| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.15259895088221268, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.007629947544110634, | |
| "grad_norm": 0.1612738072872162, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4215, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007629947544110634, | |
| "eval_loss": 0.3208867311477661, | |
| "eval_runtime": 172.1407, | |
| "eval_samples_per_second": 1.284, | |
| "eval_steps_per_second": 0.645, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.015259895088221268, | |
| "grad_norm": 0.13944248855113983, | |
| "learning_rate": 2e-05, | |
| "loss": 0.4054, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.022889842632331903, | |
| "grad_norm": 0.13380268216133118, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3532, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.030519790176442536, | |
| "grad_norm": 0.14479003846645355, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3248, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.03814973772055317, | |
| "grad_norm": 0.12759573757648468, | |
| "learning_rate": 5e-05, | |
| "loss": 0.316, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03814973772055317, | |
| "eval_loss": 0.31166839599609375, | |
| "eval_runtime": 9.7763, | |
| "eval_samples_per_second": 22.606, | |
| "eval_steps_per_second": 11.354, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.045779685264663805, | |
| "grad_norm": 0.1176108792424202, | |
| "learning_rate": 6e-05, | |
| "loss": 0.2354, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05340963280877444, | |
| "grad_norm": 0.13301990926265717, | |
| "learning_rate": 7e-05, | |
| "loss": 0.3421, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.06103958035288507, | |
| "grad_norm": 0.1429264098405838, | |
| "learning_rate": 8e-05, | |
| "loss": 0.2788, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.06866952789699571, | |
| "grad_norm": 0.16018344461917877, | |
| "learning_rate": 9e-05, | |
| "loss": 0.3339, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.07629947544110634, | |
| "grad_norm": 0.15911918878555298, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3119, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07629947544110634, | |
| "eval_loss": 0.2649441957473755, | |
| "eval_runtime": 11.2056, | |
| "eval_samples_per_second": 19.722, | |
| "eval_steps_per_second": 9.906, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08392942298521698, | |
| "grad_norm": 0.18614229559898376, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.3911, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.09155937052932761, | |
| "grad_norm": 0.14869865775108337, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.307, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.09918931807343824, | |
| "grad_norm": 0.1571897268295288, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.2328, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.10681926561754888, | |
| "grad_norm": 0.14982838928699493, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.202, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.11444921316165951, | |
| "grad_norm": 0.2151786983013153, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2658, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.11444921316165951, | |
| "eval_loss": 0.23548823595046997, | |
| "eval_runtime": 16.2601, | |
| "eval_samples_per_second": 13.592, | |
| "eval_steps_per_second": 6.827, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.12207916070577014, | |
| "grad_norm": 0.17542776465415955, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.2538, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.12970910824988077, | |
| "grad_norm": 0.22913815081119537, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.4404, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.13733905579399142, | |
| "grad_norm": 0.16341139376163483, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.2628, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.14496900333810206, | |
| "grad_norm": 0.17043715715408325, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.3106, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.15259895088221268, | |
| "grad_norm": 0.15494485199451447, | |
| "learning_rate": 0.0, | |
| "loss": 0.2552, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.15259895088221268, | |
| "eval_loss": 0.22873683273792267, | |
| "eval_runtime": 9.7741, | |
| "eval_samples_per_second": 22.611, | |
| "eval_steps_per_second": 11.357, | |
| "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": 2.8529528527650816e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |