| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 7647, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.8038446449588075e-05, | |
| "loss": 2.5909, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 2.6076892899176145e-05, | |
| "loss": 2.3461, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.4115339348764223e-05, | |
| "loss": 2.2111, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 2.2153785798352297e-05, | |
| "loss": 2.1523, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 2.019223224794037e-05, | |
| "loss": 2.0715, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.9488946199417114, | |
| "eval_runtime": 854.0815, | |
| "eval_samples_per_second": 2.984, | |
| "eval_steps_per_second": 0.747, | |
| "step": 2549 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 1.8230678697528442e-05, | |
| "loss": 1.4507, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.6269125147116516e-05, | |
| "loss": 1.3564, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.430757159670459e-05, | |
| "loss": 1.3545, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 1.2346018046292664e-05, | |
| "loss": 1.3372, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.0384464495880738e-05, | |
| "loss": 1.3039, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.8286019563674927, | |
| "eval_runtime": 849.7166, | |
| "eval_samples_per_second": 3.0, | |
| "eval_steps_per_second": 0.751, | |
| "step": 5098 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 8.42291094546881e-06, | |
| "loss": 0.987, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 6.461357395056885e-06, | |
| "loss": 0.8986, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.4998038446449585e-06, | |
| "loss": 0.9008, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 2.538250294233033e-06, | |
| "loss": 0.8903, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 5.766967438211064e-07, | |
| "loss": 0.8313, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.8424451351165771, | |
| "eval_runtime": 846.7842, | |
| "eval_samples_per_second": 3.01, | |
| "eval_steps_per_second": 0.753, | |
| "step": 7647 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7647, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.6566780520562688e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |