| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.032609138711123795, | |
| "eval_steps": 50, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00016304569355561895, | |
| "eval_loss": 0.9726263880729675, | |
| "eval_runtime": 94.6739, | |
| "eval_samples_per_second": 27.283, | |
| "eval_steps_per_second": 13.647, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008152284677780948, | |
| "grad_norm": 0.07707637548446655, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6907, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0016304569355561896, | |
| "grad_norm": 0.055902931839227676, | |
| "learning_rate": 0.0001, | |
| "loss": 0.676, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0024456854033342843, | |
| "grad_norm": 0.07526875287294388, | |
| "learning_rate": 9.98292246503335e-05, | |
| "loss": 0.701, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.003260913871112379, | |
| "grad_norm": 0.11501318961381912, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 0.7658, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.004076142338890474, | |
| "grad_norm": 0.12424055486917496, | |
| "learning_rate": 9.847001329696653e-05, | |
| "loss": 0.8096, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.004891370806668569, | |
| "grad_norm": 0.1637168526649475, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 0.8079, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.005706599274446664, | |
| "grad_norm": 0.17115910351276398, | |
| "learning_rate": 9.578866633275288e-05, | |
| "loss": 0.8622, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.006521827742224758, | |
| "grad_norm": 0.3511226177215576, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 0.9294, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0073370562100028534, | |
| "grad_norm": 0.2993008494377136, | |
| "learning_rate": 9.185832391312644e-05, | |
| "loss": 0.9747, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.008152284677780949, | |
| "grad_norm": 0.6940411925315857, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 1.188, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.008152284677780949, | |
| "eval_loss": 0.8416158556938171, | |
| "eval_runtime": 94.5672, | |
| "eval_samples_per_second": 27.314, | |
| "eval_steps_per_second": 13.662, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.008967513145559042, | |
| "grad_norm": 0.11171204596757889, | |
| "learning_rate": 8.678619553365659e-05, | |
| "loss": 0.6554, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.009782741613337137, | |
| "grad_norm": 0.13064613938331604, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 0.6926, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.010597970081115233, | |
| "grad_norm": 0.11693071573972702, | |
| "learning_rate": 8.07106356344834e-05, | |
| "loss": 0.7102, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.011413198548893328, | |
| "grad_norm": 0.1502353698015213, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 0.6994, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.012228427016671423, | |
| "grad_norm": 0.15267403423786163, | |
| "learning_rate": 7.379736965185368e-05, | |
| "loss": 0.7595, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.013043655484449516, | |
| "grad_norm": 0.2149164080619812, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 0.762, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.013858883952227612, | |
| "grad_norm": 0.2367481142282486, | |
| "learning_rate": 6.623497346023418e-05, | |
| "loss": 0.8421, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.014674112420005707, | |
| "grad_norm": 0.35609787702560425, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 0.7825, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.015489340887783802, | |
| "grad_norm": 0.5247498750686646, | |
| "learning_rate": 5.8229729514036705e-05, | |
| "loss": 0.9606, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.016304569355561897, | |
| "grad_norm": 1.1298545598983765, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 1.2002, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.016304569355561897, | |
| "eval_loss": 0.8084781765937805, | |
| "eval_runtime": 94.6181, | |
| "eval_samples_per_second": 27.299, | |
| "eval_steps_per_second": 13.655, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.017119797823339993, | |
| "grad_norm": 0.1418055295944214, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5892, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.017935026291118084, | |
| "grad_norm": 0.1185089647769928, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 0.5924, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.01875025475889618, | |
| "grad_norm": 0.13358411192893982, | |
| "learning_rate": 4.17702704859633e-05, | |
| "loss": 0.6427, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.019565483226674275, | |
| "grad_norm": 0.15520331263542175, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 0.7182, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.02038071169445237, | |
| "grad_norm": 0.20835325121879578, | |
| "learning_rate": 3.3765026539765834e-05, | |
| "loss": 0.7745, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.021195940162230465, | |
| "grad_norm": 0.15979139506816864, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 0.7509, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.02201116863000856, | |
| "grad_norm": 0.26534563302993774, | |
| "learning_rate": 2.6202630348146324e-05, | |
| "loss": 0.8185, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.022826397097786656, | |
| "grad_norm": 0.2851661145687103, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 0.8134, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.02364162556556475, | |
| "grad_norm": 0.6048436760902405, | |
| "learning_rate": 1.928936436551661e-05, | |
| "loss": 0.8977, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.024456854033342846, | |
| "grad_norm": 0.7918521761894226, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 1.1297, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.024456854033342846, | |
| "eval_loss": 0.7911348342895508, | |
| "eval_runtime": 94.5989, | |
| "eval_samples_per_second": 27.305, | |
| "eval_steps_per_second": 13.658, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.025272082501120938, | |
| "grad_norm": 0.0935765951871872, | |
| "learning_rate": 1.3213804466343421e-05, | |
| "loss": 0.6047, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.026087310968899033, | |
| "grad_norm": 0.13173741102218628, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 0.6212, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.026902539436677128, | |
| "grad_norm": 0.13724230229854584, | |
| "learning_rate": 8.141676086873572e-06, | |
| "loss": 0.7117, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.027717767904455223, | |
| "grad_norm": 0.13333597779273987, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 0.678, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.02853299637223332, | |
| "grad_norm": 0.1563432365655899, | |
| "learning_rate": 4.2113336672471245e-06, | |
| "loss": 0.8257, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.029348224840011414, | |
| "grad_norm": 0.17387519776821136, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 0.8124, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.03016345330778951, | |
| "grad_norm": 0.2212686836719513, | |
| "learning_rate": 1.5299867030334814e-06, | |
| "loss": 0.7535, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.030978681775567604, | |
| "grad_norm": 0.25696220993995667, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 0.7698, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.0317939102433457, | |
| "grad_norm": 0.4329439699649811, | |
| "learning_rate": 1.7077534966650766e-07, | |
| "loss": 0.9063, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.032609138711123795, | |
| "grad_norm": 0.7288855910301208, | |
| "learning_rate": 0.0, | |
| "loss": 1.0139, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.032609138711123795, | |
| "eval_loss": 0.7942183017730713, | |
| "eval_runtime": 94.3849, | |
| "eval_samples_per_second": 27.367, | |
| "eval_steps_per_second": 13.689, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 1.7574856463745024e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |