| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 3540, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.211864406779661, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 0.00016949152542372882, | |
| "loss": 1.4882, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.423728813559322, | |
| "grad_norm": 1.0703125, | |
| "learning_rate": 0.00019934060086830688, | |
| "loss": 1.1689, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.635593220338983, | |
| "grad_norm": 1.09375, | |
| "learning_rate": 0.00019676565602164643, | |
| "loss": 1.1984, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.847457627118644, | |
| "grad_norm": 0.9140625, | |
| "learning_rate": 0.00019229383471353536, | |
| "loss": 1.1429, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0593220338983051, | |
| "grad_norm": 1.0703125, | |
| "learning_rate": 0.00018601279710076812, | |
| "loss": 1.1166, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.271186440677966, | |
| "grad_norm": 1.0, | |
| "learning_rate": 0.00017804566902045448, | |
| "loss": 1.0057, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4830508474576272, | |
| "grad_norm": 0.796875, | |
| "learning_rate": 0.00016854862838056546, | |
| "loss": 1.003, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.694915254237288, | |
| "grad_norm": 0.68359375, | |
| "learning_rate": 0.0001577078436378025, | |
| "loss": 0.9752, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.9067796610169492, | |
| "grad_norm": 0.8125, | |
| "learning_rate": 0.00014573582437715308, | |
| "loss": 1.0262, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.1186440677966103, | |
| "grad_norm": 0.85546875, | |
| "learning_rate": 0.00013286725553194177, | |
| "loss": 0.9129, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.330508474576271, | |
| "grad_norm": 1.453125, | |
| "learning_rate": 0.00011935439690527049, | |
| "loss": 0.8845, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.542372881355932, | |
| "grad_norm": 0.78515625, | |
| "learning_rate": 0.0001054621381750442, | |
| "loss": 0.883, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.7542372881355934, | |
| "grad_norm": 0.9296875, | |
| "learning_rate": 9.146280631826239e-05, | |
| "loss": 0.8786, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.9661016949152543, | |
| "grad_norm": 1.125, | |
| "learning_rate": 7.763082724352662e-05, | |
| "loss": 0.9204, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.1779661016949152, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 6.423734627864077e-05, | |
| "loss": 0.8706, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 3.389830508474576, | |
| "grad_norm": 0.671875, | |
| "learning_rate": 5.1544912966734994e-05, | |
| "loss": 0.8667, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.601694915254237, | |
| "grad_norm": 0.92578125, | |
| "learning_rate": 3.980233436371869e-05, | |
| "loss": 0.8416, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 3.8135593220338984, | |
| "grad_norm": 0.7890625, | |
| "learning_rate": 2.9239797726771324e-05, | |
| "loss": 0.8213, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 4.02542372881356, | |
| "grad_norm": 0.734375, | |
| "learning_rate": 2.006435820276602e-05, | |
| "loss": 0.8559, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 4.237288135593221, | |
| "grad_norm": 0.953125, | |
| "learning_rate": 1.2455879970504569e-05, | |
| "loss": 0.8274, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.4491525423728815, | |
| "grad_norm": 0.84375, | |
| "learning_rate": 6.563510401684669e-06, | |
| "loss": 0.831, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 4.661016949152542, | |
| "grad_norm": 1.0859375, | |
| "learning_rate": 2.50275635686813e-06, | |
| "loss": 0.831, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.872881355932203, | |
| "grad_norm": 0.7265625, | |
| "learning_rate": 3.5321992919458146e-07, | |
| "loss": 0.8406, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 3540, | |
| "total_flos": 8.16978613968e+16, | |
| "train_loss": 0.9628061628611074, | |
| "train_runtime": 7816.2612, | |
| "train_samples_per_second": 3.621, | |
| "train_steps_per_second": 0.453 | |
| } | |
| ], | |
| "logging_steps": 150, | |
| "max_steps": 3540, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.16978613968e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |