| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0040551500405515, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.110300081103001e-05, | |
| "eval_loss": 2.1451873779296875, | |
| "eval_runtime": 238.0687, | |
| "eval_samples_per_second": 5.755, | |
| "eval_steps_per_second": 5.755, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00040551500405515005, | |
| "grad_norm": 1.3119515180587769, | |
| "learning_rate": 5e-05, | |
| "loss": 2.1309, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0008110300081103001, | |
| "grad_norm": 32.844932556152344, | |
| "learning_rate": 0.0001, | |
| "loss": 1.8094, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0012165450121654502, | |
| "grad_norm": 1.7279303073883057, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.1586, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0016220600162206002, | |
| "grad_norm": 0.7565182447433472, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.2413, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00202757502027575, | |
| "grad_norm": 35.2607307434082, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.1916, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.00202757502027575, | |
| "eval_loss": 1.5653172731399536, | |
| "eval_runtime": 235.9922, | |
| "eval_samples_per_second": 5.805, | |
| "eval_steps_per_second": 5.805, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0024330900243309003, | |
| "grad_norm": 3.194694995880127, | |
| "learning_rate": 5e-05, | |
| "loss": 1.216, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00283860502838605, | |
| "grad_norm": 0.823891818523407, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.5, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0032441200324412004, | |
| "grad_norm": 0.46792903542518616, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.209, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0036496350364963502, | |
| "grad_norm": 0.7259841561317444, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.5031, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0040551500405515, | |
| "grad_norm": 81.30354309082031, | |
| "learning_rate": 0.0, | |
| "loss": 1.4401, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0040551500405515, | |
| "eval_loss": 1.422434687614441, | |
| "eval_runtime": 236.3994, | |
| "eval_samples_per_second": 5.795, | |
| "eval_steps_per_second": 5.795, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 643355482521600.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |