| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.08683853459972862, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004341926729986432, | |
| "grad_norm": 3.3317363262176514, | |
| "learning_rate": 1e-05, | |
| "loss": 3.8736, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004341926729986432, | |
| "eval_loss": 4.2282562255859375, | |
| "eval_runtime": 69.4265, | |
| "eval_samples_per_second": 5.589, | |
| "eval_steps_per_second": 2.794, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.008683853459972864, | |
| "grad_norm": 3.3094608783721924, | |
| "learning_rate": 2e-05, | |
| "loss": 3.9504, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.013025780189959294, | |
| "grad_norm": 3.510456085205078, | |
| "learning_rate": 3e-05, | |
| "loss": 4.1357, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.017367706919945727, | |
| "grad_norm": 3.0727248191833496, | |
| "learning_rate": 4e-05, | |
| "loss": 3.6872, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.021709633649932156, | |
| "grad_norm": 3.095470428466797, | |
| "learning_rate": 5e-05, | |
| "loss": 3.7915, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.021709633649932156, | |
| "eval_loss": 4.091055870056152, | |
| "eval_runtime": 3.6207, | |
| "eval_samples_per_second": 107.163, | |
| "eval_steps_per_second": 53.581, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.026051560379918588, | |
| "grad_norm": 2.981228828430176, | |
| "learning_rate": 6e-05, | |
| "loss": 3.6453, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.03039348710990502, | |
| "grad_norm": 3.3638360500335693, | |
| "learning_rate": 7e-05, | |
| "loss": 3.9298, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.034735413839891455, | |
| "grad_norm": 3.848174810409546, | |
| "learning_rate": 8e-05, | |
| "loss": 4.1513, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03907734056987788, | |
| "grad_norm": 2.8415369987487793, | |
| "learning_rate": 9e-05, | |
| "loss": 4.1745, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04341926729986431, | |
| "grad_norm": 2.2789957523345947, | |
| "learning_rate": 0.0001, | |
| "loss": 3.2271, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04341926729986431, | |
| "eval_loss": 3.6719045639038086, | |
| "eval_runtime": 3.6766, | |
| "eval_samples_per_second": 105.531, | |
| "eval_steps_per_second": 52.766, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04776119402985075, | |
| "grad_norm": 2.937727212905884, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 3.8382, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.052103120759837175, | |
| "grad_norm": 2.4996044635772705, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 3.0171, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05644504748982361, | |
| "grad_norm": 2.593256711959839, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 3.3513, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.06078697421981004, | |
| "grad_norm": 2.700037956237793, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 3.4148, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.06512890094979647, | |
| "grad_norm": 2.4664251804351807, | |
| "learning_rate": 5e-05, | |
| "loss": 3.2387, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06512890094979647, | |
| "eval_loss": 3.2880284786224365, | |
| "eval_runtime": 3.6181, | |
| "eval_samples_per_second": 107.239, | |
| "eval_steps_per_second": 53.619, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06947082767978291, | |
| "grad_norm": 2.7127010822296143, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 3.009, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.07381275440976934, | |
| "grad_norm": 2.4328489303588867, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.683, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.07815468113975577, | |
| "grad_norm": 3.2602317333221436, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 3.215, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0824966078697422, | |
| "grad_norm": 3.8587803840637207, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 3.2797, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.08683853459972862, | |
| "grad_norm": 3.2040271759033203, | |
| "learning_rate": 0.0, | |
| "loss": 3.0642, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08683853459972862, | |
| "eval_loss": 3.188815116882324, | |
| "eval_runtime": 3.6622, | |
| "eval_samples_per_second": 105.948, | |
| "eval_steps_per_second": 52.974, | |
| "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": 977193847750656.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |