| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05859947260474656, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0023439789041898623, | |
| "grad_norm": 30.85260009765625, | |
| "learning_rate": 5e-05, | |
| "loss": 6.6922, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0023439789041898623, | |
| "eval_loss": 7.640185832977295, | |
| "eval_runtime": 375.7713, | |
| "eval_samples_per_second": 1.913, | |
| "eval_steps_per_second": 0.479, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004687957808379725, | |
| "grad_norm": 32.24323272705078, | |
| "learning_rate": 0.0001, | |
| "loss": 7.3599, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.007031936712569587, | |
| "grad_norm": 35.10329818725586, | |
| "learning_rate": 9.953429730181653e-05, | |
| "loss": 4.3299, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.007031936712569587, | |
| "eval_loss": 1.007381796836853, | |
| "eval_runtime": 377.1204, | |
| "eval_samples_per_second": 1.907, | |
| "eval_steps_per_second": 0.477, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00937591561675945, | |
| "grad_norm": 12.174251556396484, | |
| "learning_rate": 9.814586436738998e-05, | |
| "loss": 1.0876, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.011719894520949311, | |
| "grad_norm": 11.167458534240723, | |
| "learning_rate": 9.586056507527266e-05, | |
| "loss": 0.618, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.014063873425139173, | |
| "grad_norm": 16.5855770111084, | |
| "learning_rate": 9.272097022732443e-05, | |
| "loss": 0.4434, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.014063873425139173, | |
| "eval_loss": 0.5613827109336853, | |
| "eval_runtime": 377.1269, | |
| "eval_samples_per_second": 1.907, | |
| "eval_steps_per_second": 0.477, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.016407852329329037, | |
| "grad_norm": 2.6942837238311768, | |
| "learning_rate": 8.8785564535221e-05, | |
| "loss": 0.3475, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0187518312335189, | |
| "grad_norm": 4.204878807067871, | |
| "learning_rate": 8.412765716093272e-05, | |
| "loss": 0.4269, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.02109581013770876, | |
| "grad_norm": 8.653458595275879, | |
| "learning_rate": 7.883401610574336e-05, | |
| "loss": 0.8488, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02109581013770876, | |
| "eval_loss": 0.46133139729499817, | |
| "eval_runtime": 376.6374, | |
| "eval_samples_per_second": 1.909, | |
| "eval_steps_per_second": 0.478, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.023439789041898623, | |
| "grad_norm": 6.0319647789001465, | |
| "learning_rate": 7.300325188655761e-05, | |
| "loss": 0.5979, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.025783767946088484, | |
| "grad_norm": 7.0973124504089355, | |
| "learning_rate": 6.674398060854931e-05, | |
| "loss": 0.3395, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.028127746850278346, | |
| "grad_norm": 3.4368035793304443, | |
| "learning_rate": 6.01728006526317e-05, | |
| "loss": 0.4039, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.028127746850278346, | |
| "eval_loss": 0.3852250874042511, | |
| "eval_runtime": 376.4542, | |
| "eval_samples_per_second": 1.91, | |
| "eval_steps_per_second": 0.478, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.030471725754468208, | |
| "grad_norm": 4.774641990661621, | |
| "learning_rate": 5.341212066823355e-05, | |
| "loss": 0.4755, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.032815704658658074, | |
| "grad_norm": 3.387488842010498, | |
| "learning_rate": 4.658787933176646e-05, | |
| "loss": 0.3168, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03515968356284793, | |
| "grad_norm": 4.079291343688965, | |
| "learning_rate": 3.982719934736832e-05, | |
| "loss": 0.3634, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03515968356284793, | |
| "eval_loss": 0.3480370044708252, | |
| "eval_runtime": 376.8136, | |
| "eval_samples_per_second": 1.908, | |
| "eval_steps_per_second": 0.478, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0375036624670378, | |
| "grad_norm": 2.9585719108581543, | |
| "learning_rate": 3.325601939145069e-05, | |
| "loss": 0.2674, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.039847641371227656, | |
| "grad_norm": 3.4793038368225098, | |
| "learning_rate": 2.6996748113442394e-05, | |
| "loss": 0.2099, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.04219162027541752, | |
| "grad_norm": 3.6350109577178955, | |
| "learning_rate": 2.1165983894256647e-05, | |
| "loss": 0.2057, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04219162027541752, | |
| "eval_loss": 0.30665451288223267, | |
| "eval_runtime": 376.8336, | |
| "eval_samples_per_second": 1.908, | |
| "eval_steps_per_second": 0.478, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04453559917960739, | |
| "grad_norm": 2.6633102893829346, | |
| "learning_rate": 1.5872342839067306e-05, | |
| "loss": 0.2433, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.046879578083797245, | |
| "grad_norm": 3.862442970275879, | |
| "learning_rate": 1.1214435464779006e-05, | |
| "loss": 0.2717, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04922355698798711, | |
| "grad_norm": 4.074159145355225, | |
| "learning_rate": 7.2790297726755716e-06, | |
| "loss": 0.2596, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.04922355698798711, | |
| "eval_loss": 0.32189393043518066, | |
| "eval_runtime": 376.6717, | |
| "eval_samples_per_second": 1.909, | |
| "eval_steps_per_second": 0.478, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05156753589217697, | |
| "grad_norm": 4.781056880950928, | |
| "learning_rate": 4.139434924727359e-06, | |
| "loss": 0.4503, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.053911514796366834, | |
| "grad_norm": 6.350940227508545, | |
| "learning_rate": 1.8541356326100433e-06, | |
| "loss": 0.3268, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.05625549370055669, | |
| "grad_norm": 3.5974953174591064, | |
| "learning_rate": 4.6570269818346224e-07, | |
| "loss": 0.362, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.05625549370055669, | |
| "eval_loss": 0.31786561012268066, | |
| "eval_runtime": 376.8204, | |
| "eval_samples_per_second": 1.908, | |
| "eval_steps_per_second": 0.478, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.05859947260474656, | |
| "grad_norm": 5.433014392852783, | |
| "learning_rate": 0.0, | |
| "loss": 0.2848, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.508513578745856e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |