| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.056211354693648116, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005621135469364812, | |
| "eval_loss": 0.9150053262710571, | |
| "eval_runtime": 163.2125, | |
| "eval_samples_per_second": 18.363, | |
| "eval_steps_per_second": 2.298, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016863406408094434, | |
| "grad_norm": 0.2956816852092743, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9448, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.003372681281618887, | |
| "grad_norm": 0.2726631164550781, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9084, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00505902192242833, | |
| "grad_norm": 0.21325236558914185, | |
| "learning_rate": 9e-05, | |
| "loss": 0.8568, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00505902192242833, | |
| "eval_loss": 0.8506203293800354, | |
| "eval_runtime": 164.5381, | |
| "eval_samples_per_second": 18.215, | |
| "eval_steps_per_second": 2.279, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006745362563237774, | |
| "grad_norm": 0.2024015486240387, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.841, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008431703204047217, | |
| "grad_norm": 0.21213184297084808, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.8612, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01011804384485666, | |
| "grad_norm": 0.18286366760730743, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.7799, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.01011804384485666, | |
| "eval_loss": 0.7897757887840271, | |
| "eval_runtime": 164.6058, | |
| "eval_samples_per_second": 18.207, | |
| "eval_steps_per_second": 2.278, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.011804384485666104, | |
| "grad_norm": 0.1957426816225052, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.7759, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.013490725126475547, | |
| "grad_norm": 0.1535039246082306, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.7502, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01517706576728499, | |
| "grad_norm": 0.15373417735099792, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.7894, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01517706576728499, | |
| "eval_loss": 0.7679686546325684, | |
| "eval_runtime": 164.638, | |
| "eval_samples_per_second": 18.204, | |
| "eval_steps_per_second": 2.278, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.016863406408094434, | |
| "grad_norm": 0.1481175273656845, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.7242, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01854974704890388, | |
| "grad_norm": 0.14273004233837128, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.697, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.02023608768971332, | |
| "grad_norm": 0.1542915254831314, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.7494, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02023608768971332, | |
| "eval_loss": 0.7546795606613159, | |
| "eval_runtime": 164.7817, | |
| "eval_samples_per_second": 18.188, | |
| "eval_steps_per_second": 2.276, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.021922428330522766, | |
| "grad_norm": 0.1566758006811142, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.7573, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.023608768971332208, | |
| "grad_norm": 0.16109135746955872, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.7655, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.025295109612141653, | |
| "grad_norm": 0.15883490443229675, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.765, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.025295109612141653, | |
| "eval_loss": 0.7461237907409668, | |
| "eval_runtime": 164.7581, | |
| "eval_samples_per_second": 18.19, | |
| "eval_steps_per_second": 2.276, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.026981450252951095, | |
| "grad_norm": 0.15612190961837769, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.7528, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.02866779089376054, | |
| "grad_norm": 0.15889355540275574, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.7408, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.03035413153456998, | |
| "grad_norm": 0.16103751957416534, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.7259, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.03035413153456998, | |
| "eval_loss": 0.7405259013175964, | |
| "eval_runtime": 164.8345, | |
| "eval_samples_per_second": 18.182, | |
| "eval_steps_per_second": 2.275, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.03204047217537943, | |
| "grad_norm": 0.16598886251449585, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.7398, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.03372681281618887, | |
| "grad_norm": 0.16655276715755463, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.7534, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.03541315345699832, | |
| "grad_norm": 0.15536905825138092, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.7346, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.03541315345699832, | |
| "eval_loss": 0.7366407513618469, | |
| "eval_runtime": 164.7935, | |
| "eval_samples_per_second": 18.186, | |
| "eval_steps_per_second": 2.276, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.03709949409780776, | |
| "grad_norm": 0.16439783573150635, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.7293, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.0387858347386172, | |
| "grad_norm": 0.16857261955738068, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.7634, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.04047217537942664, | |
| "grad_norm": 0.15915103256702423, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.7338, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.04047217537942664, | |
| "eval_loss": 0.7342514991760254, | |
| "eval_runtime": 164.7816, | |
| "eval_samples_per_second": 18.188, | |
| "eval_steps_per_second": 2.276, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.04215851602023609, | |
| "grad_norm": 0.1585548371076584, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.7453, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.04384485666104553, | |
| "grad_norm": 0.16192683577537537, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.7035, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.045531197301854974, | |
| "grad_norm": 0.16156958043575287, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.7299, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.045531197301854974, | |
| "eval_loss": 0.7329394817352295, | |
| "eval_runtime": 164.6509, | |
| "eval_samples_per_second": 18.202, | |
| "eval_steps_per_second": 2.278, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.047217537942664416, | |
| "grad_norm": 0.16453050076961517, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.7636, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.048903878583473864, | |
| "grad_norm": 0.17823824286460876, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.7473, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.050590219224283306, | |
| "grad_norm": 0.16380201280117035, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.7086, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.050590219224283306, | |
| "eval_loss": 0.7322941422462463, | |
| "eval_runtime": 164.8671, | |
| "eval_samples_per_second": 18.178, | |
| "eval_steps_per_second": 2.275, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.05227655986509275, | |
| "grad_norm": 0.1683541089296341, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.6752, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.05396290050590219, | |
| "grad_norm": 0.16768647730350494, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.7573, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.05564924114671164, | |
| "grad_norm": 0.15775783360004425, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.6528, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.05564924114671164, | |
| "eval_loss": 0.7322031259536743, | |
| "eval_runtime": 164.8241, | |
| "eval_samples_per_second": 18.183, | |
| "eval_steps_per_second": 2.275, | |
| "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": 5.22507534729216e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |