| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.006941431670281995, | |
| "eval_steps": 50, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.470715835140998e-05, | |
| "eval_loss": 1.488693118095398, | |
| "eval_runtime": 333.2635, | |
| "eval_samples_per_second": 36.404, | |
| "eval_steps_per_second": 18.202, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001735357917570499, | |
| "grad_norm": 0.0363650768995285, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7421, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0003470715835140998, | |
| "grad_norm": 0.04388673976063728, | |
| "learning_rate": 0.0001, | |
| "loss": 0.9049, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005206073752711496, | |
| "grad_norm": 0.04997824877500534, | |
| "learning_rate": 9.98292246503335e-05, | |
| "loss": 0.9169, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0006941431670281996, | |
| "grad_norm": 0.07242757081985474, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 0.9632, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0008676789587852494, | |
| "grad_norm": 0.08268485963344574, | |
| "learning_rate": 9.847001329696653e-05, | |
| "loss": 1.1005, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0010412147505422993, | |
| "grad_norm": 0.09746357053518295, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 1.3431, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0012147505422993492, | |
| "grad_norm": 0.179508239030838, | |
| "learning_rate": 9.578866633275288e-05, | |
| "loss": 1.4761, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0013882863340563992, | |
| "grad_norm": 0.21443940699100494, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 1.8133, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.001561822125813449, | |
| "grad_norm": 0.4208425283432007, | |
| "learning_rate": 9.185832391312644e-05, | |
| "loss": 2.0637, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0017353579175704988, | |
| "grad_norm": 0.8397026062011719, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 3.094, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0017353579175704988, | |
| "eval_loss": 1.4274080991744995, | |
| "eval_runtime": 331.799, | |
| "eval_samples_per_second": 36.564, | |
| "eval_steps_per_second": 18.282, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0019088937093275488, | |
| "grad_norm": 0.04962216690182686, | |
| "learning_rate": 8.678619553365659e-05, | |
| "loss": 0.7878, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.0020824295010845985, | |
| "grad_norm": 0.06914501637220383, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 0.9213, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0022559652928416485, | |
| "grad_norm": 0.0765681117773056, | |
| "learning_rate": 8.07106356344834e-05, | |
| "loss": 0.9687, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.0024295010845986984, | |
| "grad_norm": 0.08268511295318604, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 1.0734, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.0026030368763557484, | |
| "grad_norm": 0.09841444343328476, | |
| "learning_rate": 7.379736965185368e-05, | |
| "loss": 1.1728, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0027765726681127983, | |
| "grad_norm": 0.14268898963928223, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 1.3481, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.0029501084598698483, | |
| "grad_norm": 0.21441680192947388, | |
| "learning_rate": 6.623497346023418e-05, | |
| "loss": 1.429, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.003123644251626898, | |
| "grad_norm": 0.22959406673908234, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 1.6465, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0032971800433839477, | |
| "grad_norm": 0.46887078881263733, | |
| "learning_rate": 5.8229729514036705e-05, | |
| "loss": 1.7265, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.0034707158351409977, | |
| "grad_norm": 1.4313136339187622, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 3.0376, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0034707158351409977, | |
| "eval_loss": 1.3536286354064941, | |
| "eval_runtime": 331.6476, | |
| "eval_samples_per_second": 36.581, | |
| "eval_steps_per_second": 18.29, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0036442516268980476, | |
| "grad_norm": 0.0890691876411438, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8272, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.0038177874186550976, | |
| "grad_norm": 0.08088309317827225, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 0.8692, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.0039913232104121475, | |
| "grad_norm": 0.08706238865852356, | |
| "learning_rate": 4.17702704859633e-05, | |
| "loss": 0.9777, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.004164859002169197, | |
| "grad_norm": 0.10868410766124725, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 0.9109, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.004338394793926247, | |
| "grad_norm": 0.12020254135131836, | |
| "learning_rate": 3.3765026539765834e-05, | |
| "loss": 1.1462, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.004511930585683297, | |
| "grad_norm": 0.17492125928401947, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 1.2319, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.004685466377440347, | |
| "grad_norm": 0.21846410632133484, | |
| "learning_rate": 2.6202630348146324e-05, | |
| "loss": 1.505, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.004859002169197397, | |
| "grad_norm": 0.3237517476081848, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 1.6528, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.005032537960954447, | |
| "grad_norm": 0.5105227828025818, | |
| "learning_rate": 1.928936436551661e-05, | |
| "loss": 1.8486, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.005206073752711497, | |
| "grad_norm": 1.2406115531921387, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 2.7017, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.005206073752711497, | |
| "eval_loss": 1.3264102935791016, | |
| "eval_runtime": 334.9033, | |
| "eval_samples_per_second": 36.225, | |
| "eval_steps_per_second": 18.113, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.005379609544468546, | |
| "grad_norm": 0.07182818651199341, | |
| "learning_rate": 1.3213804466343421e-05, | |
| "loss": 0.6991, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.005553145336225597, | |
| "grad_norm": 0.07443557679653168, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 0.8821, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.005726681127982646, | |
| "grad_norm": 0.08473718911409378, | |
| "learning_rate": 8.141676086873572e-06, | |
| "loss": 0.9581, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.0059002169197396965, | |
| "grad_norm": 0.10495265573263168, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 0.9973, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.006073752711496746, | |
| "grad_norm": 0.11671217530965805, | |
| "learning_rate": 4.2113336672471245e-06, | |
| "loss": 1.1315, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.006247288503253796, | |
| "grad_norm": 0.19537697732448578, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 1.1631, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.006420824295010846, | |
| "grad_norm": 0.1836855262517929, | |
| "learning_rate": 1.5299867030334814e-06, | |
| "loss": 1.2259, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.0065943600867678955, | |
| "grad_norm": 0.27528730034828186, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 1.5324, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.006767895878524946, | |
| "grad_norm": 0.41066810488700867, | |
| "learning_rate": 1.7077534966650766e-07, | |
| "loss": 2.0259, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.006941431670281995, | |
| "grad_norm": 1.2721341848373413, | |
| "learning_rate": 0.0, | |
| "loss": 2.6547, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.006941431670281995, | |
| "eval_loss": 1.3213353157043457, | |
| "eval_runtime": 333.0889, | |
| "eval_samples_per_second": 36.423, | |
| "eval_steps_per_second": 18.211, | |
| "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": 3202304118620160.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |