| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0023347030257751212, | |
| "eval_steps": 4, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.338812103100486e-05, | |
| "grad_norm": 5.653926849365234, | |
| "learning_rate": 1e-05, | |
| "loss": 3.6941, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 9.338812103100486e-05, | |
| "eval_loss": 3.7275540828704834, | |
| "eval_runtime": 1152.6714, | |
| "eval_samples_per_second": 7.824, | |
| "eval_steps_per_second": 0.979, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00018677624206200972, | |
| "grad_norm": 5.992466926574707, | |
| "learning_rate": 2e-05, | |
| "loss": 3.6692, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0002801643630930146, | |
| "grad_norm": 5.39393424987793, | |
| "learning_rate": 3e-05, | |
| "loss": 3.3814, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00037355248412401944, | |
| "grad_norm": 6.274706840515137, | |
| "learning_rate": 4e-05, | |
| "loss": 3.9748, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00037355248412401944, | |
| "eval_loss": 3.4440622329711914, | |
| "eval_runtime": 1155.325, | |
| "eval_samples_per_second": 7.806, | |
| "eval_steps_per_second": 0.976, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0004669406051550243, | |
| "grad_norm": 5.869632720947266, | |
| "learning_rate": 5e-05, | |
| "loss": 3.3975, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0005603287261860292, | |
| "grad_norm": 6.140427589416504, | |
| "learning_rate": 6e-05, | |
| "loss": 3.1325, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.000653716847217034, | |
| "grad_norm": 5.6471686363220215, | |
| "learning_rate": 7e-05, | |
| "loss": 2.6563, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0007471049682480389, | |
| "grad_norm": 4.763476371765137, | |
| "learning_rate": 8e-05, | |
| "loss": 2.1162, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0007471049682480389, | |
| "eval_loss": 1.8402541875839233, | |
| "eval_runtime": 1154.8766, | |
| "eval_samples_per_second": 7.809, | |
| "eval_steps_per_second": 0.977, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0008404930892790437, | |
| "grad_norm": 5.830313205718994, | |
| "learning_rate": 9e-05, | |
| "loss": 1.8945, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0009338812103100486, | |
| "grad_norm": 4.46910285949707, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5051, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0010272693313410535, | |
| "grad_norm": 4.201104164123535, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 1.0962, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0011206574523720584, | |
| "grad_norm": 3.578787326812744, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.9675, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0011206574523720584, | |
| "eval_loss": 0.8466350436210632, | |
| "eval_runtime": 1155.2879, | |
| "eval_samples_per_second": 7.806, | |
| "eval_steps_per_second": 0.976, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.001214045573403063, | |
| "grad_norm": 4.287249565124512, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.0047, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.001307433694434068, | |
| "grad_norm": 4.440555572509766, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.7256, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0014008218154650729, | |
| "grad_norm": 4.264206886291504, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.5648, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0014942099364960778, | |
| "grad_norm": 3.708203077316284, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.6538, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0014942099364960778, | |
| "eval_loss": 0.49420255422592163, | |
| "eval_runtime": 1155.2843, | |
| "eval_samples_per_second": 7.806, | |
| "eval_steps_per_second": 0.976, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0015875980575270825, | |
| "grad_norm": 3.8472306728363037, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.5875, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0016809861785580874, | |
| "grad_norm": 3.500655174255371, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.5928, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0017743742995890923, | |
| "grad_norm": 3.1135025024414062, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.3957, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0018677624206200972, | |
| "grad_norm": 3.320570945739746, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.3552, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0018677624206200972, | |
| "eval_loss": 0.3732757270336151, | |
| "eval_runtime": 1156.3702, | |
| "eval_samples_per_second": 7.799, | |
| "eval_steps_per_second": 0.975, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.001961150541651102, | |
| "grad_norm": 2.772953510284424, | |
| "learning_rate": 1.6543469682057106e-05, | |
| "loss": 0.3937, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.002054538662682107, | |
| "grad_norm": 2.7570528984069824, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.4561, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.002147926783713112, | |
| "grad_norm": 2.4803898334503174, | |
| "learning_rate": 4.322727117869951e-06, | |
| "loss": 0.4145, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0022413149047441168, | |
| "grad_norm": 3.38692569732666, | |
| "learning_rate": 1.0926199633097157e-06, | |
| "loss": 0.4188, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0022413149047441168, | |
| "eval_loss": 0.34390661120414734, | |
| "eval_runtime": 1154.759, | |
| "eval_samples_per_second": 7.809, | |
| "eval_steps_per_second": 0.977, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0023347030257751212, | |
| "grad_norm": 2.569624423980713, | |
| "learning_rate": 0.0, | |
| "loss": 0.3207, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.91723590385664e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |