| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.981366459627329, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.6218, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_accuracy": 0.7417102966841187, | |
| "eval_loss": 0.573567271232605, | |
| "eval_runtime": 9.2285, | |
| "eval_samples_per_second": 62.09, | |
| "eval_steps_per_second": 1.95, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.9326121764495596e-05, | |
| "loss": 0.6103, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.7399650959860384, | |
| "eval_loss": 0.5730276703834534, | |
| "eval_runtime": 9.2303, | |
| "eval_samples_per_second": 62.078, | |
| "eval_steps_per_second": 1.95, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.665063509461097e-05, | |
| "loss": 0.6105, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_accuracy": 0.7277486910994765, | |
| "eval_loss": 0.5863298177719116, | |
| "eval_runtime": 9.2823, | |
| "eval_samples_per_second": 61.73, | |
| "eval_steps_per_second": 1.939, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.215604094671835e-05, | |
| "loss": 0.6261, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.7068062827225131, | |
| "eval_loss": 0.5964760184288025, | |
| "eval_runtime": 9.2516, | |
| "eval_samples_per_second": 61.935, | |
| "eval_steps_per_second": 1.946, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.621997950501156e-05, | |
| "loss": 0.5684, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_accuracy": 0.7521815008726004, | |
| "eval_loss": 0.567564845085144, | |
| "eval_runtime": 10.2074, | |
| "eval_samples_per_second": 56.136, | |
| "eval_steps_per_second": 1.763, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 2.9341204441673266e-05, | |
| "loss": 0.5878, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_accuracy": 0.6561954624781849, | |
| "eval_loss": 0.6583427786827087, | |
| "eval_runtime": 9.2811, | |
| "eval_samples_per_second": 61.739, | |
| "eval_steps_per_second": 1.939, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.2097677146869242e-05, | |
| "loss": 0.5274, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_accuracy": 0.7521815008726004, | |
| "eval_loss": 0.5735621452331543, | |
| "eval_runtime": 9.3011, | |
| "eval_samples_per_second": 61.605, | |
| "eval_steps_per_second": 1.935, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.509800584902108e-05, | |
| "loss": 0.581, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.7399650959860384, | |
| "eval_loss": 0.5575574636459351, | |
| "eval_runtime": 9.3052, | |
| "eval_samples_per_second": 61.579, | |
| "eval_steps_per_second": 1.934, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.527, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_accuracy": 0.7539267015706806, | |
| "eval_loss": 0.5575008988380432, | |
| "eval_runtime": 9.355, | |
| "eval_samples_per_second": 61.251, | |
| "eval_steps_per_second": 1.924, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 4.112804714676594e-06, | |
| "loss": 0.5228, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_accuracy": 0.7521815008726004, | |
| "eval_loss": 0.5663809776306152, | |
| "eval_runtime": 9.3904, | |
| "eval_samples_per_second": 61.02, | |
| "eval_steps_per_second": 1.917, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 1.0502621921127776e-06, | |
| "loss": 0.4978, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "eval_accuracy": 0.7539267015706806, | |
| "eval_loss": 0.5707576274871826, | |
| "eval_runtime": 9.7621, | |
| "eval_samples_per_second": 58.697, | |
| "eval_steps_per_second": 1.844, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 0.0, | |
| "loss": 0.5051, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.7469458987783595, | |
| "eval_loss": 0.5715910792350769, | |
| "eval_runtime": 9.7356, | |
| "eval_samples_per_second": 58.856, | |
| "eval_steps_per_second": 1.849, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "step": 120, | |
| "total_flos": 3.5051816769866957e+18, | |
| "train_loss": 0.5655017614364624, | |
| "train_runtime": 483.5766, | |
| "train_samples_per_second": 31.962, | |
| "train_steps_per_second": 0.248 | |
| } | |
| ], | |
| "max_steps": 120, | |
| "num_train_epochs": 3, | |
| "total_flos": 3.5051816769866957e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |