`torch_dtype` is deprecated! Use `dtype` instead!
Loading /workspace/stage2/output/snapshots/step-4000...

Loading weights:   0%|          | 0/399 [00:00<?, ?it/s]
Loading weights:   1%|          | 2/399 [00:00<00:38, 10.28it/s]
Loading weights:   9%|▉         | 37/399 [00:00<00:02, 149.42it/s]
Loading weights:  18%|█▊        | 70/399 [00:00<00:01, 213.84it/s]
Loading weights:  26%|██▌       | 103/399 [00:00<00:01, 250.40it/s]
Loading weights:  34%|███▍      | 136/399 [00:00<00:00, 275.06it/s]
Loading weights:  42%|████▏     | 169/399 [00:00<00:00, 291.20it/s]
Loading weights:  51%|█████     | 203/399 [00:00<00:00, 299.66it/s]
Loading weights:  59%|█████▉    | 237/399 [00:00<00:00, 303.81it/s]
Loading weights:  68%|██████▊   | 270/399 [00:01<00:00, 309.01it/s]
Loading weights:  76%|███████▌  | 303/399 [00:01<00:00, 314.18it/s]
Loading weights:  84%|████████▍ | 336/399 [00:01<00:00, 316.74it/s]
Loading weights:  92%|█████████▏| 369/399 [00:01<00:00, 319.74it/s]
Loading weights: 100%|██████████| 399/399 [00:01<00:00, 283.07it/s]
The following generation flags are not valid and may be ignored: ['temperature', 'top_p', 'top_k']. Set `TRANSFORMERS_VERBOSITY=info` for more details.
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Both `max_new_tokens` (=12) and `max_length`(=40960) seem to have been set. `max_new_tokens` will take precedence. Please refer to the documentation for more information. (https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)
Model: 8.20B params, vocab=152696
Reco samples in val: 204874
Eval items: 1000
  [32/1000] 1s elapsed, ETA 16s
  [64/1000] 1s elapsed, ETA 13s
  [96/1000] 1s elapsed, ETA 12s
  [128/1000] 2s elapsed, ETA 11s
  [160/1000] 2s elapsed, ETA 10s
  [192/1000] 2s elapsed, ETA 9s
  [224/1000] 3s elapsed, ETA 9s
  [256/1000] 3s elapsed, ETA 9s
  [288/1000] 3s elapsed, ETA 8s
  [320/1000] 4s elapsed, ETA 8s
  [352/1000] 4s elapsed, ETA 7s
  [384/1000] 4s elapsed, ETA 7s
  [416/1000] 5s elapsed, ETA 7s
  [448/1000] 5s elapsed, ETA 6s
  [480/1000] 5s elapsed, ETA 6s
  [512/1000] 6s elapsed, ETA 5s
  [544/1000] 6s elapsed, ETA 5s
  [576/1000] 6s elapsed, ETA 5s
  [608/1000] 7s elapsed, ETA 4s
  [640/1000] 7s elapsed, ETA 4s
  [672/1000] 7s elapsed, ETA 4s
  [704/1000] 8s elapsed, ETA 3s
  [736/1000] 8s elapsed, ETA 3s
  [768/1000] 8s elapsed, ETA 3s
  [800/1000] 9s elapsed, ETA 2s
  [832/1000] 9s elapsed, ETA 2s
  [864/1000] 9s elapsed, ETA 1s
  [896/1000] 10s elapsed, ETA 1s
  [928/1000] 10s elapsed, ETA 1s
  [960/1000] 10s elapsed, ETA 0s
  [992/1000] 11s elapsed, ETA 0s
  [1000/1000] 11s elapsed, ETA 0s

=== Results (n=1000, time=11s) ===
  valid_format: 100.0%
  level_A:      7.0%
  level_AB:     0.1%
  level_ABC:    0.0%
  exact (Hit@1):0.00%

=== By type ===
  copurchase_backward            n= 357  valid=100.0%  A=  3.9%  exact= 0.00%
  copurchase_forward             n= 361  valid=100.0%  A=  7.8%  exact= 0.00%
  seq_last_2                     n= 124  valid=100.0%  A=  7.3%  exact= 0.00%
  seq_last_3                     n= 104  valid=100.0%  A= 11.5%  exact= 0.00%
  seq_last_5                     n=  54  valid=100.0%  A= 13.0%  exact= 0.00%