{ "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "eval_steps": 500, "global_step": 282, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.10638297872340426, "grad_norm": 3.6553943157196045, "learning_rate": 4.840425531914894e-05, "loss": 0.9456, "step": 10 }, { "epoch": 0.2127659574468085, "grad_norm": 3.9983913898468018, "learning_rate": 4.663120567375887e-05, "loss": 0.8552, "step": 20 }, { "epoch": 0.3191489361702128, "grad_norm": 3.7922072410583496, "learning_rate": 4.48581560283688e-05, "loss": 1.0052, "step": 30 }, { "epoch": 0.425531914893617, "grad_norm": 3.6140127182006836, "learning_rate": 4.3262411347517734e-05, "loss": 0.9121, "step": 40 }, { "epoch": 0.5319148936170213, "grad_norm": 3.507614850997925, "learning_rate": 4.148936170212766e-05, "loss": 0.8598, "step": 50 }, { "epoch": 0.6382978723404256, "grad_norm": 4.006348133087158, "learning_rate": 3.971631205673759e-05, "loss": 0.9708, "step": 60 }, { "epoch": 0.7446808510638298, "grad_norm": 4.814829349517822, "learning_rate": 3.794326241134752e-05, "loss": 0.8753, "step": 70 }, { "epoch": 0.851063829787234, "grad_norm": 4.398285388946533, "learning_rate": 3.617021276595745e-05, "loss": 0.9466, "step": 80 }, { "epoch": 0.9574468085106383, "grad_norm": 4.186042785644531, "learning_rate": 3.4397163120567377e-05, "loss": 0.9408, "step": 90 }, { "epoch": 1.0, "eval_loss": 2.226564645767212, "eval_runtime": 9.5187, "eval_samples_per_second": 4.412, "eval_steps_per_second": 1.156, "step": 94 }, { "epoch": 1.0638297872340425, "grad_norm": 3.774606227874756, "learning_rate": 3.262411347517731e-05, "loss": 0.8182, "step": 100 }, { "epoch": 1.1702127659574468, "grad_norm": 3.641083240509033, "learning_rate": 3.085106382978723e-05, "loss": 0.7849, "step": 110 }, { "epoch": 1.2765957446808511, "grad_norm": 5.060397624969482, "learning_rate": 2.9078014184397162e-05, "loss": 0.7795, "step": 120 }, { "epoch": 1.3829787234042552, "grad_norm": 3.457118272781372, "learning_rate": 2.7304964539007094e-05, "loss": 0.723, "step": 130 }, { "epoch": 1.4893617021276595, "grad_norm": 3.421212673187256, "learning_rate": 2.5531914893617022e-05, "loss": 0.6819, "step": 140 }, { "epoch": 1.5957446808510638, "grad_norm": 3.1989083290100098, "learning_rate": 2.3758865248226954e-05, "loss": 0.6891, "step": 150 }, { "epoch": 1.702127659574468, "grad_norm": 3.486337661743164, "learning_rate": 2.1985815602836882e-05, "loss": 0.7369, "step": 160 }, { "epoch": 1.8085106382978724, "grad_norm": 3.4989173412323, "learning_rate": 2.0212765957446807e-05, "loss": 0.7042, "step": 170 }, { "epoch": 1.9148936170212765, "grad_norm": 2.965428590774536, "learning_rate": 1.8439716312056736e-05, "loss": 0.712, "step": 180 }, { "epoch": 2.0, "eval_loss": 2.416543483734131, "eval_runtime": 9.5843, "eval_samples_per_second": 4.382, "eval_steps_per_second": 1.148, "step": 188 }, { "epoch": 2.021276595744681, "grad_norm": 4.359320640563965, "learning_rate": 1.6666666666666667e-05, "loss": 0.7045, "step": 190 }, { "epoch": 2.127659574468085, "grad_norm": 3.513986825942993, "learning_rate": 1.4893617021276596e-05, "loss": 0.6259, "step": 200 }, { "epoch": 2.2340425531914896, "grad_norm": 4.127784252166748, "learning_rate": 1.3120567375886524e-05, "loss": 0.6355, "step": 210 }, { "epoch": 2.3404255319148937, "grad_norm": 3.798154592514038, "learning_rate": 1.1347517730496454e-05, "loss": 0.5645, "step": 220 }, { "epoch": 2.4468085106382977, "grad_norm": 3.0239367485046387, "learning_rate": 9.574468085106383e-06, "loss": 0.6694, "step": 230 }, { "epoch": 2.5531914893617023, "grad_norm": 3.148362398147583, "learning_rate": 7.801418439716313e-06, "loss": 0.6103, "step": 240 }, { "epoch": 2.6595744680851063, "grad_norm": 4.111635684967041, "learning_rate": 6.028368794326241e-06, "loss": 0.5299, "step": 250 }, { "epoch": 2.7659574468085104, "grad_norm": 3.042776107788086, "learning_rate": 4.255319148936171e-06, "loss": 0.5928, "step": 260 }, { "epoch": 2.872340425531915, "grad_norm": 3.5736513137817383, "learning_rate": 2.4822695035460995e-06, "loss": 0.6444, "step": 270 }, { "epoch": 2.978723404255319, "grad_norm": 3.4036762714385986, "learning_rate": 7.092198581560284e-07, "loss": 0.5531, "step": 280 }, { "epoch": 3.0, "eval_loss": 2.504506826400757, "eval_runtime": 14.1693, "eval_samples_per_second": 2.964, "eval_steps_per_second": 0.776, "step": 282 } ], "logging_steps": 10, "max_steps": 282, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1047574383427584.0, "train_batch_size": 4, "trial_name": null, "trial_params": null }