| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.059273894788836747, | |
| "eval_steps": 8, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001975796492961225, | |
| "grad_norm": 7.703702926635742, | |
| "learning_rate": 1e-05, | |
| "loss": 20.6007, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001975796492961225, | |
| "eval_loss": NaN, | |
| "eval_runtime": 161.515, | |
| "eval_samples_per_second": 5.281, | |
| "eval_steps_per_second": 2.644, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00395159298592245, | |
| "grad_norm": 10.786239624023438, | |
| "learning_rate": 2e-05, | |
| "loss": 19.4341, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.005927389478883675, | |
| "grad_norm": 10.11152172088623, | |
| "learning_rate": 3e-05, | |
| "loss": 20.713, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0079031859718449, | |
| "grad_norm": 7.585945129394531, | |
| "learning_rate": 4e-05, | |
| "loss": 18.3887, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.009878982464806126, | |
| "grad_norm": 10.86007308959961, | |
| "learning_rate": 5e-05, | |
| "loss": 19.6414, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01185477895776735, | |
| "grad_norm": 10.720023155212402, | |
| "learning_rate": 6e-05, | |
| "loss": 22.0141, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.013830575450728575, | |
| "grad_norm": 7.739660263061523, | |
| "learning_rate": 7e-05, | |
| "loss": 22.056, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0158063719436898, | |
| "grad_norm": 5.171518325805664, | |
| "learning_rate": 8e-05, | |
| "loss": 18.8531, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0158063719436898, | |
| "eval_loss": NaN, | |
| "eval_runtime": 146.1491, | |
| "eval_samples_per_second": 5.837, | |
| "eval_steps_per_second": 2.922, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.017782168436651025, | |
| "grad_norm": 5.789409637451172, | |
| "learning_rate": 9e-05, | |
| "loss": 18.2752, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01975796492961225, | |
| "grad_norm": 7.014954090118408, | |
| "learning_rate": 0.0001, | |
| "loss": 17.7699, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.021733761422573473, | |
| "grad_norm": 8.307316780090332, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 18.9728, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0237095579155347, | |
| "grad_norm": 5.6757659912109375, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 16.5401, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.025685354408495925, | |
| "grad_norm": 4.361359596252441, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 18.2953, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02766115090145715, | |
| "grad_norm": 4.344420433044434, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 18.8302, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.029636947394418373, | |
| "grad_norm": 3.632826566696167, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 15.5949, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0316127438873796, | |
| "grad_norm": 5.199756622314453, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 16.3128, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0316127438873796, | |
| "eval_loss": NaN, | |
| "eval_runtime": 145.9235, | |
| "eval_samples_per_second": 5.846, | |
| "eval_steps_per_second": 2.926, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.03358854038034083, | |
| "grad_norm": 3.910904884338379, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 15.6483, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.03556433687330205, | |
| "grad_norm": 3.871004104614258, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 17.0178, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03754013336626327, | |
| "grad_norm": 4.427883625030518, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 18.9645, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0395159298592245, | |
| "grad_norm": 3.5598199367523193, | |
| "learning_rate": 5e-05, | |
| "loss": 17.7472, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.041491726352185725, | |
| "grad_norm": 3.578526020050049, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 17.6056, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.04346752284514695, | |
| "grad_norm": 2.900790214538574, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 17.3251, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.045443319338108176, | |
| "grad_norm": 3.5540361404418945, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 16.6689, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0474191158310694, | |
| "grad_norm": 2.947385311126709, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 16.161, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0474191158310694, | |
| "eval_loss": NaN, | |
| "eval_runtime": 145.4977, | |
| "eval_samples_per_second": 5.863, | |
| "eval_steps_per_second": 2.935, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04939491232403063, | |
| "grad_norm": 4.943809509277344, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 17.7916, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05137070881699185, | |
| "grad_norm": 5.202208518981934, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 16.5579, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.05334650530995307, | |
| "grad_norm": 2.9827098846435547, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 16.4835, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0553223018029143, | |
| "grad_norm": 5.730014324188232, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 16.5029, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.057298098295875524, | |
| "grad_norm": 3.476895570755005, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 16.4587, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.059273894788836747, | |
| "grad_norm": 3.6241424083709717, | |
| "learning_rate": 0.0, | |
| "loss": 17.1663, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.842039240163328e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |