| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.28293545534924847, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014146772767462422, | |
| "grad_norm": 4.968000411987305, | |
| "learning_rate": 1e-05, | |
| "loss": 8.3533, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.014146772767462422, | |
| "eval_loss": 7.851544380187988, | |
| "eval_runtime": 85.5599, | |
| "eval_samples_per_second": 1.403, | |
| "eval_steps_per_second": 0.701, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.028293545534924844, | |
| "grad_norm": 4.63974142074585, | |
| "learning_rate": 2e-05, | |
| "loss": 7.7908, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.042440318302387266, | |
| "grad_norm": 5.394748687744141, | |
| "learning_rate": 3e-05, | |
| "loss": 7.5542, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05658709106984969, | |
| "grad_norm": 4.698472023010254, | |
| "learning_rate": 4e-05, | |
| "loss": 7.7776, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.07073386383731212, | |
| "grad_norm": 5.1544108390808105, | |
| "learning_rate": 5e-05, | |
| "loss": 7.3847, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07073386383731212, | |
| "eval_loss": 7.290081977844238, | |
| "eval_runtime": 5.8165, | |
| "eval_samples_per_second": 20.631, | |
| "eval_steps_per_second": 10.316, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.08488063660477453, | |
| "grad_norm": 5.7256669998168945, | |
| "learning_rate": 6e-05, | |
| "loss": 7.4766, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09902740937223696, | |
| "grad_norm": 5.811617374420166, | |
| "learning_rate": 7e-05, | |
| "loss": 7.0652, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.11317418213969938, | |
| "grad_norm": 6.549645900726318, | |
| "learning_rate": 8e-05, | |
| "loss": 6.8741, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1273209549071618, | |
| "grad_norm": 7.107234477996826, | |
| "learning_rate": 9e-05, | |
| "loss": 6.21, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.14146772767462423, | |
| "grad_norm": 6.856642246246338, | |
| "learning_rate": 0.0001, | |
| "loss": 5.349, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14146772767462423, | |
| "eval_loss": 4.707101345062256, | |
| "eval_runtime": 5.8383, | |
| "eval_samples_per_second": 20.554, | |
| "eval_steps_per_second": 10.277, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15561450044208666, | |
| "grad_norm": 7.615804195404053, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 4.8278, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.16976127320954906, | |
| "grad_norm": 7.701565742492676, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 3.9847, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1839080459770115, | |
| "grad_norm": 6.278581619262695, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 3.133, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.19805481874447392, | |
| "grad_norm": 6.357863903045654, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.2536, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.21220159151193635, | |
| "grad_norm": 4.74558687210083, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9522, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.21220159151193635, | |
| "eval_loss": 1.695694088935852, | |
| "eval_runtime": 5.7854, | |
| "eval_samples_per_second": 20.742, | |
| "eval_steps_per_second": 10.371, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.22634836427939875, | |
| "grad_norm": 4.90279483795166, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.6686, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.24049513704686118, | |
| "grad_norm": 4.2148027420043945, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.4045, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.2546419098143236, | |
| "grad_norm": 4.100856304168701, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.3224, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.268788682581786, | |
| "grad_norm": 2.986417531967163, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.1764, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.28293545534924847, | |
| "grad_norm": 4.208008766174316, | |
| "learning_rate": 0.0, | |
| "loss": 1.7821, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.28293545534924847, | |
| "eval_loss": 1.1980129480361938, | |
| "eval_runtime": 5.8173, | |
| "eval_samples_per_second": 20.628, | |
| "eval_steps_per_second": 10.314, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3388561252352e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |