| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.07355645457888929, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001471129091577786, | |
| "eval_loss": 2.445326089859009, | |
| "eval_runtime": 103.0161, | |
| "eval_samples_per_second": 11.115, | |
| "eval_steps_per_second": 1.398, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004413387274733358, | |
| "grad_norm": 1.040910005569458, | |
| "learning_rate": 1.5e-05, | |
| "loss": 9.5736, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00735564545788893, | |
| "eval_loss": 2.441359758377075, | |
| "eval_runtime": 103.7991, | |
| "eval_samples_per_second": 11.031, | |
| "eval_steps_per_second": 1.387, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008826774549466716, | |
| "grad_norm": 1.0511504411697388, | |
| "learning_rate": 3e-05, | |
| "loss": 9.7365, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.013240161824200073, | |
| "grad_norm": 1.3530985116958618, | |
| "learning_rate": 4.5e-05, | |
| "loss": 9.7514, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01471129091577786, | |
| "eval_loss": 2.3896217346191406, | |
| "eval_runtime": 103.8136, | |
| "eval_samples_per_second": 11.029, | |
| "eval_steps_per_second": 1.387, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.017653549098933432, | |
| "grad_norm": 1.658168077468872, | |
| "learning_rate": 4.9692208514878444e-05, | |
| "loss": 9.3972, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02206693637366679, | |
| "grad_norm": 1.7478517293930054, | |
| "learning_rate": 4.8096988312782174e-05, | |
| "loss": 9.1795, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02206693637366679, | |
| "eval_loss": 2.2163238525390625, | |
| "eval_runtime": 103.8055, | |
| "eval_samples_per_second": 11.03, | |
| "eval_steps_per_second": 1.387, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.026480323648400146, | |
| "grad_norm": 1.6007380485534668, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 8.6175, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02942258183155572, | |
| "eval_loss": 2.0439629554748535, | |
| "eval_runtime": 103.8112, | |
| "eval_samples_per_second": 11.03, | |
| "eval_steps_per_second": 1.387, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.030893710923133505, | |
| "grad_norm": 1.5845534801483154, | |
| "learning_rate": 4.123620120825459e-05, | |
| "loss": 8.3277, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.035307098197866864, | |
| "grad_norm": 1.4348173141479492, | |
| "learning_rate": 3.634976249348867e-05, | |
| "loss": 7.9879, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.036778227289444645, | |
| "eval_loss": 1.9180830717086792, | |
| "eval_runtime": 103.7801, | |
| "eval_samples_per_second": 11.033, | |
| "eval_steps_per_second": 1.388, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03972048547260022, | |
| "grad_norm": 1.5030665397644043, | |
| "learning_rate": 3.083613409639764e-05, | |
| "loss": 7.6862, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04413387274733358, | |
| "grad_norm": 1.541954517364502, | |
| "learning_rate": 2.5e-05, | |
| "loss": 7.2887, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04413387274733358, | |
| "eval_loss": 1.839240550994873, | |
| "eval_runtime": 103.7652, | |
| "eval_samples_per_second": 11.035, | |
| "eval_steps_per_second": 1.388, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04854726002206693, | |
| "grad_norm": 1.4720203876495361, | |
| "learning_rate": 1.9163865903602374e-05, | |
| "loss": 7.1466, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.05148951820522251, | |
| "eval_loss": 1.7863552570343018, | |
| "eval_runtime": 103.7831, | |
| "eval_samples_per_second": 11.033, | |
| "eval_steps_per_second": 1.388, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.05296064729680029, | |
| "grad_norm": 1.4572272300720215, | |
| "learning_rate": 1.3650237506511331e-05, | |
| "loss": 7.2036, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05737403457153365, | |
| "grad_norm": 1.4571818113327026, | |
| "learning_rate": 8.763798791745411e-06, | |
| "loss": 6.8959, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.05884516366311144, | |
| "eval_loss": 1.7566184997558594, | |
| "eval_runtime": 103.7914, | |
| "eval_samples_per_second": 11.032, | |
| "eval_steps_per_second": 1.387, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06178742184626701, | |
| "grad_norm": 1.4031445980072021, | |
| "learning_rate": 4.7745751406263165e-06, | |
| "loss": 6.9848, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.06620080912100036, | |
| "grad_norm": 1.314190149307251, | |
| "learning_rate": 1.9030116872178316e-06, | |
| "loss": 7.0053, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06620080912100036, | |
| "eval_loss": 1.7448129653930664, | |
| "eval_runtime": 103.8, | |
| "eval_samples_per_second": 11.031, | |
| "eval_steps_per_second": 1.387, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.07061419639573373, | |
| "grad_norm": 1.3296202421188354, | |
| "learning_rate": 3.077914851215585e-07, | |
| "loss": 6.9776, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.07355645457888929, | |
| "eval_loss": 1.742841124534607, | |
| "eval_runtime": 103.7854, | |
| "eval_samples_per_second": 11.032, | |
| "eval_steps_per_second": 1.387, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.53464644354048e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |