| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 222, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.13513513513513514, |
| "grad_norm": 4.831328868865967, |
| "learning_rate": 2.9985590229902075e-05, |
| "loss": 0.8729, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.2702702702702703, |
| "grad_norm": 3.3961567878723145, |
| "learning_rate": 2.9730185992421254e-05, |
| "loss": 0.9016, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.40540540540540543, |
| "grad_norm": 3.0312085151672363, |
| "learning_rate": 2.9160833414327637e-05, |
| "loss": 0.9115, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.5405405405405406, |
| "grad_norm": 2.8146772384643555, |
| "learning_rate": 2.828966725734167e-05, |
| "loss": 0.8631, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.6756756756756757, |
| "grad_norm": 3.070246934890747, |
| "learning_rate": 2.7135254915624213e-05, |
| "loss": 0.8696, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.8108108108108109, |
| "grad_norm": 2.657665729522705, |
| "learning_rate": 2.5722200684005718e-05, |
| "loss": 0.9504, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.9459459459459459, |
| "grad_norm": 2.7406342029571533, |
| "learning_rate": 2.408062136015872e-05, |
| "loss": 0.7631, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.0810810810810811, |
| "grad_norm": 2.587312698364258, |
| "learning_rate": 2.224550435734272e-05, |
| "loss": 0.6282, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.2162162162162162, |
| "grad_norm": 2.679584503173828, |
| "learning_rate": 2.025596200843394e-05, |
| "loss": 0.4663, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.3513513513513513, |
| "grad_norm": 2.1560540199279785, |
| "learning_rate": 1.8154397954454996e-05, |
| "loss": 0.4216, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.4864864864864864, |
| "grad_norm": 3.28458833694458, |
| "learning_rate": 1.5985603384585545e-05, |
| "loss": 0.4337, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.6216216216216215, |
| "grad_norm": 2.6721553802490234, |
| "learning_rate": 1.3795802389729185e-05, |
| "loss": 0.4256, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.7567567567567568, |
| "grad_norm": 2.691438674926758, |
| "learning_rate": 1.1631666776267804e-05, |
| "loss": 0.4605, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.8918918918918919, |
| "grad_norm": 2.9018781185150146, |
| "learning_rate": 9.5393213375378e-06, |
| "loss": 0.4006, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.027027027027027, |
| "grad_norm": 1.536293864250183, |
| "learning_rate": 7.563360783939724e-06, |
| "loss": 0.4041, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.1621621621621623, |
| "grad_norm": 2.0216639041900635, |
| "learning_rate": 5.7458992841091544e-06, |
| "loss": 0.1893, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.2972972972972974, |
| "grad_norm": 2.6039671897888184, |
| "learning_rate": 4.125672874528797e-06, |
| "loss": 0.1901, |
| "step": 170 |
| }, |
| { |
| "epoch": 2.4324324324324325, |
| "grad_norm": 2.349864959716797, |
| "learning_rate": 2.7372138681624246e-06, |
| "loss": 0.165, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.5675675675675675, |
| "grad_norm": 1.7878508567810059, |
| "learning_rate": 1.6101148581573276e-06, |
| "loss": 0.1793, |
| "step": 190 |
| }, |
| { |
| "epoch": 2.7027027027027026, |
| "grad_norm": 1.9807162284851074, |
| "learning_rate": 7.683980030985655e-07, |
| "loss": 0.169, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.8378378378378377, |
| "grad_norm": 2.102729558944702, |
| "learning_rate": 2.300030364048139e-07, |
| "loss": 0.1955, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.972972972972973, |
| "grad_norm": 1.9260261058807373, |
| "learning_rate": 6.404912056610091e-09, |
| "loss": 0.1692, |
| "step": 220 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 222, |
| "total_flos": 2.842963155615744e+16, |
| "train_loss": 0.49890191973866643, |
| "train_runtime": 996.419, |
| "train_samples_per_second": 3.541, |
| "train_steps_per_second": 0.223 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 222, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": false, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.842963155615744e+16, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
