| { | |
| "best_metric": 1.2377259731292725, | |
| "best_model_checkpoint": "./outputs/checkpoint-2000", | |
| "epoch": 2.6666666666666665, | |
| "eval_steps": 100, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4887, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 1.360513687133789, | |
| "eval_runtime": 127.6803, | |
| "eval_samples_per_second": 15.116, | |
| "eval_steps_per_second": 1.895, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1616, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 1.3187366724014282, | |
| "eval_runtime": 126.439, | |
| "eval_samples_per_second": 15.264, | |
| "eval_steps_per_second": 1.914, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1374, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 1.3053778409957886, | |
| "eval_runtime": 126.588, | |
| "eval_samples_per_second": 15.246, | |
| "eval_steps_per_second": 1.912, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002, | |
| "loss": 1.125, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 1.2945491075515747, | |
| "eval_runtime": 126.4785, | |
| "eval_samples_per_second": 15.26, | |
| "eval_steps_per_second": 1.913, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1119, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 1.2894083261489868, | |
| "eval_runtime": 126.4649, | |
| "eval_samples_per_second": 15.261, | |
| "eval_steps_per_second": 1.914, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1142, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.2811017036437988, | |
| "eval_runtime": 126.9356, | |
| "eval_samples_per_second": 15.205, | |
| "eval_steps_per_second": 1.906, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1078, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 1.2782505750656128, | |
| "eval_runtime": 126.2875, | |
| "eval_samples_per_second": 15.283, | |
| "eval_steps_per_second": 1.916, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0976, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 1.2754321098327637, | |
| "eval_runtime": 126.6616, | |
| "eval_samples_per_second": 15.237, | |
| "eval_steps_per_second": 1.911, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0956, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 1.2672399282455444, | |
| "eval_runtime": 126.5881, | |
| "eval_samples_per_second": 15.246, | |
| "eval_steps_per_second": 1.912, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0927, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 1.266547441482544, | |
| "eval_runtime": 126.4531, | |
| "eval_samples_per_second": 15.263, | |
| "eval_steps_per_second": 1.914, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0885, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_loss": 1.2609761953353882, | |
| "eval_runtime": 126.7802, | |
| "eval_samples_per_second": 15.223, | |
| "eval_steps_per_second": 1.909, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0775, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.2579678297042847, | |
| "eval_runtime": 126.3602, | |
| "eval_samples_per_second": 15.274, | |
| "eval_steps_per_second": 1.915, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0837, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_loss": 1.2565162181854248, | |
| "eval_runtime": 126.7636, | |
| "eval_samples_per_second": 15.225, | |
| "eval_steps_per_second": 1.909, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0845, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_loss": 1.2504855394363403, | |
| "eval_runtime": 126.8158, | |
| "eval_samples_per_second": 15.219, | |
| "eval_steps_per_second": 1.908, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0781, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.252923607826233, | |
| "eval_runtime": 127.0578, | |
| "eval_samples_per_second": 15.19, | |
| "eval_steps_per_second": 1.905, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0761, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_loss": 1.2466760873794556, | |
| "eval_runtime": 126.5255, | |
| "eval_samples_per_second": 15.254, | |
| "eval_steps_per_second": 1.913, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0726, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_loss": 1.2422055006027222, | |
| "eval_runtime": 126.6317, | |
| "eval_samples_per_second": 15.241, | |
| "eval_steps_per_second": 1.911, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0666, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 1.2425215244293213, | |
| "eval_runtime": 126.4267, | |
| "eval_samples_per_second": 15.266, | |
| "eval_steps_per_second": 1.914, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0601, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 1.2401798963546753, | |
| "eval_runtime": 126.5967, | |
| "eval_samples_per_second": 15.245, | |
| "eval_steps_per_second": 1.912, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0715, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 1.2377259731292725, | |
| "eval_runtime": 126.6342, | |
| "eval_samples_per_second": 15.241, | |
| "eval_steps_per_second": 1.911, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 7.090695034659226e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |