| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06874328678839957, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0034371643394199786, | |
| "grad_norm": 1.3658827543258667, | |
| "learning_rate": 1e-05, | |
| "loss": 2.9807, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0034371643394199786, | |
| "eval_loss": 2.908154249191284, | |
| "eval_runtime": 76.6233, | |
| "eval_samples_per_second": 6.395, | |
| "eval_steps_per_second": 3.197, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006874328678839957, | |
| "grad_norm": 0.9233030080795288, | |
| "learning_rate": 2e-05, | |
| "loss": 2.8263, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.010311493018259935, | |
| "grad_norm": 1.083367109298706, | |
| "learning_rate": 3e-05, | |
| "loss": 2.6734, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.013748657357679914, | |
| "grad_norm": 1.1248096227645874, | |
| "learning_rate": 4e-05, | |
| "loss": 2.957, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.017185821697099892, | |
| "grad_norm": 1.562717318534851, | |
| "learning_rate": 5e-05, | |
| "loss": 2.8726, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.017185821697099892, | |
| "eval_loss": 2.8320984840393066, | |
| "eval_runtime": 24.711, | |
| "eval_samples_per_second": 19.829, | |
| "eval_steps_per_second": 9.915, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02062298603651987, | |
| "grad_norm": 1.1580060720443726, | |
| "learning_rate": 6e-05, | |
| "loss": 2.7856, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.02406015037593985, | |
| "grad_norm": 1.037327527999878, | |
| "learning_rate": 7e-05, | |
| "loss": 2.7426, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02749731471535983, | |
| "grad_norm": 1.1322250366210938, | |
| "learning_rate": 8e-05, | |
| "loss": 2.762, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.030934479054779807, | |
| "grad_norm": 0.8920819759368896, | |
| "learning_rate": 9e-05, | |
| "loss": 2.6572, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.034371643394199784, | |
| "grad_norm": 0.7211071252822876, | |
| "learning_rate": 0.0001, | |
| "loss": 2.5354, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.034371643394199784, | |
| "eval_loss": 2.593578577041626, | |
| "eval_runtime": 26.2073, | |
| "eval_samples_per_second": 18.697, | |
| "eval_steps_per_second": 9.349, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03780880773361976, | |
| "grad_norm": 0.857118546962738, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.6361, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.04124597207303974, | |
| "grad_norm": 1.2428054809570312, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.567, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04468313641245972, | |
| "grad_norm": 1.1504372358322144, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.3349, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0481203007518797, | |
| "grad_norm": 1.156445860862732, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.4002, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.05155746509129968, | |
| "grad_norm": 0.9205731153488159, | |
| "learning_rate": 5e-05, | |
| "loss": 2.3525, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.05155746509129968, | |
| "eval_loss": 2.407013416290283, | |
| "eval_runtime": 37.4382, | |
| "eval_samples_per_second": 13.088, | |
| "eval_steps_per_second": 6.544, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.05499462943071966, | |
| "grad_norm": 0.7372962832450867, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.4173, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.058431793770139635, | |
| "grad_norm": 0.7817841172218323, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.3822, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.06186895810955961, | |
| "grad_norm": 0.7767869234085083, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 2.4632, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0653061224489796, | |
| "grad_norm": 0.7745295166969299, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 2.4015, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.06874328678839957, | |
| "grad_norm": 0.8225830793380737, | |
| "learning_rate": 0.0, | |
| "loss": 2.3646, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06874328678839957, | |
| "eval_loss": 2.3743178844451904, | |
| "eval_runtime": 37.5929, | |
| "eval_samples_per_second": 13.034, | |
| "eval_steps_per_second": 6.517, | |
| "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": 1.479651398516736e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |