| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 3125, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 2.823965072631836, | |
| "learning_rate": 1.68064e-05, | |
| "loss": 0.4565, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9214636932551902, | |
| "eval_f1": 0.806775887856969, | |
| "eval_loss": 0.26512816548347473, | |
| "eval_precision": 0.7941518865986441, | |
| "eval_recall": 0.8198077194966775, | |
| "eval_runtime": 37.78, | |
| "eval_samples_per_second": 264.691, | |
| "eval_steps_per_second": 33.086, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 5.351424217224121, | |
| "learning_rate": 1.3606400000000002e-05, | |
| "loss": 0.2612, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9257350749975166, | |
| "eval_f1": 0.8162128281913782, | |
| "eval_loss": 0.2490304708480835, | |
| "eval_precision": 0.8042821850123525, | |
| "eval_recall": 0.8285027569630992, | |
| "eval_runtime": 38.4394, | |
| "eval_samples_per_second": 260.15, | |
| "eval_steps_per_second": 32.519, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 3.4635002613067627, | |
| "learning_rate": 1.0406400000000001e-05, | |
| "loss": 0.2184, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9293856163703189, | |
| "eval_f1": 0.8262937062937062, | |
| "eval_loss": 0.2470722198486328, | |
| "eval_precision": 0.8174899681748997, | |
| "eval_recall": 0.8352891276685989, | |
| "eval_runtime": 38.5739, | |
| "eval_samples_per_second": 259.242, | |
| "eval_steps_per_second": 32.405, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 2.8742923736572266, | |
| "learning_rate": 7.2064e-06, | |
| "loss": 0.1874, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.9541894197463989, | |
| "learning_rate": 4.012800000000001e-06, | |
| "loss": 0.1636, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9307638819906625, | |
| "eval_f1": 0.8312837484758753, | |
| "eval_loss": 0.2493128478527069, | |
| "eval_precision": 0.819493097053369, | |
| "eval_recall": 0.8434186342428955, | |
| "eval_runtime": 36.2649, | |
| "eval_samples_per_second": 275.749, | |
| "eval_steps_per_second": 34.469, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 3.6088337898254395, | |
| "learning_rate": 8.128000000000001e-07, | |
| "loss": 0.1408, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9318317274262442, | |
| "eval_f1": 0.8372271964185787, | |
| "eval_loss": 0.2519814968109131, | |
| "eval_precision": 0.8285338502007477, | |
| "eval_recall": 0.8461049059804892, | |
| "eval_runtime": 37.7598, | |
| "eval_samples_per_second": 264.832, | |
| "eval_steps_per_second": 33.104, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 3125, | |
| "total_flos": 1384127485887552.0, | |
| "train_loss": 0.23445223083496095, | |
| "train_runtime": 2722.1951, | |
| "train_samples_per_second": 36.735, | |
| "train_steps_per_second": 1.148 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3125, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1384127485887552.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |