| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.6052631578947367, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02631578947368421, | |
| "eval_loss": 1.6184024810791016, | |
| "eval_runtime": 4.4294, | |
| "eval_samples_per_second": 14.449, | |
| "eval_steps_per_second": 1.806, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07894736842105263, | |
| "grad_norm": 26.1455020904541, | |
| "learning_rate": 3e-05, | |
| "loss": 6.2533, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.15789473684210525, | |
| "grad_norm": 20.71628761291504, | |
| "learning_rate": 6e-05, | |
| "loss": 6.0813, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.23684210526315788, | |
| "grad_norm": 17.294282913208008, | |
| "learning_rate": 9e-05, | |
| "loss": 4.9981, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.23684210526315788, | |
| "eval_loss": 1.035604476928711, | |
| "eval_runtime": 4.5385, | |
| "eval_samples_per_second": 14.102, | |
| "eval_steps_per_second": 1.763, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "grad_norm": 11.629775047302246, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 4.0086, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.39473684210526316, | |
| "grad_norm": 10.649725914001465, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.8607, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "grad_norm": 9.19792366027832, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.954, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "eval_loss": 0.8341699838638306, | |
| "eval_runtime": 4.56, | |
| "eval_samples_per_second": 14.035, | |
| "eval_steps_per_second": 1.754, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5526315789473685, | |
| "grad_norm": 8.501199722290039, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 3.2274, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 10.514701843261719, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.3657, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.7105263157894737, | |
| "grad_norm": 8.145644187927246, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 3.5328, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7105263157894737, | |
| "eval_loss": 0.7960277199745178, | |
| "eval_runtime": 4.5667, | |
| "eval_samples_per_second": 14.015, | |
| "eval_steps_per_second": 1.752, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7894736842105263, | |
| "grad_norm": 9.123873710632324, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 3.727, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.868421052631579, | |
| "grad_norm": 9.119816780090332, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 3.2364, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "grad_norm": 9.756755828857422, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 3.4796, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "eval_loss": 0.7768706679344177, | |
| "eval_runtime": 4.5592, | |
| "eval_samples_per_second": 14.038, | |
| "eval_steps_per_second": 1.755, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.0263157894736843, | |
| "grad_norm": 6.43715238571167, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 3.1494, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.1052631578947367, | |
| "grad_norm": 8.016036033630371, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.5075, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.1842105263157894, | |
| "grad_norm": 6.886407375335693, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.4376, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.1842105263157894, | |
| "eval_loss": 0.7812040448188782, | |
| "eval_runtime": 4.5624, | |
| "eval_samples_per_second": 14.028, | |
| "eval_steps_per_second": 1.753, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.263157894736842, | |
| "grad_norm": 7.848536014556885, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.4708, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.3421052631578947, | |
| "grad_norm": 8.303677558898926, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.6233, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.4210526315789473, | |
| "grad_norm": 8.685861587524414, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.6195, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.4210526315789473, | |
| "eval_loss": 0.765683650970459, | |
| "eval_runtime": 4.5582, | |
| "eval_samples_per_second": 14.041, | |
| "eval_steps_per_second": 1.755, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 6.90402364730835, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.3836, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.5789473684210527, | |
| "grad_norm": 7.729911804199219, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.1328, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.6578947368421053, | |
| "grad_norm": 8.159226417541504, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.2059, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.6578947368421053, | |
| "eval_loss": 0.771750271320343, | |
| "eval_runtime": 4.5594, | |
| "eval_samples_per_second": 14.037, | |
| "eval_steps_per_second": 1.755, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.736842105263158, | |
| "grad_norm": 9.652267456054688, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.2225, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.8157894736842106, | |
| "grad_norm": 8.405353546142578, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.1885, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.8947368421052633, | |
| "grad_norm": 8.815597534179688, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.3782, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.8947368421052633, | |
| "eval_loss": 0.7682980298995972, | |
| "eval_runtime": 4.5642, | |
| "eval_samples_per_second": 14.022, | |
| "eval_steps_per_second": 1.753, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.973684210526316, | |
| "grad_norm": 7.776499271392822, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 1.9714, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.0526315789473686, | |
| "grad_norm": 6.823435306549072, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 1.7944, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.1315789473684212, | |
| "grad_norm": 7.627166271209717, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 1.5251, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.1315789473684212, | |
| "eval_loss": 0.7713338732719421, | |
| "eval_runtime": 4.5611, | |
| "eval_samples_per_second": 14.032, | |
| "eval_steps_per_second": 1.754, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.2105263157894735, | |
| "grad_norm": 7.461455821990967, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 1.5446, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.2894736842105265, | |
| "grad_norm": 6.697714328765869, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 1.5799, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.3684210526315788, | |
| "grad_norm": 7.722606182098389, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 1.5105, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.3684210526315788, | |
| "eval_loss": 0.7937512993812561, | |
| "eval_runtime": 4.5625, | |
| "eval_samples_per_second": 14.027, | |
| "eval_steps_per_second": 1.753, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.4473684210526314, | |
| "grad_norm": 6.9743523597717285, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 1.5261, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.526315789473684, | |
| "grad_norm": 7.8254241943359375, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 1.4583, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.6052631578947367, | |
| "grad_norm": 7.711915016174316, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 1.4774, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.6052631578947367, | |
| "eval_loss": 0.7982271313667297, | |
| "eval_runtime": 4.5631, | |
| "eval_samples_per_second": 14.025, | |
| "eval_steps_per_second": 1.753, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3860456117357773e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |