  "bos_token_id": 128245,
  "eos_token_id": 151643,
  "pad_token_id": 151643
}

Sliding Window Attention is enabled but not implemented for `eager`; unexpected results may be encountered.
Sliding Window Attention is enabled but not implemented for `eager`; unexpected results may be encountered.
Sliding Window Attention is enabled but not implemented for `eager`; unexpected results may be encountered.
Sliding Window Attention is enabled but not implemented for `eager`; unexpected results may be encountered.
Sliding Window Attention is enabled but not implemented for `eager`; unexpected results may be encountered.
Sliding Window Attention is enabled but not implemented for `eager`; unexpected results may be encountered.
Sliding Window Attention is enabled but not implemented for `eager`; unexpected results may be encountered.
All model checkpoint weights were used when initializing Qwen2ForCausalLM.

All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k.
If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2ForCausalLM for predictions without further training.
All model checkpoint weights were used when initializing Qwen2ForCausalLM.

All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k.
If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2ForCausalLM for predictions without further training.
All model checkpoint weights were used when initializing Qwen2ForCausalLM.

All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k.
If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2ForCausalLM for predictions without further training.
All model checkpoint weights were used when initializing Qwen2ForCausalLM.

All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k.
If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2ForCausalLM for predictions without further training.
All model checkpoint weights were used when initializing Qwen2ForCausalLM.

All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k.
If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2ForCausalLM for predictions without further training.
Generation config file not found, using a generation config created from the model config.
Generation config file not found, using a generation config created from the model config.
Generation config file not found, using a generation config created from the model config.
Generation config file not found, using a generation config created from the model config.
Generation config file not found, using a generation config created from the model config.
All model checkpoint weights were used when initializing Qwen2ForCausalLM.

All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k.
If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2ForCausalLM for predictions without further training.
All model checkpoint weights were used when initializing Qwen2ForCausalLM.

All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k.
If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2ForCausalLM for predictions without further training.
Generation config file not found, using a generation config created from the model config.
Generation config file not found, using a generation config created from the model config.
loading file vocab.json
loading file merges.txt
loading file tokenizer.json
loading file added_tokens.json
loading file vocab.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file merges.txt
loading file chat_template.jinja
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file vocab.json
loading file vocab.json
loading file merges.txt
loading file merges.txt
loading file tokenizer.json
loading file tokenizer.json
loading file added_tokens.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file chat_template.jinja
loading file vocab.json
loading file merges.txt
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file vocab.json
loading file merges.txt
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
loading file vocab.json
loading file merges.txt
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
All model checkpoint weights were used when initializing Qwen2ForCausalLM.

All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k.
If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2ForCausalLM for predictions without further training.
Generation config file not found, using a generation config created from the model config.
loading file vocab.json
loading file merges.txt
loading file tokenizer.json
loading file added_tokens.json
loading file special_tokens_map.json
loading file tokenizer_config.json
loading file chat_template.jinja
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...


Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Using /home/hansirui_1st/.cache/torch_extensions/py311_cu124 as PyTorch extensions root...
Detected CUDA files, patching ldflags
Emitting ninja build file /home/hansirui_1st/.cache/torch_extensions/py311_cu124/fused_adam/build.ninja...
/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/torch/utils/cpp_extension.py:2059: UserWarning: TORCH_CUDA_ARCH_LIST is not set, all archs for visible cards are included for compilation. 
If this is not desired, please set os.environ['TORCH_CUDA_ARCH_LIST'].
  warnings.warn(
Building extension module fused_adam...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
Loading extension module fused_adam...
wandb: Using wandb-core as the SDK backend.  Please refer to https://wandb.me/wandb-core for more information.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.
wandb: Currently logged in as: xtom to https://api.wandb.ai. Use `wandb login --relogin` to force relogin
wandb: Tracking run with wandb version 0.19.8
wandb: Run data is saved locally in /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k-Q2-2k/wandb/run-20250528_220942-06gqeb33
wandb: Run `wandb offline` to turn off syncing.
wandb: Syncing run qwen-4b-s3-Q1-10k-Q2-2k
wandb: ⭐️ View project at https://wandb.ai/xtom/Inverse_Alignment
wandb: 🚀 View run at https://wandb.ai/xtom/Inverse_Alignment/runs/06gqeb33

Training 1/1 epoch:   0%|          | 0/63 [00:00<?, ?it/s]`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`.

Training 1/1 epoch (loss 1.7146):   0%|          | 0/63 [00:07<?, ?it/s]
Training 1/1 epoch (loss 1.7146):   2%|▏         | 1/63 [00:07<07:23,  7.16s/it]
Training 1/1 epoch (loss 1.6844):   2%|▏         | 1/63 [00:09<07:23,  7.16s/it]
Training 1/1 epoch (loss 1.6844):   3%|▎         | 2/63 [00:09<04:29,  4.41s/it]
Training 1/1 epoch (loss 1.7634):   3%|▎         | 2/63 [00:10<04:29,  4.41s/it]
Training 1/1 epoch (loss 1.7634):   5%|▍         | 3/63 [00:10<02:41,  2.70s/it]
Training 1/1 epoch (loss 1.6159):   5%|▍         | 3/63 [00:10<02:41,  2.70s/it]
Training 1/1 epoch (loss 1.6159):   6%|▋         | 4/63 [00:10<01:52,  1.90s/it]
Training 1/1 epoch (loss 1.6885):   6%|▋         | 4/63 [00:11<01:52,  1.90s/it]
Training 1/1 epoch (loss 1.6885):   8%|▊         | 5/63 [00:11<01:24,  1.46s/it]
Training 1/1 epoch (loss 1.7611):   8%|▊         | 5/63 [00:12<01:24,  1.46s/it]
Training 1/1 epoch (loss 1.7611):  10%|▉         | 6/63 [00:12<01:08,  1.19s/it]
Training 1/1 epoch (loss 1.7074):  10%|▉         | 6/63 [00:13<01:08,  1.19s/it]
Training 1/1 epoch (loss 1.7074):  11%|█         | 7/63 [00:13<00:57,  1.02s/it]
Training 1/1 epoch (loss 1.7700):  11%|█         | 7/63 [00:13<00:57,  1.02s/it]
Training 1/1 epoch (loss 1.7700):  13%|█▎        | 8/63 [00:13<00:54,  1.01it/s]
Training 1/1 epoch (loss 1.6734):  13%|█▎        | 8/63 [00:14<00:54,  1.01it/s]
Training 1/1 epoch (loss 1.6734):  14%|█▍        | 9/63 [00:14<00:48,  1.12it/s]
Training 1/1 epoch (loss 1.6348):  14%|█▍        | 9/63 [00:15<00:48,  1.12it/s]
Training 1/1 epoch (loss 1.6348):  16%|█▌        | 10/63 [00:15<00:43,  1.22it/s]
Training 1/1 epoch (loss 1.7091):  16%|█▌        | 10/63 [00:15<00:43,  1.22it/s]
Training 1/1 epoch (loss 1.7091):  17%|█▋        | 11/63 [00:15<00:40,  1.30it/s]
Training 1/1 epoch (loss 1.7345):  17%|█▋        | 11/63 [00:16<00:40,  1.30it/s]
Training 1/1 epoch (loss 1.7345):  19%|█▉        | 12/63 [00:16<00:37,  1.37it/s]
Training 1/1 epoch (loss 1.7241):  19%|█▉        | 12/63 [00:17<00:37,  1.37it/s]
Training 1/1 epoch (loss 1.7241):  21%|██        | 13/63 [00:17<00:35,  1.41it/s]
Training 1/1 epoch (loss 1.6559):  21%|██        | 13/63 [00:17<00:35,  1.41it/s]
Training 1/1 epoch (loss 1.6559):  22%|██▏       | 14/63 [00:17<00:34,  1.41it/s]
Training 1/1 epoch (loss 1.6420):  22%|██▏       | 14/63 [00:18<00:34,  1.41it/s]
Training 1/1 epoch (loss 1.6420):  24%|██▍       | 15/63 [00:18<00:33,  1.45it/s]
Training 1/1 epoch (loss 1.6910):  24%|██▍       | 15/63 [00:19<00:33,  1.45it/s]
Training 1/1 epoch (loss 1.6910):  25%|██▌       | 16/63 [00:19<00:32,  1.44it/s]
Training 1/1 epoch (loss 1.8611):  25%|██▌       | 16/63 [00:19<00:32,  1.44it/s]
Training 1/1 epoch (loss 1.8611):  27%|██▋       | 17/63 [00:19<00:31,  1.46it/s]
Training 1/1 epoch (loss 1.7476):  27%|██▋       | 17/63 [00:20<00:31,  1.46it/s]
Training 1/1 epoch (loss 1.7476):  29%|██▊       | 18/63 [00:20<00:30,  1.49it/s]
Training 1/1 epoch (loss 1.8110):  29%|██▊       | 18/63 [00:21<00:30,  1.49it/s]
Training 1/1 epoch (loss 1.8110):  30%|███       | 19/63 [00:21<00:29,  1.49it/s]
Training 1/1 epoch (loss 1.8342):  30%|███       | 19/63 [00:21<00:29,  1.49it/s]
Training 1/1 epoch (loss 1.8342):  32%|███▏      | 20/63 [00:21<00:28,  1.52it/s]
Training 1/1 epoch (loss 1.7625):  32%|███▏      | 20/63 [00:22<00:28,  1.52it/s]
Training 1/1 epoch (loss 1.7625):  33%|███▎      | 21/63 [00:22<00:27,  1.51it/s]
Training 1/1 epoch (loss 1.7169):  33%|███▎      | 21/63 [00:23<00:27,  1.51it/s]
Training 1/1 epoch (loss 1.7169):  35%|███▍      | 22/63 [00:23<00:27,  1.47it/s]
Training 1/1 epoch (loss 1.7240):  35%|███▍      | 22/63 [00:23<00:27,  1.47it/s]
Training 1/1 epoch (loss 1.7240):  37%|███▋      | 23/63 [00:23<00:26,  1.50it/s]
Training 1/1 epoch (loss 1.7048):  37%|███▋      | 23/63 [00:24<00:26,  1.50it/s]
Training 1/1 epoch (loss 1.7048):  38%|███▊      | 24/63 [00:24<00:26,  1.46it/s]
Training 1/1 epoch (loss 1.6184):  38%|███▊      | 24/63 [00:25<00:26,  1.46it/s]
Training 1/1 epoch (loss 1.6184):  40%|███▉      | 25/63 [00:25<00:25,  1.47it/s]
Training 1/1 epoch (loss 1.6789):  40%|███▉      | 25/63 [00:25<00:25,  1.47it/s]
Training 1/1 epoch (loss 1.6789):  41%|████▏     | 26/63 [00:25<00:24,  1.50it/s]
Training 1/1 epoch (loss 1.7730):  41%|████▏     | 26/63 [00:26<00:24,  1.50it/s]
Training 1/1 epoch (loss 1.7730):  43%|████▎     | 27/63 [00:26<00:23,  1.50it/s]
Training 1/1 epoch (loss 1.7459):  43%|████▎     | 27/63 [00:27<00:23,  1.50it/s]
Training 1/1 epoch (loss 1.7459):  44%|████▍     | 28/63 [00:27<00:23,  1.51it/s]
Training 1/1 epoch (loss 1.7494):  44%|████▍     | 28/63 [00:27<00:23,  1.51it/s]
Training 1/1 epoch (loss 1.7494):  46%|████▌     | 29/63 [00:27<00:22,  1.49it/s]
Training 1/1 epoch (loss 1.6847):  46%|████▌     | 29/63 [00:28<00:22,  1.49it/s]
Training 1/1 epoch (loss 1.6847):  48%|████▊     | 30/63 [00:28<00:23,  1.43it/s]
Training 1/1 epoch (loss 1.8040):  48%|████▊     | 30/63 [00:29<00:23,  1.43it/s]
Training 1/1 epoch (loss 1.8040):  49%|████▉     | 31/63 [00:29<00:22,  1.42it/s]
Training 1/1 epoch (loss 1.7556):  49%|████▉     | 31/63 [00:30<00:22,  1.42it/s]
Training 1/1 epoch (loss 1.7556):  51%|█████     | 32/63 [00:30<00:22,  1.35it/s]
Training 1/1 epoch (loss 1.6802):  51%|█████     | 32/63 [00:31<00:22,  1.35it/s]
Training 1/1 epoch (loss 1.6802):  52%|█████▏    | 33/63 [00:31<00:22,  1.34it/s]
Training 1/1 epoch (loss 1.7203):  52%|█████▏    | 33/63 [00:31<00:22,  1.34it/s]
Training 1/1 epoch (loss 1.7203):  54%|█████▍    | 34/63 [00:31<00:21,  1.33it/s]
Training 1/1 epoch (loss 1.6790):  54%|█████▍    | 34/63 [00:32<00:21,  1.33it/s]
Training 1/1 epoch (loss 1.6790):  56%|█████▌    | 35/63 [00:32<00:20,  1.35it/s]
Training 1/1 epoch (loss 1.7255):  56%|█████▌    | 35/63 [00:33<00:20,  1.35it/s]
Training 1/1 epoch (loss 1.7255):  57%|█████▋    | 36/63 [00:33<00:20,  1.34it/s]
Training 1/1 epoch (loss 1.6554):  57%|█████▋    | 36/63 [00:34<00:20,  1.34it/s]
Training 1/1 epoch (loss 1.6554):  59%|█████▊    | 37/63 [00:34<00:19,  1.34it/s]
Training 1/1 epoch (loss 1.6295):  59%|█████▊    | 37/63 [00:34<00:19,  1.34it/s]
Training 1/1 epoch (loss 1.6295):  60%|██████    | 38/63 [00:34<00:18,  1.35it/s]
Training 1/1 epoch (loss 1.7125):  60%|██████    | 38/63 [00:35<00:18,  1.35it/s]
Training 1/1 epoch (loss 1.7125):  62%|██████▏   | 39/63 [00:35<00:18,  1.33it/s]
Training 1/1 epoch (loss 1.7542):  62%|██████▏   | 39/63 [00:36<00:18,  1.33it/s]
Training 1/1 epoch (loss 1.7542):  63%|██████▎   | 40/63 [00:36<00:17,  1.30it/s]
Training 1/1 epoch (loss 1.7085):  63%|██████▎   | 40/63 [00:37<00:17,  1.30it/s]
Training 1/1 epoch (loss 1.7085):  65%|██████▌   | 41/63 [00:37<00:16,  1.30it/s]
Training 1/1 epoch (loss 1.5987):  65%|██████▌   | 41/63 [00:37<00:16,  1.30it/s]
Training 1/1 epoch (loss 1.5987):  67%|██████▋   | 42/63 [00:37<00:15,  1.32it/s]
Training 1/1 epoch (loss 1.7950):  67%|██████▋   | 42/63 [00:38<00:15,  1.32it/s]
Training 1/1 epoch (loss 1.7950):  68%|██████▊   | 43/63 [00:38<00:15,  1.33it/s]
Training 1/1 epoch (loss 1.6070):  68%|██████▊   | 43/63 [00:39<00:15,  1.33it/s]
Training 1/1 epoch (loss 1.6070):  70%|██████▉   | 44/63 [00:39<00:14,  1.33it/s]
Training 1/1 epoch (loss 1.7500):  70%|██████▉   | 44/63 [00:40<00:14,  1.33it/s]
Training 1/1 epoch (loss 1.7500):  71%|███████▏  | 45/63 [00:40<00:13,  1.33it/s]
Training 1/1 epoch (loss 1.6836):  71%|███████▏  | 45/63 [00:40<00:13,  1.33it/s]
Training 1/1 epoch (loss 1.6836):  73%|███████▎  | 46/63 [00:40<00:12,  1.37it/s]
Training 1/1 epoch (loss 1.6940):  73%|███████▎  | 46/63 [00:41<00:12,  1.37it/s]
Training 1/1 epoch (loss 1.6940):  75%|███████▍  | 47/63 [00:41<00:11,  1.38it/s]
Training 1/1 epoch (loss 1.6777):  75%|███████▍  | 47/63 [00:42<00:11,  1.38it/s]
Training 1/1 epoch (loss 1.6777):  76%|███████▌  | 48/63 [00:42<00:10,  1.39it/s]
Training 1/1 epoch (loss 1.7644):  76%|███████▌  | 48/63 [00:42<00:10,  1.39it/s]
Training 1/1 epoch (loss 1.7644):  78%|███████▊  | 49/63 [00:42<00:10,  1.36it/s]
Training 1/1 epoch (loss 1.6275):  78%|███████▊  | 49/63 [00:43<00:10,  1.36it/s]
Training 1/1 epoch (loss 1.6275):  79%|███████▉  | 50/63 [00:43<00:09,  1.38it/s]
Training 1/1 epoch (loss 1.6958):  79%|███████▉  | 50/63 [00:44<00:09,  1.38it/s]
Training 1/1 epoch (loss 1.6958):  81%|████████  | 51/63 [00:44<00:08,  1.37it/s]
Training 1/1 epoch (loss 1.6001):  81%|████████  | 51/63 [00:45<00:08,  1.37it/s]
Training 1/1 epoch (loss 1.6001):  83%|████████▎ | 52/63 [00:45<00:07,  1.40it/s]
Training 1/1 epoch (loss 1.6260):  83%|████████▎ | 52/63 [00:45<00:07,  1.40it/s]
Training 1/1 epoch (loss 1.6260):  84%|████████▍ | 53/63 [00:45<00:07,  1.42it/s]
Training 1/1 epoch (loss 1.7592):  84%|████████▍ | 53/63 [00:46<00:07,  1.42it/s]
Training 1/1 epoch (loss 1.7592):  86%|████████▌ | 54/63 [00:46<00:06,  1.45it/s]
Training 1/1 epoch (loss 1.7547):  86%|████████▌ | 54/63 [00:47<00:06,  1.45it/s]
Training 1/1 epoch (loss 1.7547):  87%|████████▋ | 55/63 [00:47<00:05,  1.49it/s]
Training 1/1 epoch (loss 1.6434):  87%|████████▋ | 55/63 [00:47<00:05,  1.49it/s]
Training 1/1 epoch (loss 1.6434):  89%|████████▉ | 56/63 [00:47<00:04,  1.46it/s]
Training 1/1 epoch (loss 1.5495):  89%|████████▉ | 56/63 [00:48<00:04,  1.46it/s]
Training 1/1 epoch (loss 1.5495):  90%|█████████ | 57/63 [00:48<00:04,  1.48it/s]
Training 1/1 epoch (loss 1.7184):  90%|█████████ | 57/63 [00:49<00:04,  1.48it/s]
Training 1/1 epoch (loss 1.7184):  92%|█████████▏| 58/63 [00:49<00:03,  1.49it/s]
Training 1/1 epoch (loss 1.7613):  92%|█████████▏| 58/63 [00:49<00:03,  1.49it/s]
Training 1/1 epoch (loss 1.7613):  94%|█████████▎| 59/63 [00:49<00:02,  1.46it/s]
Training 1/1 epoch (loss 1.7492):  94%|█████████▎| 59/63 [00:50<00:02,  1.46it/s]
Training 1/1 epoch (loss 1.7492):  95%|█████████▌| 60/63 [00:50<00:02,  1.46it/s]
Training 1/1 epoch (loss 1.6354):  95%|█████████▌| 60/63 [00:51<00:02,  1.46it/s]
Training 1/1 epoch (loss 1.6354):  97%|█████████▋| 61/63 [00:51<00:01,  1.41it/s]
Training 1/1 epoch (loss 1.6877):  97%|█████████▋| 61/63 [00:51<00:01,  1.41it/s]
Training 1/1 epoch (loss 1.6877):  98%|█████████▊| 62/63 [00:51<00:00,  1.44it/s]
Training 1/1 epoch (loss 1.6930):  98%|█████████▊| 62/63 [00:52<00:00,  1.44it/s]
Training 1/1 epoch (loss 1.6930): 100%|██████████| 63/63 [00:52<00:00,  1.48it/s]
Training 1/1 epoch (loss 1.6930): 100%|██████████| 63/63 [00:52<00:00,  1.20it/s]
tokenizer config file saved in /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k-Q2-2k/tokenizer_config.json
Special tokens file saved in /aifs4su/hansirui_1st/boyuan/resist/setting3-safety/Qwen1.5-4B/Qwen1.5-4B-s3-Q1-10k-Q2-2k/special_tokens_map.json
wandb: ERROR Problem finishing run
Exception ignored in atexit callback: <bound method rank_zero_only.<locals>.wrapper of <safe_rlhf.logger.Logger object at 0x1551041f9150>>
Traceback (most recent call last):
  File "/home/hansirui_1st/jiayi/resist/setting3/safe_rlhf/utils.py", line 212, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/hansirui_1st/jiayi/resist/setting3/safe_rlhf/logger.py", line 183, in close
    self.wandb.finish()
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 449, in wrapper
    return func(self, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 391, in wrapper
    return func(self, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2106, in finish
    return self._finish(exit_code)
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2127, in _finish
    self._atexit_cleanup(exit_code=exit_code)
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2352, in _atexit_cleanup
    self._on_finish()
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/wandb_run.py", line 2609, in _on_finish
    wait_with_progress(
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/mailbox/wait_with_progress.py", line 24, in wait_with_progress
    return wait_all_with_progress(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/mailbox/wait_with_progress.py", line 87, in wait_all_with_progress
    return asyncio_compat.run(progress_loop_with_timeout)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/site-packages/wandb/sdk/lib/asyncio_compat.py", line 27, in run
    future = executor.submit(runner.run, fn)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/aifs4su/hansirui_1st/miniconda3/envs/by-align/lib/python3.11/concurrent/futures/thread.py", line 169, in submit
    raise RuntimeError('cannot schedule new futures after '
RuntimeError: cannot schedule new futures after interpreter shutdown