| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.07267441860465117, | |
| "eval_steps": 9, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0029069767441860465, | |
| "grad_norm": 0.6269262433052063, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4149, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0029069767441860465, | |
| "eval_loss": 0.32177430391311646, | |
| "eval_runtime": 32.1501, | |
| "eval_samples_per_second": 9.02, | |
| "eval_steps_per_second": 1.151, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005813953488372093, | |
| "grad_norm": 0.40142378211021423, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2732, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00872093023255814, | |
| "grad_norm": 0.3893200755119324, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2537, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.011627906976744186, | |
| "grad_norm": 0.44946348667144775, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2946, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.014534883720930232, | |
| "grad_norm": 0.46690016984939575, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2881, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01744186046511628, | |
| "grad_norm": 0.4786141812801361, | |
| "learning_rate": 6e-05, | |
| "loss": 0.2854, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.020348837209302327, | |
| "grad_norm": 0.637237548828125, | |
| "learning_rate": 7e-05, | |
| "loss": 0.3118, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.023255813953488372, | |
| "grad_norm": 0.35781624913215637, | |
| "learning_rate": 8e-05, | |
| "loss": 0.2149, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.02616279069767442, | |
| "grad_norm": 0.4675982594490051, | |
| "learning_rate": 9e-05, | |
| "loss": 0.1809, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02616279069767442, | |
| "eval_loss": 0.18669699132442474, | |
| "eval_runtime": 31.781, | |
| "eval_samples_per_second": 9.125, | |
| "eval_steps_per_second": 1.164, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.029069767441860465, | |
| "grad_norm": 1.1234859228134155, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1509, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03197674418604651, | |
| "grad_norm": 0.597476601600647, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.1963, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03488372093023256, | |
| "grad_norm": 0.45688414573669434, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.0975, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0377906976744186, | |
| "grad_norm": 0.43219056725502014, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.0719, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.040697674418604654, | |
| "grad_norm": 0.5428093075752258, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.1036, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0436046511627907, | |
| "grad_norm": 0.7716228365898132, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.081, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.046511627906976744, | |
| "grad_norm": 0.8211978673934937, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.1221, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04941860465116279, | |
| "grad_norm": 0.9486254453659058, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.1489, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.05232558139534884, | |
| "grad_norm": 0.5020085573196411, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.1001, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05232558139534884, | |
| "eval_loss": 0.06299833208322525, | |
| "eval_runtime": 31.8162, | |
| "eval_samples_per_second": 9.115, | |
| "eval_steps_per_second": 1.163, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.055232558139534885, | |
| "grad_norm": 0.3962463140487671, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0522, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05813953488372093, | |
| "grad_norm": 0.774312436580658, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.1395, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.061046511627906974, | |
| "grad_norm": 0.4788380265235901, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0497, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.06395348837209303, | |
| "grad_norm": 0.4114557206630707, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.0671, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.06686046511627906, | |
| "grad_norm": 0.18382249772548676, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.0168, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06976744186046512, | |
| "grad_norm": 0.30654194951057434, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0493, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07267441860465117, | |
| "grad_norm": 0.5613061189651489, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.0914, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 1.8918125730791424e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |