| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.038461538461538, | |
| "eval_steps": 210, | |
| "global_step": 420, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00019427890345649585, | |
| "loss": 1.0041, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00018831942789034566, | |
| "loss": 0.9074, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00018235995232419546, | |
| "loss": 0.9026, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.0001764004767580453, | |
| "loss": 1.0049, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00017044100119189512, | |
| "loss": 0.722, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00016448152562574496, | |
| "loss": 0.5386, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.00015852205005959479, | |
| "loss": 0.5596, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00015256257449344456, | |
| "loss": 0.5727, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_loss": 0.25414443016052246, | |
| "eval_runtime": 539.3917, | |
| "eval_samples_per_second": 0.384, | |
| "eval_steps_per_second": 0.048, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 0.0001466030989272944, | |
| "loss": 0.3522, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.00014064362336114423, | |
| "loss": 0.2049, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.00013468414779499406, | |
| "loss": 0.2453, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.0001287246722288439, | |
| "loss": 0.2045, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.0001227651966626937, | |
| "loss": 0.1815, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 0.00011680572109654351, | |
| "loss": 0.093, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 0.00011084624553039333, | |
| "loss": 0.1013, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 0.00010488676996424316, | |
| "loss": 0.0648, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "eval_loss": 0.03103998862206936, | |
| "eval_runtime": 535.4733, | |
| "eval_samples_per_second": 0.387, | |
| "eval_steps_per_second": 0.049, | |
| "step": 420 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 840, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 210, | |
| "total_flos": 7.101562243598254e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |