| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.025350952245143707, | |
| "eval_steps": 50, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00012675476122571855, | |
| "eval_loss": 1.4765392541885376, | |
| "eval_runtime": 262.6502, | |
| "eval_samples_per_second": 12.648, | |
| "eval_steps_per_second": 6.324, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006337738061285927, | |
| "grad_norm": 5.484171390533447, | |
| "learning_rate": 5e-05, | |
| "loss": 3.2689, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0012675476122571854, | |
| "grad_norm": 5.716429710388184, | |
| "learning_rate": 0.0001, | |
| "loss": 3.1198, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0019013214183857781, | |
| "grad_norm": 5.180782794952393, | |
| "learning_rate": 9.98292246503335e-05, | |
| "loss": 3.5239, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.002535095224514371, | |
| "grad_norm": 6.50588846206665, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 3.5008, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0031688690306429634, | |
| "grad_norm": 7.72395133972168, | |
| "learning_rate": 9.847001329696653e-05, | |
| "loss": 3.4503, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0038026428367715563, | |
| "grad_norm": 9.041804313659668, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 3.5008, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.004436416642900149, | |
| "grad_norm": 9.004871368408203, | |
| "learning_rate": 9.578866633275288e-05, | |
| "loss": 3.5921, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.005070190449028742, | |
| "grad_norm": 9.65662670135498, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 3.8512, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.005703964255157334, | |
| "grad_norm": 10.689106941223145, | |
| "learning_rate": 9.185832391312644e-05, | |
| "loss": 3.6242, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.006337738061285927, | |
| "grad_norm": 25.17250633239746, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 4.5744, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.006337738061285927, | |
| "eval_loss": 0.8999184966087341, | |
| "eval_runtime": 264.74, | |
| "eval_samples_per_second": 12.548, | |
| "eval_steps_per_second": 6.274, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.00697151186741452, | |
| "grad_norm": 4.206664085388184, | |
| "learning_rate": 8.678619553365659e-05, | |
| "loss": 2.9932, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.007605285673543113, | |
| "grad_norm": 3.9618825912475586, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 2.9488, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.008239059479671705, | |
| "grad_norm": 4.880193710327148, | |
| "learning_rate": 8.07106356344834e-05, | |
| "loss": 2.7563, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.008872833285800298, | |
| "grad_norm": 6.973112106323242, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 3.7767, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.00950660709192889, | |
| "grad_norm": 5.642597675323486, | |
| "learning_rate": 7.379736965185368e-05, | |
| "loss": 3.0909, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.010140380898057483, | |
| "grad_norm": 8.437419891357422, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 3.2673, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.010774154704186077, | |
| "grad_norm": 8.672842979431152, | |
| "learning_rate": 6.623497346023418e-05, | |
| "loss": 3.7468, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.011407928510314668, | |
| "grad_norm": 8.314080238342285, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 3.1235, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.012041702316443262, | |
| "grad_norm": 21.81874656677246, | |
| "learning_rate": 5.8229729514036705e-05, | |
| "loss": 2.977, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.012675476122571853, | |
| "grad_norm": 8.898733139038086, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 3.6767, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.012675476122571853, | |
| "eval_loss": 0.8634766936302185, | |
| "eval_runtime": 264.836, | |
| "eval_samples_per_second": 12.544, | |
| "eval_steps_per_second": 6.272, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.013309249928700447, | |
| "grad_norm": 3.161619186401367, | |
| "learning_rate": 5e-05, | |
| "loss": 2.861, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.01394302373482904, | |
| "grad_norm": 3.4773592948913574, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 3.2931, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.014576797540957632, | |
| "grad_norm": 3.9867374897003174, | |
| "learning_rate": 4.17702704859633e-05, | |
| "loss": 3.4508, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.015210571347086225, | |
| "grad_norm": 4.51062536239624, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 3.1185, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.01584434515321482, | |
| "grad_norm": 5.138334274291992, | |
| "learning_rate": 3.3765026539765834e-05, | |
| "loss": 3.1852, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.01647811895934341, | |
| "grad_norm": 7.484903812408447, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 3.5996, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.017111892765472002, | |
| "grad_norm": 7.916966915130615, | |
| "learning_rate": 2.6202630348146324e-05, | |
| "loss": 3.7636, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.017745666571600597, | |
| "grad_norm": 11.922800064086914, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 3.8454, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.01837944037772919, | |
| "grad_norm": 10.14339542388916, | |
| "learning_rate": 1.928936436551661e-05, | |
| "loss": 4.1687, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.01901321418385778, | |
| "grad_norm": 20.12777328491211, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 3.831, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.01901321418385778, | |
| "eval_loss": 0.8437188267707825, | |
| "eval_runtime": 264.6703, | |
| "eval_samples_per_second": 12.551, | |
| "eval_steps_per_second": 6.276, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.019646987989986375, | |
| "grad_norm": 2.489067792892456, | |
| "learning_rate": 1.3213804466343421e-05, | |
| "loss": 2.7438, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.020280761796114967, | |
| "grad_norm": 3.4757256507873535, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 2.8515, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.02091453560224356, | |
| "grad_norm": 4.207156181335449, | |
| "learning_rate": 8.141676086873572e-06, | |
| "loss": 3.3305, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.021548309408372154, | |
| "grad_norm": 4.16589879989624, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 3.2926, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.022182083214500745, | |
| "grad_norm": 5.062546253204346, | |
| "learning_rate": 4.2113336672471245e-06, | |
| "loss": 2.8543, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.022815857020629337, | |
| "grad_norm": 5.446523666381836, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 3.4054, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.02344963082675793, | |
| "grad_norm": 8.401721954345703, | |
| "learning_rate": 1.5299867030334814e-06, | |
| "loss": 3.582, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.024083404632886524, | |
| "grad_norm": 9.53354263305664, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 3.8842, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.024717178439015115, | |
| "grad_norm": 9.483086585998535, | |
| "learning_rate": 1.7077534966650766e-07, | |
| "loss": 3.6604, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.025350952245143707, | |
| "grad_norm": 11.274394035339355, | |
| "learning_rate": 0.0, | |
| "loss": 3.9655, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.025350952245143707, | |
| "eval_loss": 0.8424938917160034, | |
| "eval_runtime": 264.7832, | |
| "eval_samples_per_second": 12.546, | |
| "eval_steps_per_second": 6.273, | |
| "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": 7.092676449848525e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |