| { | |
| "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": 1.1773, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1468, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1279, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1021, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0965, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_loss": 1.1430302858352661, | |
| "eval_runtime": 15.7711, | |
| "eval_samples_per_second": 16.613, | |
| "eval_steps_per_second": 4.185, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0874, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0742, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1263, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1085, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0622, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 1.1251620054244995, | |
| "eval_runtime": 15.7412, | |
| "eval_samples_per_second": 16.644, | |
| "eval_steps_per_second": 4.193, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0416, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5e-05, | |
| "loss": 1.102, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0927, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0662, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 5e-05, | |
| "loss": 1.022, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.1217572689056396, | |
| "eval_runtime": 15.7347, | |
| "eval_samples_per_second": 16.651, | |
| "eval_steps_per_second": 4.195, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9443, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9185, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8956, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9513, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 1.143324613571167, | |
| "eval_runtime": 15.8072, | |
| "eval_samples_per_second": 16.575, | |
| "eval_steps_per_second": 4.175, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9233, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8422, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 5e-05, | |
| "loss": 0.928, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9318, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9607, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 1.1436448097229004, | |
| "eval_runtime": 15.7943, | |
| "eval_samples_per_second": 16.588, | |
| "eval_steps_per_second": 4.179, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9034, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 5e-05, | |
| "loss": 0.977, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8984, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8655, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 294, | |
| "total_flos": 7.021903495711949e+16, | |
| "train_loss": 1.0100539411817278, | |
| "train_runtime": 867.3678, | |
| "train_samples_per_second": 5.433, | |
| "train_steps_per_second": 0.339 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 294, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 250, | |
| "total_flos": 7.021903495711949e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |