| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9936305732484076, | |
| "eval_steps": 500, | |
| "global_step": 78, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.012738853503184714, | |
| "grad_norm": 9.419306860000189, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.2082, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06369426751592357, | |
| "grad_norm": 11.154561988265748, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.1572, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12738853503184713, | |
| "grad_norm": 0.9580718367338527, | |
| "learning_rate": 1.9959742939952393e-05, | |
| "loss": 0.926, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1910828025477707, | |
| "grad_norm": 0.6962618582606526, | |
| "learning_rate": 1.9510565162951538e-05, | |
| "loss": 0.8462, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.25477707006369427, | |
| "grad_norm": 0.652360226118771, | |
| "learning_rate": 1.8584487936018663e-05, | |
| "loss": 0.8207, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3184713375796178, | |
| "grad_norm": 0.6036498850889884, | |
| "learning_rate": 1.7227948638273918e-05, | |
| "loss": 0.8328, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3821656050955414, | |
| "grad_norm": 0.6106437850436714, | |
| "learning_rate": 1.5508969814521026e-05, | |
| "loss": 0.8184, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.445859872611465, | |
| "grad_norm": 0.6112027883233256, | |
| "learning_rate": 1.3513748240813429e-05, | |
| "loss": 0.829, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5095541401273885, | |
| "grad_norm": 0.5896867860825263, | |
| "learning_rate": 1.1342332658176556e-05, | |
| "loss": 0.8223, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5732484076433121, | |
| "grad_norm": 0.6019677830367037, | |
| "learning_rate": 9.103606910965666e-06, | |
| "loss": 0.7841, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6369426751592356, | |
| "grad_norm": 0.5830221008800777, | |
| "learning_rate": 6.909830056250527e-06, | |
| "loss": 0.8322, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7006369426751592, | |
| "grad_norm": 0.5839501242867685, | |
| "learning_rate": 4.87100722594094e-06, | |
| "loss": 0.8069, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.7643312101910829, | |
| "grad_norm": 0.5957554055545055, | |
| "learning_rate": 3.089373510131354e-06, | |
| "loss": 0.7856, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8280254777070064, | |
| "grad_norm": 0.5870067617046513, | |
| "learning_rate": 1.6542674627869738e-06, | |
| "loss": 0.8141, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.89171974522293, | |
| "grad_norm": 0.557541376858347, | |
| "learning_rate": 6.37651293602628e-07, | |
| "loss": 0.7947, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9554140127388535, | |
| "grad_norm": 0.5998913867246424, | |
| "learning_rate": 9.0502382320653e-08, | |
| "loss": 0.8347, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.9936305732484076, | |
| "eval_loss": 0.8263773322105408, | |
| "eval_runtime": 9.6507, | |
| "eval_samples_per_second": 51.81, | |
| "eval_steps_per_second": 1.658, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.9936305732484076, | |
| "step": 78, | |
| "total_flos": 11234023833600.0, | |
| "train_loss": 0.8476498554914426, | |
| "train_runtime": 656.6949, | |
| "train_samples_per_second": 15.228, | |
| "train_steps_per_second": 0.119 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 78, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 11234023833600.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |