8k_v3/code/wandb/debug.log

2026-05-03 21:16:31,573 INFO    MainThread:312947 [wandb_setup.py:_flush():67] Current SDK version is 0.19.8
2026-05-03 21:16:31,573 INFO    MainThread:312947 [wandb_setup.py:_flush():67] Configure stats pid to 312947
2026-05-03 21:16:31,573 INFO    MainThread:312947 [wandb_setup.py:_flush():67] Loading settings from /admin/home/craffel/.config/wandb/settings
2026-05-03 21:16:31,573 INFO    MainThread:312947 [wandb_setup.py:_flush():67] Loading settings from /fsx/craffel/lingua_logs/8k_v3/code/wandb/settings
2026-05-03 21:16:31,573 INFO    MainThread:312947 [wandb_setup.py:_flush():67] Loading settings from environment variables
2026-05-03 21:16:31,573 INFO    MainThread:312947 [wandb_init.py:setup_run_log_directory():647] Logging user logs to /fsx/craffel/lingua_logs/8k_v3/code/wandb/run-20260503_211631-d2owlw9s/logs/debug.log
2026-05-03 21:16:31,574 INFO    MainThread:312947 [wandb_init.py:setup_run_log_directory():648] Logging internal logs to /fsx/craffel/lingua_logs/8k_v3/code/wandb/run-20260503_211631-d2owlw9s/logs/debug-internal.log
2026-05-03 21:16:31,574 INFO    MainThread:312947 [wandb_init.py:init():761] calling init triggers
2026-05-03 21:16:31,574 INFO    MainThread:312947 [wandb_init.py:init():766] wandb.init called with sweep_config: {}
config: {'name': 'evals', 'dump_dir': '/fsx/craffel/lingua_logs/8k_v3/evals/0000080000', 'metric_log_dir': '/fsx/craffel/lingua_logs/8k_v3', 'ckpt_dir': '/fsx/craffel/lingua_logs/checkpoints/8k_v3/0000080000', 'generator': {'temperature': 0.0, 'top_p': None, 'top_k': None, 'max_gen_len': 512, 'max_tokens': 16384, 'max_prompt_len': None, 'until': [], 'compile_prefilling': False, 'reduce_generation_overhead': False, 'show_progress': False, 'dtype': 'bf16', 'device': 'cuda', 'add_bos': False}, 'harness': {'tasks': ['hellaswag', 'piqa', 'arc_easy', 'arc_challenge', 'include_base_44_arabic', 'include_base_44_chinese', 'include_base_44_german', 'include_base_44_greek', 'include_base_44_persian', 'include_base_44_french', 'include_base_44_hungarian', 'include_base_44_indonesian', 'include_base_44_italian', 'include_base_44_japanese', 'include_base_44_dutch', 'include_base_44_polish', 'include_base_44_portuguese', 'include_base_44_russian', 'include_base_44_spanish', 'include_base_44_turkish', 'include_base_44_vietnamese', 'belebele_arb_Arab', 'belebele_ces_Latn', 'belebele_zho_Hans', 'belebele_dan_Latn', 'belebele_deu_Latn', 'belebele_ell_Grek', 'belebele_pes_Arab', 'belebele_fra_Latn', 'belebele_hun_Latn', 'belebele_ind_Latn', 'belebele_ita_Latn', 'belebele_jpn_Jpan', 'belebele_nld_Latn', 'belebele_pol_Latn', 'belebele_por_Latn', 'belebele_rus_Cyrl', 'belebele_spa_Latn', 'belebele_swe_Latn', 'belebele_tur_Latn', 'belebele_vie_Latn', 'belebele_eng_Latn', 'xnli_ar', 'xnli_zh', 'xnli_de', 'xnli_el', 'xnli_en', 'xnli_es', 'xnli_fr', 'xnli_ru', 'xnli_tr', 'xnli_vi'], 'num_fewshot': None, 'device': None, 'use_cache': None, 'cache_requests': False, 'rewrite_requests_cache': False, 'delete_requests_cache': False, 'limit': None, 'bootstrap_iters': 100000, 'check_integrity': False, 'write_out': False, 'log_samples': True, 'system_instruction': None, 'apply_chat_template': False, 'fewshot_as_multiturn': False, 'gen_kwargs': None, 'predict_only': False, 'random_seed': 0, 'numpy_random_seed': 1234, 'torch_random_seed': 1234, 'fewshot_random_seed': 1234}, 'validation': None, 'wandb': {'job_type': None, 'dir': None, 'project': 'MoTo-Training', 'entity': 'raffel-reports', 'tags': None, 'group': None, 'name': '8k_v3', 'notes': None, 'config_exclude_keys': None, 'config_include_keys': None, 'anonymous': None, 'mode': None, 'allow_val_change': None, 'resume': None, 'force': None, 'tensorboard': None, 'sync_tensorboard': None, 'monitor_gym': None, 'save_code': None, 'id': None, 'fork_from': None, 'resume_from': None}, 'global_step': 80000, 'tokenizer': None, 'routing': None, '_wandb': {}}
2026-05-03 21:16:31,574 INFO    MainThread:312947 [wandb_init.py:init():784] starting backend
2026-05-03 21:16:31,574 INFO    MainThread:312947 [wandb_init.py:init():788] sending inform_init request
2026-05-03 21:16:31,603 INFO    MainThread:312947 [backend.py:_multiprocessing_setup():101] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
2026-05-03 21:16:31,603 INFO    MainThread:312947 [wandb_init.py:init():798] backend started and connected
2026-05-03 21:16:31,607 INFO    MainThread:312947 [wandb_init.py:init():891] updated telemetry
2026-05-03 21:16:31,631 INFO    MainThread:312947 [wandb_init.py:init():915] communicating run to backend with 90.0 second timeout
2026-05-03 21:16:32,244 INFO    MainThread:312947 [wandb_init.py:init():990] starting run threads in backend
2026-05-03 21:16:34,744 INFO    MainThread:312947 [wandb_run.py:_console_start():2375] atexit reg
2026-05-03 21:16:34,744 INFO    MainThread:312947 [wandb_run.py:_redirect():2227] redirect: wrap_raw
2026-05-03 21:16:34,744 INFO    MainThread:312947 [wandb_run.py:_redirect():2292] Wrapping output streams.
2026-05-03 21:16:34,745 INFO    MainThread:312947 [wandb_run.py:_redirect():2315] Redirects installed.
2026-05-03 21:16:34,809 INFO    MainThread:312947 [wandb_init.py:init():1032] run started, returning control to user process