| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.37523452157598497, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00375234521575985, | |
| "eval_loss": 10.385416984558105, | |
| "eval_runtime": 16.6568, | |
| "eval_samples_per_second": 13.508, | |
| "eval_steps_per_second": 3.422, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01125703564727955, | |
| "grad_norm": 9.703173637390137, | |
| "learning_rate": 3e-05, | |
| "loss": 10.4586, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0225140712945591, | |
| "grad_norm": 8.736605644226074, | |
| "learning_rate": 6e-05, | |
| "loss": 9.0406, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.03377110694183865, | |
| "grad_norm": 9.648565292358398, | |
| "learning_rate": 9e-05, | |
| "loss": 5.2104, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03377110694183865, | |
| "eval_loss": 1.994680404663086, | |
| "eval_runtime": 16.3241, | |
| "eval_samples_per_second": 13.783, | |
| "eval_steps_per_second": 3.492, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0450281425891182, | |
| "grad_norm": 5.035146713256836, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.5712, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05628517823639775, | |
| "grad_norm": 6.059263706207275, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.6058, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0675422138836773, | |
| "grad_norm": 2.648972511291504, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.3104, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0675422138836773, | |
| "eval_loss": 0.18601766228675842, | |
| "eval_runtime": 16.4012, | |
| "eval_samples_per_second": 13.719, | |
| "eval_steps_per_second": 3.475, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07879924953095685, | |
| "grad_norm": 7.399054050445557, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.2436, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0900562851782364, | |
| "grad_norm": 3.5452539920806885, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.246, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.10131332082551595, | |
| "grad_norm": 3.253471851348877, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.1533, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10131332082551595, | |
| "eval_loss": 0.11802710592746735, | |
| "eval_runtime": 16.4414, | |
| "eval_samples_per_second": 13.685, | |
| "eval_steps_per_second": 3.467, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1125703564727955, | |
| "grad_norm": 0.3799557685852051, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.045, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12382739212007504, | |
| "grad_norm": 0.9986750483512878, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0625, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.1350844277673546, | |
| "grad_norm": 0.397352010011673, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0679, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1350844277673546, | |
| "eval_loss": 0.05288050323724747, | |
| "eval_runtime": 16.4545, | |
| "eval_samples_per_second": 13.674, | |
| "eval_steps_per_second": 3.464, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.14634146341463414, | |
| "grad_norm": 0.8268300890922546, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.0183, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1575984990619137, | |
| "grad_norm": 0.09965112060308456, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0271, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.16885553470919323, | |
| "grad_norm": 2.12029767036438, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0276, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.16885553470919323, | |
| "eval_loss": 0.039369650185108185, | |
| "eval_runtime": 16.4644, | |
| "eval_samples_per_second": 13.666, | |
| "eval_steps_per_second": 3.462, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1801125703564728, | |
| "grad_norm": 0.2447308748960495, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0194, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.19136960600375236, | |
| "grad_norm": 0.11860853433609009, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.0215, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.2026266416510319, | |
| "grad_norm": 0.042394720017910004, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0028, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.2026266416510319, | |
| "eval_loss": 0.020199263468384743, | |
| "eval_runtime": 16.4568, | |
| "eval_samples_per_second": 13.672, | |
| "eval_steps_per_second": 3.464, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.21388367729831145, | |
| "grad_norm": 0.01664595864713192, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.0006, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.225140712945591, | |
| "grad_norm": 2.271406888961792, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0372, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.23639774859287055, | |
| "grad_norm": 0.012259761802852154, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.021, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.23639774859287055, | |
| "eval_loss": 0.005822421051561832, | |
| "eval_runtime": 16.4688, | |
| "eval_samples_per_second": 13.662, | |
| "eval_steps_per_second": 3.461, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.24765478424015008, | |
| "grad_norm": 0.1131921261548996, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0062, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.2589118198874296, | |
| "grad_norm": 0.6602727770805359, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.007, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.2701688555347092, | |
| "grad_norm": 0.03234098479151726, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0012, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2701688555347092, | |
| "eval_loss": 0.004046585410833359, | |
| "eval_runtime": 16.4739, | |
| "eval_samples_per_second": 13.658, | |
| "eval_steps_per_second": 3.46, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.28142589118198874, | |
| "grad_norm": 0.008015623316168785, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0002, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.2926829268292683, | |
| "grad_norm": 0.09628616273403168, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.001, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.30393996247654786, | |
| "grad_norm": 0.019364288076758385, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.0024, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.30393996247654786, | |
| "eval_loss": 0.0014989529736340046, | |
| "eval_runtime": 16.4648, | |
| "eval_samples_per_second": 13.666, | |
| "eval_steps_per_second": 3.462, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.3151969981238274, | |
| "grad_norm": 0.004858713131397963, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0003, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.32645403377110693, | |
| "grad_norm": 0.002612839685752988, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0009, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.33771106941838647, | |
| "grad_norm": 0.015446677803993225, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0006, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.33771106941838647, | |
| "eval_loss": 0.001242813770659268, | |
| "eval_runtime": 16.4726, | |
| "eval_samples_per_second": 13.659, | |
| "eval_steps_per_second": 3.46, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.34896810506566606, | |
| "grad_norm": 0.005139649845659733, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0003, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.3602251407129456, | |
| "grad_norm": 0.0025040253531187773, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0007, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.3714821763602251, | |
| "grad_norm": 0.0036368141882121563, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.0003, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.3714821763602251, | |
| "eval_loss": 0.0012390989577397704, | |
| "eval_runtime": 16.4714, | |
| "eval_samples_per_second": 13.66, | |
| "eval_steps_per_second": 3.461, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 7.1094434267136e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |