| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02915408967091217, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00029448575425163807, | |
| "eval_loss": 2.0751140117645264, | |
| "eval_runtime": 392.124, | |
| "eval_samples_per_second": 14.587, | |
| "eval_steps_per_second": 1.823, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008834572627549142, | |
| "grad_norm": 45.209259033203125, | |
| "learning_rate": 3e-05, | |
| "loss": 8.1417, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0017669145255098284, | |
| "grad_norm": 18.42889404296875, | |
| "learning_rate": 6e-05, | |
| "loss": 7.1146, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.002650371788264743, | |
| "grad_norm": 13.495532035827637, | |
| "learning_rate": 9e-05, | |
| "loss": 4.8841, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002650371788264743, | |
| "eval_loss": 1.0568773746490479, | |
| "eval_runtime": 394.6909, | |
| "eval_samples_per_second": 14.492, | |
| "eval_steps_per_second": 1.812, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.003533829051019657, | |
| "grad_norm": 11.649324417114258, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.78, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004417286313774571, | |
| "grad_norm": 8.40984058380127, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.8775, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.005300743576529486, | |
| "grad_norm": 7.721532344818115, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.3807, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.005300743576529486, | |
| "eval_loss": 0.9211275577545166, | |
| "eval_runtime": 394.5819, | |
| "eval_samples_per_second": 14.496, | |
| "eval_steps_per_second": 1.812, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0061842008392844, | |
| "grad_norm": 7.611982345581055, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 3.5986, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.007067658102039314, | |
| "grad_norm": 7.277873992919922, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.7016, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.007951115364794228, | |
| "grad_norm": 9.115721702575684, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 3.4477, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.007951115364794228, | |
| "eval_loss": 0.8963199853897095, | |
| "eval_runtime": 394.8293, | |
| "eval_samples_per_second": 14.487, | |
| "eval_steps_per_second": 1.811, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.008834572627549142, | |
| "grad_norm": 7.103928565979004, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 3.4513, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.009718029890304056, | |
| "grad_norm": 7.205202102661133, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 3.4688, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.010601487153058971, | |
| "grad_norm": 7.348625183105469, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 3.675, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.010601487153058971, | |
| "eval_loss": 0.8810924291610718, | |
| "eval_runtime": 394.7481, | |
| "eval_samples_per_second": 14.49, | |
| "eval_steps_per_second": 1.811, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.011484944415813885, | |
| "grad_norm": 6.225783348083496, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 3.5757, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0123684016785688, | |
| "grad_norm": 5.358260631561279, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 3.7153, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.013251858941323713, | |
| "grad_norm": 5.572663307189941, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 3.1355, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.013251858941323713, | |
| "eval_loss": 0.8735527396202087, | |
| "eval_runtime": 394.6194, | |
| "eval_samples_per_second": 14.495, | |
| "eval_steps_per_second": 1.812, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.014135316204078627, | |
| "grad_norm": 6.943516254425049, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 3.6308, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.015018773466833541, | |
| "grad_norm": 7.251932144165039, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 3.4569, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.015902230729588455, | |
| "grad_norm": 6.627459526062012, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 3.3764, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.015902230729588455, | |
| "eval_loss": 0.86681067943573, | |
| "eval_runtime": 394.4174, | |
| "eval_samples_per_second": 14.502, | |
| "eval_steps_per_second": 1.813, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01678568799234337, | |
| "grad_norm": 6.218847751617432, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 3.5507, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.017669145255098283, | |
| "grad_norm": 6.713202953338623, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 3.2378, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0185526025178532, | |
| "grad_norm": 6.8192548751831055, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 3.4102, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0185526025178532, | |
| "eval_loss": 0.8609820604324341, | |
| "eval_runtime": 394.5997, | |
| "eval_samples_per_second": 14.496, | |
| "eval_steps_per_second": 1.812, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.01943605978060811, | |
| "grad_norm": 6.667771816253662, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 3.5032, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.020319517043363027, | |
| "grad_norm": 6.182212829589844, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 3.9349, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.021202974306117943, | |
| "grad_norm": 5.752615928649902, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 3.2835, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.021202974306117943, | |
| "eval_loss": 0.8611263036727905, | |
| "eval_runtime": 394.6571, | |
| "eval_samples_per_second": 14.494, | |
| "eval_steps_per_second": 1.812, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.022086431568872855, | |
| "grad_norm": 5.570738792419434, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 3.4432, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02296988883162777, | |
| "grad_norm": 5.480838775634766, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 3.2455, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.023853346094382683, | |
| "grad_norm": 5.031819820404053, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 3.3175, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.023853346094382683, | |
| "eval_loss": 0.8552263379096985, | |
| "eval_runtime": 394.699, | |
| "eval_samples_per_second": 14.492, | |
| "eval_steps_per_second": 1.812, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0247368033571376, | |
| "grad_norm": 5.9167256355285645, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 3.5925, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.02562026061989251, | |
| "grad_norm": 6.1761155128479, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 3.0721, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.026503717882647427, | |
| "grad_norm": 5.942464351654053, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 3.1059, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.026503717882647427, | |
| "eval_loss": 0.853427529335022, | |
| "eval_runtime": 394.7149, | |
| "eval_samples_per_second": 14.491, | |
| "eval_steps_per_second": 1.811, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.027387175145402343, | |
| "grad_norm": 5.856942653656006, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 3.0394, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.028270632408157255, | |
| "grad_norm": 5.515998363494873, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 3.1025, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.02915408967091217, | |
| "grad_norm": 5.73545503616333, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 3.2693, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.02915408967091217, | |
| "eval_loss": 0.8529771566390991, | |
| "eval_runtime": 394.735, | |
| "eval_samples_per_second": 14.491, | |
| "eval_steps_per_second": 1.811, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3928086979267789e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |