| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.46296296296296297, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004629629629629629, | |
| "eval_loss": 0.8352945446968079, | |
| "eval_runtime": 37.1582, | |
| "eval_samples_per_second": 9.796, | |
| "eval_steps_per_second": 1.238, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.013888888888888888, | |
| "grad_norm": 3.743543863296509, | |
| "learning_rate": 3e-05, | |
| "loss": 3.5477, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.027777777777777776, | |
| "grad_norm": 2.9249703884124756, | |
| "learning_rate": 6e-05, | |
| "loss": 3.2728, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.041666666666666664, | |
| "grad_norm": 1.9571139812469482, | |
| "learning_rate": 9e-05, | |
| "loss": 2.4194, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.041666666666666664, | |
| "eval_loss": 0.5313647389411926, | |
| "eval_runtime": 38.0465, | |
| "eval_samples_per_second": 9.567, | |
| "eval_steps_per_second": 1.209, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05555555555555555, | |
| "grad_norm": 2.0869150161743164, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.9516, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.06944444444444445, | |
| "grad_norm": 1.3190362453460693, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.7807, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08333333333333333, | |
| "grad_norm": 1.532779574394226, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.8552, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08333333333333333, | |
| "eval_loss": 0.4421681761741638, | |
| "eval_runtime": 38.1844, | |
| "eval_samples_per_second": 9.533, | |
| "eval_steps_per_second": 1.205, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.09722222222222222, | |
| "grad_norm": 1.317895770072937, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.7407, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 1.1588882207870483, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.5437, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 1.1349965333938599, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.6061, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "eval_loss": 0.4252684414386749, | |
| "eval_runtime": 38.1826, | |
| "eval_samples_per_second": 9.533, | |
| "eval_steps_per_second": 1.205, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1388888888888889, | |
| "grad_norm": 1.0014010667800903, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.5592, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1527777777777778, | |
| "grad_norm": 0.9527996182441711, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 1.5534, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 1.0669186115264893, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.6002, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "eval_loss": 0.41367948055267334, | |
| "eval_runtime": 38.1988, | |
| "eval_samples_per_second": 9.529, | |
| "eval_steps_per_second": 1.204, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.18055555555555555, | |
| "grad_norm": 0.9425536394119263, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.626, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.19444444444444445, | |
| "grad_norm": 1.298422932624817, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 1.6206, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.20833333333333334, | |
| "grad_norm": 0.9776718616485596, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 1.6982, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.20833333333333334, | |
| "eval_loss": 0.4053753912448883, | |
| "eval_runtime": 38.148, | |
| "eval_samples_per_second": 9.542, | |
| "eval_steps_per_second": 1.206, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 0.9453088045120239, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 1.5541, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.2361111111111111, | |
| "grad_norm": 0.8852896094322205, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 1.5389, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.2749226093292236, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 1.5278, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.4023856222629547, | |
| "eval_runtime": 38.139, | |
| "eval_samples_per_second": 9.544, | |
| "eval_steps_per_second": 1.206, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.2638888888888889, | |
| "grad_norm": 1.0323377847671509, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 1.5173, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 1.0452280044555664, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 1.536, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2916666666666667, | |
| "grad_norm": 1.195465087890625, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 1.5933, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.2916666666666667, | |
| "eval_loss": 0.39836037158966064, | |
| "eval_runtime": 38.1947, | |
| "eval_samples_per_second": 9.53, | |
| "eval_steps_per_second": 1.204, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.3055555555555556, | |
| "grad_norm": 0.9835031032562256, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 1.5645, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.3194444444444444, | |
| "grad_norm": 1.1227846145629883, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 1.4236, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 0.9475577473640442, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 1.614, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "eval_loss": 0.3959140479564667, | |
| "eval_runtime": 38.2146, | |
| "eval_samples_per_second": 9.525, | |
| "eval_steps_per_second": 1.204, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3472222222222222, | |
| "grad_norm": 1.0517405271530151, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 1.4999, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.3611111111111111, | |
| "grad_norm": 1.2465274333953857, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 1.7315, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 1.0259209871292114, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 1.3891, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "eval_loss": 0.3939442038536072, | |
| "eval_runtime": 38.1987, | |
| "eval_samples_per_second": 9.529, | |
| "eval_steps_per_second": 1.204, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.3888888888888889, | |
| "grad_norm": 1.0821902751922607, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 1.5456, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.4027777777777778, | |
| "grad_norm": 0.9735687971115112, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 1.5431, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.4166666666666667, | |
| "grad_norm": 1.082150936126709, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 1.575, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4166666666666667, | |
| "eval_loss": 0.3928387761116028, | |
| "eval_runtime": 38.151, | |
| "eval_samples_per_second": 9.541, | |
| "eval_steps_per_second": 1.206, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4305555555555556, | |
| "grad_norm": 1.0945594310760498, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 1.4757, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 0.995895266532898, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 1.546, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.4583333333333333, | |
| "grad_norm": 1.0993353128433228, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 1.6258, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.4583333333333333, | |
| "eval_loss": 0.392814964056015, | |
| "eval_runtime": 38.1529, | |
| "eval_samples_per_second": 9.541, | |
| "eval_steps_per_second": 1.206, | |
| "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": 2.28728047140864e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |