| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.99660441426146, | |
| "eval_steps": 50, | |
| "global_step": 294, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1891, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1472, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1311, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1016, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0974, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_loss": 1.142778754234314, | |
| "eval_runtime": 15.973, | |
| "eval_samples_per_second": 16.403, | |
| "eval_steps_per_second": 4.132, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0897, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0727, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1307, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1089, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0626, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 1.1251928806304932, | |
| "eval_runtime": 15.965, | |
| "eval_samples_per_second": 16.411, | |
| "eval_steps_per_second": 4.134, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 5e-05, | |
| "loss": 1.041, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1059, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0942, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0676, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0209, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.1206752061843872, | |
| "eval_runtime": 16.0175, | |
| "eval_samples_per_second": 16.357, | |
| "eval_steps_per_second": 4.12, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9453, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0044, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9175, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8933, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9509, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 1.142478346824646, | |
| "eval_runtime": 15.9666, | |
| "eval_samples_per_second": 16.409, | |
| "eval_steps_per_second": 4.134, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9252, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 5e-05, | |
| "loss": 0.841, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9285, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9325, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9616, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 1.1395381689071655, | |
| "eval_runtime": 15.961, | |
| "eval_samples_per_second": 16.415, | |
| "eval_steps_per_second": 4.135, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 5e-05, | |
| "loss": 0.896, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9771, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8961, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8632, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 294, | |
| "total_flos": 7.196690777643418e+16, | |
| "train_loss": 1.01067645939029, | |
| "train_runtime": 877.5897, | |
| "train_samples_per_second": 5.369, | |
| "train_steps_per_second": 0.335 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 294, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 250, | |
| "total_flos": 7.196690777643418e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |