| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5154639175257731, | |
| "eval_steps": 9, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.020618556701030927, | |
| "grad_norm": 3.2483479976654053, | |
| "learning_rate": 1e-05, | |
| "loss": 3.488, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.020618556701030927, | |
| "eval_loss": 3.5999021530151367, | |
| "eval_runtime": 1.3665, | |
| "eval_samples_per_second": 30.003, | |
| "eval_steps_per_second": 4.391, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.041237113402061855, | |
| "grad_norm": 3.2766311168670654, | |
| "learning_rate": 2e-05, | |
| "loss": 3.5399, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.061855670103092786, | |
| "grad_norm": 3.8570547103881836, | |
| "learning_rate": 3e-05, | |
| "loss": 3.5997, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08247422680412371, | |
| "grad_norm": 3.21763014793396, | |
| "learning_rate": 4e-05, | |
| "loss": 3.4567, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.10309278350515463, | |
| "grad_norm": 3.6865789890289307, | |
| "learning_rate": 5e-05, | |
| "loss": 3.5648, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12371134020618557, | |
| "grad_norm": 3.4130187034606934, | |
| "learning_rate": 6e-05, | |
| "loss": 3.4646, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.14432989690721648, | |
| "grad_norm": 3.0942752361297607, | |
| "learning_rate": 7e-05, | |
| "loss": 3.6277, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.16494845360824742, | |
| "grad_norm": 2.5932466983795166, | |
| "learning_rate": 8e-05, | |
| "loss": 3.3956, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.18556701030927836, | |
| "grad_norm": 1.657394289970398, | |
| "learning_rate": 9e-05, | |
| "loss": 3.2032, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.18556701030927836, | |
| "eval_loss": 3.2119903564453125, | |
| "eval_runtime": 0.4139, | |
| "eval_samples_per_second": 99.05, | |
| "eval_steps_per_second": 14.495, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "grad_norm": 1.3544855117797852, | |
| "learning_rate": 0.0001, | |
| "loss": 3.1933, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2268041237113402, | |
| "grad_norm": 1.2699248790740967, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 3.2105, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.24742268041237114, | |
| "grad_norm": 1.3860818147659302, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.0634, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.26804123711340205, | |
| "grad_norm": 1.409780502319336, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 3.1134, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.28865979381443296, | |
| "grad_norm": 1.366316318511963, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 3.061, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.30927835051546393, | |
| "grad_norm": 1.3080459833145142, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.0381, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.32989690721649484, | |
| "grad_norm": 1.0734084844589233, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 2.8557, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.35051546391752575, | |
| "grad_norm": 1.0526894330978394, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 2.9582, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.3711340206185567, | |
| "grad_norm": 1.0270817279815674, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.995, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3711340206185567, | |
| "eval_loss": 2.911524772644043, | |
| "eval_runtime": 0.4107, | |
| "eval_samples_per_second": 99.835, | |
| "eval_steps_per_second": 14.61, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3917525773195876, | |
| "grad_norm": 0.9986397624015808, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.9881, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.41237113402061853, | |
| "grad_norm": 0.9932209849357605, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 2.8199, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4329896907216495, | |
| "grad_norm": 0.9946835041046143, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.8456, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.4536082474226804, | |
| "grad_norm": 0.8482059240341187, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 2.7425, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.4742268041237113, | |
| "grad_norm": 0.9843814969062805, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 2.727, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.4948453608247423, | |
| "grad_norm": 0.7832030057907104, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 2.7126, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5154639175257731, | |
| "grad_norm": 0.813600480556488, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.7697, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 347093965209600.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |