| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.2727272727272725, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 0.12492965161800385, | |
| "learning_rate": 1e-05, | |
| "loss": 3.3057, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.09090909090909091, | |
| "eval_loss": 1.6252607107162476, | |
| "eval_runtime": 2.7168, | |
| "eval_samples_per_second": 3.681, | |
| "eval_steps_per_second": 0.736, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 0.1324862390756607, | |
| "learning_rate": 2e-05, | |
| "loss": 3.2578, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2727272727272727, | |
| "grad_norm": 0.125848650932312, | |
| "learning_rate": 3e-05, | |
| "loss": 3.272, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2727272727272727, | |
| "eval_loss": 1.6250454187393188, | |
| "eval_runtime": 2.7157, | |
| "eval_samples_per_second": 3.682, | |
| "eval_steps_per_second": 0.736, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.14208167791366577, | |
| "learning_rate": 4e-05, | |
| "loss": 3.2557, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.45454545454545453, | |
| "grad_norm": 0.14454525709152222, | |
| "learning_rate": 5e-05, | |
| "loss": 3.31, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 0.14084851741790771, | |
| "learning_rate": 6e-05, | |
| "loss": 3.2925, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "eval_loss": 1.6203479766845703, | |
| "eval_runtime": 2.7169, | |
| "eval_samples_per_second": 3.681, | |
| "eval_steps_per_second": 0.736, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6363636363636364, | |
| "grad_norm": 0.1460854858160019, | |
| "learning_rate": 7e-05, | |
| "loss": 3.1833, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.1497233808040619, | |
| "learning_rate": 8e-05, | |
| "loss": 3.2238, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8181818181818182, | |
| "grad_norm": 0.1676853448152542, | |
| "learning_rate": 9e-05, | |
| "loss": 3.1513, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.8181818181818182, | |
| "eval_loss": 1.6018472909927368, | |
| "eval_runtime": 2.7205, | |
| "eval_samples_per_second": 3.676, | |
| "eval_steps_per_second": 0.735, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 0.23749925196170807, | |
| "learning_rate": 0.0001, | |
| "loss": 3.2127, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.23205164074897766, | |
| "learning_rate": 9.953429730181653e-05, | |
| "loss": 3.1366, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.2311161458492279, | |
| "learning_rate": 9.814586436738998e-05, | |
| "loss": 3.1547, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "eval_loss": 1.5531282424926758, | |
| "eval_runtime": 2.7195, | |
| "eval_samples_per_second": 3.677, | |
| "eval_steps_per_second": 0.735, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.1818181818181819, | |
| "grad_norm": 0.22115594148635864, | |
| "learning_rate": 9.586056507527266e-05, | |
| "loss": 3.004, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 0.23537488281726837, | |
| "learning_rate": 9.272097022732443e-05, | |
| "loss": 3.1026, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 0.20603659749031067, | |
| "learning_rate": 8.8785564535221e-05, | |
| "loss": 3.0531, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.3636363636363638, | |
| "eval_loss": 1.5018327236175537, | |
| "eval_runtime": 2.7165, | |
| "eval_samples_per_second": 3.681, | |
| "eval_steps_per_second": 0.736, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.28566017746925354, | |
| "learning_rate": 8.412765716093272e-05, | |
| "loss": 3.0285, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.5454545454545454, | |
| "grad_norm": 0.2917863130569458, | |
| "learning_rate": 7.883401610574336e-05, | |
| "loss": 2.9987, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": 0.2859671711921692, | |
| "learning_rate": 7.300325188655761e-05, | |
| "loss": 3.0485, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "eval_loss": 1.447091817855835, | |
| "eval_runtime": 2.7157, | |
| "eval_samples_per_second": 3.682, | |
| "eval_steps_per_second": 0.736, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.7272727272727273, | |
| "grad_norm": 0.28030309081077576, | |
| "learning_rate": 6.674398060854931e-05, | |
| "loss": 2.8872, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.3111637532711029, | |
| "learning_rate": 6.01728006526317e-05, | |
| "loss": 2.9068, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.9090909090909092, | |
| "grad_norm": 0.26893553137779236, | |
| "learning_rate": 5.341212066823355e-05, | |
| "loss": 2.8959, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.9090909090909092, | |
| "eval_loss": 1.3965473175048828, | |
| "eval_runtime": 2.7175, | |
| "eval_samples_per_second": 3.68, | |
| "eval_steps_per_second": 0.736, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.2833939492702484, | |
| "learning_rate": 4.658787933176646e-05, | |
| "loss": 2.7861, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.090909090909091, | |
| "grad_norm": 0.30153441429138184, | |
| "learning_rate": 3.982719934736832e-05, | |
| "loss": 2.8688, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.33502814173698425, | |
| "learning_rate": 3.325601939145069e-05, | |
| "loss": 2.7297, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "eval_loss": 1.359919786453247, | |
| "eval_runtime": 2.7165, | |
| "eval_samples_per_second": 3.681, | |
| "eval_steps_per_second": 0.736, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.2727272727272725, | |
| "grad_norm": 0.3763825297355652, | |
| "learning_rate": 2.6996748113442394e-05, | |
| "loss": 2.7016, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 33, | |
| "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": 6.3478299623424e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |