| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.7467248908296944, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.017467248908296942, | |
| "eval_loss": 2.185687303543091, | |
| "eval_runtime": 8.5055, | |
| "eval_samples_per_second": 11.404, | |
| "eval_steps_per_second": 1.528, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05240174672489083, | |
| "grad_norm": 3.7532355785369873, | |
| "learning_rate": 3e-05, | |
| "loss": 2.1745, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.10480349344978165, | |
| "grad_norm": 3.853607177734375, | |
| "learning_rate": 6e-05, | |
| "loss": 2.0202, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1572052401746725, | |
| "grad_norm": 2.7308859825134277, | |
| "learning_rate": 9e-05, | |
| "loss": 1.2437, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1572052401746725, | |
| "eval_loss": 0.7501007318496704, | |
| "eval_runtime": 8.6626, | |
| "eval_samples_per_second": 11.198, | |
| "eval_steps_per_second": 1.501, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2096069868995633, | |
| "grad_norm": 1.2397022247314453, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.6726, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.26200873362445415, | |
| "grad_norm": 0.9129244089126587, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.4927, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.314410480349345, | |
| "grad_norm": 0.7560355067253113, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.4404, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.314410480349345, | |
| "eval_loss": 0.3994991183280945, | |
| "eval_runtime": 8.7008, | |
| "eval_samples_per_second": 11.148, | |
| "eval_steps_per_second": 1.494, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.36681222707423583, | |
| "grad_norm": 0.7913711071014404, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.3265, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.4192139737991266, | |
| "grad_norm": 0.6146355271339417, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.3328, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.47161572052401746, | |
| "grad_norm": 0.762830913066864, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.3289, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.47161572052401746, | |
| "eval_loss": 0.3602748215198517, | |
| "eval_runtime": 8.7103, | |
| "eval_samples_per_second": 11.136, | |
| "eval_steps_per_second": 1.492, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5240174672489083, | |
| "grad_norm": 0.7961021661758423, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.3447, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5764192139737991, | |
| "grad_norm": 0.6712778210639954, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.3778, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.62882096069869, | |
| "grad_norm": 0.5632585883140564, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.2967, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.62882096069869, | |
| "eval_loss": 0.3325504958629608, | |
| "eval_runtime": 8.6977, | |
| "eval_samples_per_second": 11.152, | |
| "eval_steps_per_second": 1.495, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6812227074235808, | |
| "grad_norm": 0.4797973930835724, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.2961, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.7336244541484717, | |
| "grad_norm": 0.5742776989936829, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.3068, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.7860262008733624, | |
| "grad_norm": 0.6687835454940796, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.2926, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7860262008733624, | |
| "eval_loss": 0.31762516498565674, | |
| "eval_runtime": 8.7094, | |
| "eval_samples_per_second": 11.137, | |
| "eval_steps_per_second": 1.493, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.8384279475982532, | |
| "grad_norm": 0.8334570527076721, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.324, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.8908296943231441, | |
| "grad_norm": 0.730705976486206, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.2717, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.9432314410480349, | |
| "grad_norm": 0.7632585763931274, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.335, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.9432314410480349, | |
| "eval_loss": 0.31074467301368713, | |
| "eval_runtime": 8.7163, | |
| "eval_samples_per_second": 11.129, | |
| "eval_steps_per_second": 1.491, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.9956331877729258, | |
| "grad_norm": 0.6203595399856567, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.2952, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.0480349344978166, | |
| "grad_norm": 0.6411959528923035, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.3086, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.1004366812227073, | |
| "grad_norm": 0.5493583083152771, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.2665, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.1004366812227073, | |
| "eval_loss": 0.30614474415779114, | |
| "eval_runtime": 8.7092, | |
| "eval_samples_per_second": 11.138, | |
| "eval_steps_per_second": 1.493, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.1528384279475983, | |
| "grad_norm": 0.8020545840263367, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.1996, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.205240174672489, | |
| "grad_norm": 0.5899050235748291, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.2316, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.25764192139738, | |
| "grad_norm": 0.793855607509613, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.234, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.25764192139738, | |
| "eval_loss": 0.3036545515060425, | |
| "eval_runtime": 8.7238, | |
| "eval_samples_per_second": 11.119, | |
| "eval_steps_per_second": 1.49, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.3100436681222707, | |
| "grad_norm": 0.7186128497123718, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.205, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.3624454148471616, | |
| "grad_norm": 0.621354877948761, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.2249, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.4148471615720524, | |
| "grad_norm": 0.820740818977356, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.2493, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.4148471615720524, | |
| "eval_loss": 0.29848697781562805, | |
| "eval_runtime": 8.7163, | |
| "eval_samples_per_second": 11.129, | |
| "eval_steps_per_second": 1.491, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.467248908296943, | |
| "grad_norm": 0.5625925660133362, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.1867, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.519650655021834, | |
| "grad_norm": 0.6035133004188538, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.2114, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.572052401746725, | |
| "grad_norm": 0.598051905632019, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.2172, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.572052401746725, | |
| "eval_loss": 0.29645535349845886, | |
| "eval_runtime": 8.7188, | |
| "eval_samples_per_second": 11.125, | |
| "eval_steps_per_second": 1.491, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.6244541484716157, | |
| "grad_norm": 0.899366021156311, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.2307, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.6768558951965065, | |
| "grad_norm": 0.6130616068840027, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.2302, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.7292576419213974, | |
| "grad_norm": 0.7148019075393677, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.2076, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.7292576419213974, | |
| "eval_loss": 0.29628854990005493, | |
| "eval_runtime": 8.709, | |
| "eval_samples_per_second": 11.138, | |
| "eval_steps_per_second": 1.493, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.717877747126108e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |