| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.07010369504892654, | |
| "eval_steps": 8, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0023367898349642177, | |
| "grad_norm": 0.3870636224746704, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0099, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0023367898349642177, | |
| "eval_loss": 2.423468828201294, | |
| "eval_runtime": 133.3362, | |
| "eval_samples_per_second": 5.407, | |
| "eval_steps_per_second": 2.707, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004673579669928435, | |
| "grad_norm": 0.48035353422164917, | |
| "learning_rate": 2e-05, | |
| "loss": 2.3747, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0070103695048926535, | |
| "grad_norm": 0.533902108669281, | |
| "learning_rate": 3e-05, | |
| "loss": 2.4928, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00934715933985687, | |
| "grad_norm": 0.50484299659729, | |
| "learning_rate": 4e-05, | |
| "loss": 1.9791, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.011683949174821089, | |
| "grad_norm": 0.518813967704773, | |
| "learning_rate": 5e-05, | |
| "loss": 2.3034, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.014020739009785307, | |
| "grad_norm": 0.4816703796386719, | |
| "learning_rate": 6e-05, | |
| "loss": 1.8373, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.016357528844749527, | |
| "grad_norm": 0.6508989334106445, | |
| "learning_rate": 7e-05, | |
| "loss": 2.2096, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.01869431867971374, | |
| "grad_norm": 0.644939661026001, | |
| "learning_rate": 8e-05, | |
| "loss": 2.2933, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01869431867971374, | |
| "eval_loss": 2.344738006591797, | |
| "eval_runtime": 44.173, | |
| "eval_samples_per_second": 16.322, | |
| "eval_steps_per_second": 8.172, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.02103110851467796, | |
| "grad_norm": 0.6885388493537903, | |
| "learning_rate": 9e-05, | |
| "loss": 2.4027, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.023367898349642178, | |
| "grad_norm": 0.5526238083839417, | |
| "learning_rate": 0.0001, | |
| "loss": 1.8518, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.025704688184606396, | |
| "grad_norm": 1.0618926286697388, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 2.0879, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.028041478019570614, | |
| "grad_norm": 0.857267439365387, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.8582, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.030378267854534832, | |
| "grad_norm": 0.7300710082054138, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 1.7029, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.032715057689499054, | |
| "grad_norm": 0.8072000741958618, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.1175, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03505184752446327, | |
| "grad_norm": 0.7812466621398926, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.4707, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03738863735942748, | |
| "grad_norm": 1.0912774801254272, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.6905, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.03738863735942748, | |
| "eval_loss": 2.0501091480255127, | |
| "eval_runtime": 44.2225, | |
| "eval_samples_per_second": 16.304, | |
| "eval_steps_per_second": 8.163, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.039725427194391705, | |
| "grad_norm": 0.856139063835144, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 1.9171, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.04206221702935592, | |
| "grad_norm": 0.8442476987838745, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.1821, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04439900686432014, | |
| "grad_norm": 1.2221778631210327, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 2.0464, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.046735796699284356, | |
| "grad_norm": 0.8681303858757019, | |
| "learning_rate": 5e-05, | |
| "loss": 2.0358, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04907258653424858, | |
| "grad_norm": 1.1155294179916382, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 1.9598, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05140937636921279, | |
| "grad_norm": 1.2673661708831787, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.6758, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.053746166204177014, | |
| "grad_norm": 1.268276572227478, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 2.1396, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.05608295603914123, | |
| "grad_norm": 0.8677752614021301, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.0531, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.05608295603914123, | |
| "eval_loss": 1.8527228832244873, | |
| "eval_runtime": 44.1953, | |
| "eval_samples_per_second": 16.314, | |
| "eval_steps_per_second": 8.168, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.05841974587410545, | |
| "grad_norm": 0.7778655886650085, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.9664, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.060756535709069664, | |
| "grad_norm": 1.1115986108779907, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.8281, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06309332554403388, | |
| "grad_norm": 0.8952155709266663, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 2.0629, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06543011537899811, | |
| "grad_norm": 0.6139501929283142, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.9817, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.06776690521396232, | |
| "grad_norm": 0.6845227479934692, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 1.9068, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.07010369504892654, | |
| "grad_norm": 0.7899869680404663, | |
| "learning_rate": 0.0, | |
| "loss": 1.746, | |
| "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": 3.135045208375296e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |