| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04792571514153063, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0023962857570765314, | |
| "grad_norm": 0.20548972487449646, | |
| "learning_rate": 1e-05, | |
| "loss": 1.0815, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0023962857570765314, | |
| "eval_loss": 1.0520811080932617, | |
| "eval_runtime": 298.192, | |
| "eval_samples_per_second": 2.358, | |
| "eval_steps_per_second": 1.18, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004792571514153063, | |
| "grad_norm": 0.20183496177196503, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0013, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.007188857271229594, | |
| "grad_norm": 0.17294946312904358, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0227, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.009585143028306126, | |
| "grad_norm": 0.1912940889596939, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0544, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.011981428785382657, | |
| "grad_norm": 0.18699872493743896, | |
| "learning_rate": 5e-05, | |
| "loss": 1.058, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.011981428785382657, | |
| "eval_loss": 1.0332369804382324, | |
| "eval_runtime": 91.6381, | |
| "eval_samples_per_second": 7.671, | |
| "eval_steps_per_second": 3.841, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.014377714542459188, | |
| "grad_norm": 0.18617145717144012, | |
| "learning_rate": 6e-05, | |
| "loss": 1.0427, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.01677400029953572, | |
| "grad_norm": 0.2107728272676468, | |
| "learning_rate": 7e-05, | |
| "loss": 1.0103, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.01917028605661225, | |
| "grad_norm": 0.2871200144290924, | |
| "learning_rate": 8e-05, | |
| "loss": 1.0056, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.021566571813688783, | |
| "grad_norm": 0.35235795378685, | |
| "learning_rate": 9e-05, | |
| "loss": 1.0133, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.023962857570765314, | |
| "grad_norm": 0.3103502094745636, | |
| "learning_rate": 0.0001, | |
| "loss": 0.9999, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.023962857570765314, | |
| "eval_loss": 0.9529623985290527, | |
| "eval_runtime": 91.6471, | |
| "eval_samples_per_second": 7.671, | |
| "eval_steps_per_second": 3.841, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.026359143327841845, | |
| "grad_norm": 0.25037920475006104, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.9432, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.028755429084918377, | |
| "grad_norm": 0.2890665829181671, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.9455, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03115171484199491, | |
| "grad_norm": 0.25389617681503296, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.8939, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03354800059907144, | |
| "grad_norm": 0.2094624787569046, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.8387, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03594428635614797, | |
| "grad_norm": 0.21110887825489044, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8639, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03594428635614797, | |
| "eval_loss": 0.8595573306083679, | |
| "eval_runtime": 91.6996, | |
| "eval_samples_per_second": 7.666, | |
| "eval_steps_per_second": 3.839, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0383405721132245, | |
| "grad_norm": 0.20239923894405365, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.8725, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.040736857870301034, | |
| "grad_norm": 0.19121764600276947, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.8105, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.043133143627377565, | |
| "grad_norm": 0.19437186419963837, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.8758, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0455294293844541, | |
| "grad_norm": 0.20055896043777466, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.8736, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.04792571514153063, | |
| "grad_norm": 0.21000787615776062, | |
| "learning_rate": 0.0, | |
| "loss": 0.8396, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04792571514153063, | |
| "eval_loss": 0.8422181010246277, | |
| "eval_runtime": 91.6047, | |
| "eval_samples_per_second": 7.674, | |
| "eval_steps_per_second": 3.843, | |
| "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": 2.628277562769408e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |