| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.23255813953488372, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002325581395348837, | |
| "eval_loss": 5.767578601837158, | |
| "eval_runtime": 10.4189, | |
| "eval_samples_per_second": 69.489, | |
| "eval_steps_per_second": 8.734, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0069767441860465115, | |
| "grad_norm": 0.7177151441574097, | |
| "learning_rate": 3e-05, | |
| "loss": 5.7725, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.013953488372093023, | |
| "grad_norm": 0.6705913543701172, | |
| "learning_rate": 6e-05, | |
| "loss": 5.6508, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.020930232558139535, | |
| "grad_norm": 0.9152357578277588, | |
| "learning_rate": 9e-05, | |
| "loss": 5.7661, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.020930232558139535, | |
| "eval_loss": 5.740058898925781, | |
| "eval_runtime": 10.4015, | |
| "eval_samples_per_second": 69.605, | |
| "eval_steps_per_second": 8.749, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.027906976744186046, | |
| "grad_norm": 0.8683664798736572, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 5.7731, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03488372093023256, | |
| "grad_norm": 1.1061254739761353, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 5.4242, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04186046511627907, | |
| "grad_norm": 0.99778813123703, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 5.2717, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04186046511627907, | |
| "eval_loss": 5.513608932495117, | |
| "eval_runtime": 10.4282, | |
| "eval_samples_per_second": 69.427, | |
| "eval_steps_per_second": 8.726, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04883720930232558, | |
| "grad_norm": 1.3651623725891113, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 5.5186, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05581395348837209, | |
| "grad_norm": 1.554046630859375, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 5.3002, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06279069767441861, | |
| "grad_norm": 1.4339934587478638, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 5.1694, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06279069767441861, | |
| "eval_loss": 5.061354637145996, | |
| "eval_runtime": 10.4152, | |
| "eval_samples_per_second": 69.514, | |
| "eval_steps_per_second": 8.737, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06976744186046512, | |
| "grad_norm": 1.417418360710144, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 4.9417, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07674418604651163, | |
| "grad_norm": 1.4718533754348755, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 4.5364, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.08372093023255814, | |
| "grad_norm": 1.4996187686920166, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 4.6917, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.08372093023255814, | |
| "eval_loss": 4.580844402313232, | |
| "eval_runtime": 10.4483, | |
| "eval_samples_per_second": 69.294, | |
| "eval_steps_per_second": 8.71, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09069767441860466, | |
| "grad_norm": 1.4007647037506104, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 4.4604, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.09767441860465116, | |
| "grad_norm": 1.404478907585144, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 4.2265, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.10465116279069768, | |
| "grad_norm": 1.4139697551727295, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 4.1217, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10465116279069768, | |
| "eval_loss": 4.179461479187012, | |
| "eval_runtime": 10.381, | |
| "eval_samples_per_second": 69.743, | |
| "eval_steps_per_second": 8.766, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11162790697674418, | |
| "grad_norm": 1.4293689727783203, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 3.9897, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1186046511627907, | |
| "grad_norm": 1.2574753761291504, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 4.0715, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.12558139534883722, | |
| "grad_norm": 1.068290114402771, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 3.8083, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.12558139534883722, | |
| "eval_loss": 3.9013290405273438, | |
| "eval_runtime": 10.3754, | |
| "eval_samples_per_second": 69.78, | |
| "eval_steps_per_second": 8.771, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.1325581395348837, | |
| "grad_norm": 1.0287175178527832, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 3.9619, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.13953488372093023, | |
| "grad_norm": 1.1125946044921875, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 3.6843, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.14651162790697675, | |
| "grad_norm": 0.8303821086883545, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 3.6188, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.14651162790697675, | |
| "eval_loss": 3.723705768585205, | |
| "eval_runtime": 10.3865, | |
| "eval_samples_per_second": 69.706, | |
| "eval_steps_per_second": 8.761, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.15348837209302327, | |
| "grad_norm": 1.2538200616836548, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 3.5128, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.16046511627906976, | |
| "grad_norm": 0.9986847639083862, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 3.5111, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.16744186046511628, | |
| "grad_norm": 1.1189088821411133, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 3.654, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.16744186046511628, | |
| "eval_loss": 3.6171345710754395, | |
| "eval_runtime": 10.3906, | |
| "eval_samples_per_second": 69.678, | |
| "eval_steps_per_second": 8.758, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1744186046511628, | |
| "grad_norm": 0.94745934009552, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 3.7317, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1813953488372093, | |
| "grad_norm": 0.8033557534217834, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 3.7478, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.1883720930232558, | |
| "grad_norm": 0.942162036895752, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 3.4858, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.1883720930232558, | |
| "eval_loss": 3.5632054805755615, | |
| "eval_runtime": 10.3798, | |
| "eval_samples_per_second": 69.751, | |
| "eval_steps_per_second": 8.767, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.19534883720930232, | |
| "grad_norm": 0.8051215410232544, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 3.3875, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.20232558139534884, | |
| "grad_norm": 1.0949082374572754, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 3.6471, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.20930232558139536, | |
| "grad_norm": 0.9996029138565063, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 3.5462, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.20930232558139536, | |
| "eval_loss": 3.5418734550476074, | |
| "eval_runtime": 10.3865, | |
| "eval_samples_per_second": 69.706, | |
| "eval_steps_per_second": 8.761, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.21627906976744185, | |
| "grad_norm": 1.184890627861023, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 3.6468, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.22325581395348837, | |
| "grad_norm": 1.0085941553115845, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 3.5317, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2302325581395349, | |
| "grad_norm": 0.8848790526390076, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 3.4723, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.2302325581395349, | |
| "eval_loss": 3.538015842437744, | |
| "eval_runtime": 10.3899, | |
| "eval_samples_per_second": 69.683, | |
| "eval_steps_per_second": 8.759, | |
| "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": 6271342215168000.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |