| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06146281499692686, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001229256299938537, | |
| "eval_loss": 1.8433973789215088, | |
| "eval_runtime": 107.6131, | |
| "eval_samples_per_second": 12.74, | |
| "eval_steps_per_second": 1.598, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0036877688998156115, | |
| "grad_norm": 8.21639347076416, | |
| "learning_rate": 3e-05, | |
| "loss": 1.906, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.006146281499692686, | |
| "eval_loss": 0.9512829184532166, | |
| "eval_runtime": 107.5751, | |
| "eval_samples_per_second": 12.745, | |
| "eval_steps_per_second": 1.599, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.007375537799631223, | |
| "grad_norm": 3.9271481037139893, | |
| "learning_rate": 6e-05, | |
| "loss": 1.1987, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.011063306699446834, | |
| "grad_norm": 2.8276703357696533, | |
| "learning_rate": 9e-05, | |
| "loss": 0.753, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.012292562999385371, | |
| "eval_loss": 0.6514846086502075, | |
| "eval_runtime": 107.8745, | |
| "eval_samples_per_second": 12.709, | |
| "eval_steps_per_second": 1.594, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.014751075599262446, | |
| "grad_norm": 2.486225128173828, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.6877, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01843884449907806, | |
| "grad_norm": 1.9635486602783203, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.6047, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01843884449907806, | |
| "eval_loss": 0.5996317267417908, | |
| "eval_runtime": 107.9179, | |
| "eval_samples_per_second": 12.704, | |
| "eval_steps_per_second": 1.594, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.022126613398893668, | |
| "grad_norm": 2.0274317264556885, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.5949, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.024585125998770743, | |
| "eval_loss": 0.5480629205703735, | |
| "eval_runtime": 108.0826, | |
| "eval_samples_per_second": 12.685, | |
| "eval_steps_per_second": 1.591, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02581438229870928, | |
| "grad_norm": 1.7041795253753662, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.5983, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.02950215119852489, | |
| "grad_norm": 1.4558817148208618, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.5055, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03073140749846343, | |
| "eval_loss": 0.5338066816329956, | |
| "eval_runtime": 107.8606, | |
| "eval_samples_per_second": 12.711, | |
| "eval_steps_per_second": 1.595, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03318992009834051, | |
| "grad_norm": 1.7881927490234375, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.5399, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03687768899815612, | |
| "grad_norm": 1.5611547231674194, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5698, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03687768899815612, | |
| "eval_loss": 0.5137801170349121, | |
| "eval_runtime": 107.8291, | |
| "eval_samples_per_second": 12.715, | |
| "eval_steps_per_second": 1.595, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.040565457897971724, | |
| "grad_norm": 1.404949426651001, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.4988, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0430239704978488, | |
| "eval_loss": 0.4986587464809418, | |
| "eval_runtime": 107.8743, | |
| "eval_samples_per_second": 12.709, | |
| "eval_steps_per_second": 1.594, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.044253226797787336, | |
| "grad_norm": 1.4394307136535645, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.539, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04794099569760295, | |
| "grad_norm": 1.472872018814087, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.4637, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.049170251997541485, | |
| "eval_loss": 0.48103824257850647, | |
| "eval_runtime": 107.9289, | |
| "eval_samples_per_second": 12.703, | |
| "eval_steps_per_second": 1.594, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05162876459741856, | |
| "grad_norm": 1.4950206279754639, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.4697, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.05531653349723417, | |
| "grad_norm": 1.9292529821395874, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.553, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05531653349723417, | |
| "eval_loss": 0.473427951335907, | |
| "eval_runtime": 107.8864, | |
| "eval_samples_per_second": 12.708, | |
| "eval_steps_per_second": 1.594, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05900430239704978, | |
| "grad_norm": 1.1889021396636963, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.443, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06146281499692686, | |
| "eval_loss": 0.4718257188796997, | |
| "eval_runtime": 107.867, | |
| "eval_samples_per_second": 12.71, | |
| "eval_steps_per_second": 1.595, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 50, | |
| "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": 7.881983448947098e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |