| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9093931837073983, | |
| "eval_steps": 500, | |
| "global_step": 7000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.20781379883624274, | |
| "grad_norm": 0.6509162187576294, | |
| "learning_rate": 1.862011637572735e-05, | |
| "loss": 0.5079, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41562759767248547, | |
| "grad_norm": 0.8360931873321533, | |
| "learning_rate": 1.72346910501524e-05, | |
| "loss": 0.4047, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6234413965087282, | |
| "grad_norm": 0.43808215856552124, | |
| "learning_rate": 1.5849265724577447e-05, | |
| "loss": 0.3844, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8312551953449709, | |
| "grad_norm": 0.4758487045764923, | |
| "learning_rate": 1.4463840399002496e-05, | |
| "loss": 0.3804, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.30536511540412903, | |
| "eval_runtime": 26.8313, | |
| "eval_samples_per_second": 18.411, | |
| "eval_steps_per_second": 4.621, | |
| "step": 2406 | |
| }, | |
| { | |
| "epoch": 1.0390689941812137, | |
| "grad_norm": 0.5247063636779785, | |
| "learning_rate": 1.3078415073427544e-05, | |
| "loss": 0.3753, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.2468827930174564, | |
| "grad_norm": 0.36628538370132446, | |
| "learning_rate": 1.1692989747852592e-05, | |
| "loss": 0.347, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4546965918536992, | |
| "grad_norm": 0.3838510513305664, | |
| "learning_rate": 1.030756442227764e-05, | |
| "loss": 0.3417, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.6625103906899419, | |
| "grad_norm": 0.7672457098960876, | |
| "learning_rate": 8.92213909670269e-06, | |
| "loss": 0.3444, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.8703241895261846, | |
| "grad_norm": 0.5239430069923401, | |
| "learning_rate": 7.536713771127737e-06, | |
| "loss": 0.3383, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.28803136944770813, | |
| "eval_runtime": 26.8252, | |
| "eval_samples_per_second": 18.416, | |
| "eval_steps_per_second": 4.623, | |
| "step": 4812 | |
| }, | |
| { | |
| "epoch": 2.0781379883624274, | |
| "grad_norm": 0.5267770886421204, | |
| "learning_rate": 6.154059296203935e-06, | |
| "loss": 0.3237, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.28595178719867, | |
| "grad_norm": 0.8022969961166382, | |
| "learning_rate": 4.7686339706289835e-06, | |
| "loss": 0.3291, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.493765586034913, | |
| "grad_norm": 0.7274072766304016, | |
| "learning_rate": 3.3832086450540318e-06, | |
| "loss": 0.3238, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.7015793848711556, | |
| "grad_norm": 0.7056533694267273, | |
| "learning_rate": 1.99778331947908e-06, | |
| "loss": 0.3229, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.9093931837073983, | |
| "grad_norm": 0.5834936499595642, | |
| "learning_rate": 6.123579939041286e-07, | |
| "loss": 0.3166, | |
| "step": 7000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7218, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.704960301989888e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |