| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_setup.py:_flush():79] Current SDK version is 0.18.6 | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_setup.py:_flush():79] Configure stats pid to 16680 | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_setup.py:_flush():79] Loading settings from C:\Users\silxs\.config\wandb\settings | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_setup.py:_flush():79] Loading settings from C:\sakana\enwik8-model\wandb\settings | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_setup.py:_flush():79] Loading settings from environment variables: {} | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': None, '_disable_service': None} | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'train_dtat.py', 'program_abspath': 'C:\\sakana\\enwik8-model\\train_dtat.py', 'program': 'C:\\sakana\\enwik8-model\\train_dtat.py'} | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_setup.py:_flush():79] Applying login settings: {} | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_init.py:_log_setup():533] Logging user logs to C:\sakana\enwik8-model\wandb\run-20241230_125819-geso4xvw\logs\debug.log | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_init.py:_log_setup():534] Logging internal logs to C:\sakana\enwik8-model\wandb\run-20241230_125819-geso4xvw\logs\debug-internal.log | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_init.py:init():619] calling init triggers | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_init.py:init():626] wandb.init called with sweep_config: {} | |
| config: {'architecture': 'DTAT', 'dataset': 'enwik8', 'batch_size': 32, 'learning_rate': 0.0006, 'warmup_iters': 2000, 'block_size': 1024, 'n_layer': 12, 'n_head': 8, 'n_embd': 512, 'dropout': 0.1, 'sparse_topk': 32} | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_init.py:init():669] starting backend | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [wandb_init.py:init():673] sending inform_init request | |
| 2024-12-30 12:58:19,921 INFO MainThread:16680 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=spawn, using: spawn | |
| 2024-12-30 12:58:19,924 INFO MainThread:16680 [wandb_init.py:init():686] backend started and connected | |
| 2024-12-30 12:58:19,927 INFO MainThread:16680 [wandb_init.py:init():781] updated telemetry | |
| 2024-12-30 12:58:19,977 INFO MainThread:16680 [wandb_init.py:init():814] communicating run to backend with 90.0 second timeout | |
| 2024-12-30 12:58:20,892 INFO MainThread:16680 [wandb_init.py:init():867] starting run threads in backend | |
| 2024-12-30 12:58:21,272 INFO MainThread:16680 [wandb_run.py:_console_start():2451] atexit reg | |
| 2024-12-30 12:58:21,272 INFO MainThread:16680 [wandb_run.py:_redirect():2299] redirect: wrap_raw | |
| 2024-12-30 12:58:21,272 INFO MainThread:16680 [wandb_run.py:_redirect():2364] Wrapping output streams. | |
| 2024-12-30 12:58:21,272 INFO MainThread:16680 [wandb_run.py:_redirect():2389] Redirects installed. | |
| 2024-12-30 12:58:21,277 INFO MainThread:16680 [wandb_init.py:init():911] run started, returning control to user process | |
| 2024-12-30 12:58:21,889 WARNING MsgRouterThr:16680 [router.py:message_loop():75] message_loop has been closed | |