| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05966809621480515, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0029834048107402573, | |
| "grad_norm": 2.264500856399536, | |
| "learning_rate": 1e-05, | |
| "loss": 2.9527, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0029834048107402573, | |
| "eval_loss": 2.8173000812530518, | |
| "eval_runtime": 437.0205, | |
| "eval_samples_per_second": 1.293, | |
| "eval_steps_per_second": 0.648, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005966809621480515, | |
| "grad_norm": 2.272521734237671, | |
| "learning_rate": 2e-05, | |
| "loss": 3.0445, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.008950214432220773, | |
| "grad_norm": 1.8918967247009277, | |
| "learning_rate": 3e-05, | |
| "loss": 2.5151, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01193361924296103, | |
| "grad_norm": 2.1396842002868652, | |
| "learning_rate": 4e-05, | |
| "loss": 2.8278, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.014917024053701287, | |
| "grad_norm": 2.394638776779175, | |
| "learning_rate": 5e-05, | |
| "loss": 2.7964, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.014917024053701287, | |
| "eval_loss": 2.63140869140625, | |
| "eval_runtime": 37.4393, | |
| "eval_samples_per_second": 15.091, | |
| "eval_steps_per_second": 7.559, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.017900428864441546, | |
| "grad_norm": 2.2498764991760254, | |
| "learning_rate": 6e-05, | |
| "loss": 2.854, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0208838336751818, | |
| "grad_norm": 1.8807456493377686, | |
| "learning_rate": 7e-05, | |
| "loss": 2.2667, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02386723848592206, | |
| "grad_norm": 2.422107696533203, | |
| "learning_rate": 8e-05, | |
| "loss": 2.3471, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.026850643296662317, | |
| "grad_norm": 2.0063250064849854, | |
| "learning_rate": 9e-05, | |
| "loss": 2.131, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.029834048107402575, | |
| "grad_norm": 2.074761152267456, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1234, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.029834048107402575, | |
| "eval_loss": 1.8020358085632324, | |
| "eval_runtime": 39.6627, | |
| "eval_samples_per_second": 14.245, | |
| "eval_steps_per_second": 7.135, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03281745291814283, | |
| "grad_norm": 1.388445258140564, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.6868, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03580085772888309, | |
| "grad_norm": 1.4602677822113037, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.6535, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.038784262539623346, | |
| "grad_norm": 1.17501699924469, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.4424, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0417676673503636, | |
| "grad_norm": 1.192784309387207, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.4175, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.04475107216110386, | |
| "grad_norm": 1.4021286964416504, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5799, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04475107216110386, | |
| "eval_loss": 1.4331947565078735, | |
| "eval_runtime": 37.2111, | |
| "eval_samples_per_second": 15.184, | |
| "eval_steps_per_second": 7.605, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04773447697184412, | |
| "grad_norm": 1.331388235092163, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.5606, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05071788178258437, | |
| "grad_norm": 1.2103185653686523, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.3576, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.05370128659332463, | |
| "grad_norm": 1.212957501411438, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.4321, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05668469140406489, | |
| "grad_norm": 0.9084908366203308, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.9734, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05966809621480515, | |
| "grad_norm": 1.1958684921264648, | |
| "learning_rate": 0.0, | |
| "loss": 1.386, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05966809621480515, | |
| "eval_loss": 1.3843706846237183, | |
| "eval_runtime": 24.7218, | |
| "eval_samples_per_second": 22.854, | |
| "eval_steps_per_second": 11.447, | |
| "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": 1.6470458779041792e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |