| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.975609756097561, | |
| "eval_steps": 500, | |
| "global_step": 15, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06504065040650407, | |
| "grad_norm": 0.5584165453910828, | |
| "learning_rate": 4.9453690018345144e-05, | |
| "loss": 0.7344, | |
| "num_input_tokens_seen": 2097152, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.13008130081300814, | |
| "grad_norm": 0.4928242862224579, | |
| "learning_rate": 4.783863644106502e-05, | |
| "loss": 0.7113, | |
| "num_input_tokens_seen": 4194304, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1951219512195122, | |
| "grad_norm": 0.4566553831100464, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 0.7009, | |
| "num_input_tokens_seen": 6291456, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2601626016260163, | |
| "grad_norm": 0.38513678312301636, | |
| "learning_rate": 4.172826515897146e-05, | |
| "loss": 0.6704, | |
| "num_input_tokens_seen": 8388608, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.3252032520325203, | |
| "grad_norm": 0.36934641003608704, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.6504, | |
| "num_input_tokens_seen": 10485760, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3902439024390244, | |
| "grad_norm": 0.33891424536705017, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 0.6217, | |
| "num_input_tokens_seen": 12582912, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.45528455284552843, | |
| "grad_norm": 0.30929532647132874, | |
| "learning_rate": 2.761321158169134e-05, | |
| "loss": 0.6052, | |
| "num_input_tokens_seen": 14680064, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5203252032520326, | |
| "grad_norm": 0.28951436281204224, | |
| "learning_rate": 2.238678841830867e-05, | |
| "loss": 0.6302, | |
| "num_input_tokens_seen": 16777216, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5853658536585366, | |
| "grad_norm": 0.29053160548210144, | |
| "learning_rate": 1.7274575140626318e-05, | |
| "loss": 0.5964, | |
| "num_input_tokens_seen": 18874368, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.6504065040650406, | |
| "grad_norm": 0.2903811037540436, | |
| "learning_rate": 1.2500000000000006e-05, | |
| "loss": 0.6233, | |
| "num_input_tokens_seen": 20971520, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7154471544715447, | |
| "grad_norm": 0.28426122665405273, | |
| "learning_rate": 8.271734841028553e-06, | |
| "loss": 0.5964, | |
| "num_input_tokens_seen": 23068672, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.7804878048780488, | |
| "grad_norm": 0.25174424052238464, | |
| "learning_rate": 4.7745751406263165e-06, | |
| "loss": 0.6106, | |
| "num_input_tokens_seen": 25165824, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8455284552845529, | |
| "grad_norm": 0.24576599895954132, | |
| "learning_rate": 2.1613635589349756e-06, | |
| "loss": 0.5916, | |
| "num_input_tokens_seen": 27262976, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.9105691056910569, | |
| "grad_norm": 0.24116040766239166, | |
| "learning_rate": 5.463099816548579e-07, | |
| "loss": 0.5893, | |
| "num_input_tokens_seen": 29360128, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.975609756097561, | |
| "grad_norm": 0.23365961015224457, | |
| "learning_rate": 0.0, | |
| "loss": 0.6298, | |
| "num_input_tokens_seen": 31457280, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.975609756097561, | |
| "num_input_tokens_seen": 31457280, | |
| "step": 15, | |
| "total_flos": 1.2251230144089293e+18, | |
| "train_loss": 0.6374592224756876, | |
| "train_runtime": 1520.8651, | |
| "train_samples_per_second": 5.175, | |
| "train_steps_per_second": 0.01 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 15, | |
| "num_input_tokens_seen": 31457280, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2251230144089293e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |