| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9999220394480393, | |
| "global_step": 6413, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.610166848588804e-05, | |
| "loss": 0.6465, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_gen_len": 43.125025166096236, | |
| "eval_loss": 0.09457100182771683, | |
| "eval_runtime": 1631.4106, | |
| "eval_samples_per_second": 3.045, | |
| "eval_steps_per_second": 0.381, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.220333697177608e-05, | |
| "loss": 0.0749, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_gen_len": 40.72941413327965, | |
| "eval_loss": 0.08562018722295761, | |
| "eval_runtime": 1556.1203, | |
| "eval_samples_per_second": 3.192, | |
| "eval_steps_per_second": 0.399, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3.830500545766412e-05, | |
| "loss": 0.0695, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_gen_len": 41.34185625125831, | |
| "eval_loss": 0.08309376239776611, | |
| "eval_runtime": 1494.5878, | |
| "eval_samples_per_second": 3.323, | |
| "eval_steps_per_second": 0.415, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3.440667394355216e-05, | |
| "loss": 0.066, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_gen_len": 40.56633782967586, | |
| "eval_loss": 0.07833520323038101, | |
| "eval_runtime": 1580.4677, | |
| "eval_samples_per_second": 3.143, | |
| "eval_steps_per_second": 0.393, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.05083424294402e-05, | |
| "loss": 0.065, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_gen_len": 42.21783772901148, | |
| "eval_loss": 0.07629594951868057, | |
| "eval_runtime": 1515.9865, | |
| "eval_samples_per_second": 3.276, | |
| "eval_steps_per_second": 0.41, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.6610010915328243e-05, | |
| "loss": 0.0628, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_gen_len": 40.085363398429635, | |
| "eval_loss": 0.07593819499015808, | |
| "eval_runtime": 1486.4085, | |
| "eval_samples_per_second": 3.342, | |
| "eval_steps_per_second": 0.418, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.271167940121628e-05, | |
| "loss": 0.0605, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_gen_len": 40.57298167908194, | |
| "eval_loss": 0.07500995695590973, | |
| "eval_runtime": 1478.1922, | |
| "eval_samples_per_second": 3.36, | |
| "eval_steps_per_second": 0.42, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.881334788710432e-05, | |
| "loss": 0.0604, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_gen_len": 41.43708475941212, | |
| "eval_loss": 0.07322078198194504, | |
| "eval_runtime": 1508.2222, | |
| "eval_samples_per_second": 3.293, | |
| "eval_steps_per_second": 0.412, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.491501637299236e-05, | |
| "loss": 0.0597, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_gen_len": 42.330380511375076, | |
| "eval_loss": 0.0725751593708992, | |
| "eval_runtime": 1504.6447, | |
| "eval_samples_per_second": 3.301, | |
| "eval_steps_per_second": 0.413, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.1016684858880399e-05, | |
| "loss": 0.0592, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_gen_len": 42.24562109925508, | |
| "eval_loss": 0.07068128883838654, | |
| "eval_runtime": 1517.39, | |
| "eval_samples_per_second": 3.273, | |
| "eval_steps_per_second": 0.409, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.11835334476844e-06, | |
| "loss": 0.0586, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_gen_len": 43.367827662572985, | |
| "eval_loss": 0.07018809765577316, | |
| "eval_runtime": 1537.9058, | |
| "eval_samples_per_second": 3.23, | |
| "eval_steps_per_second": 0.404, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.220021830656479e-06, | |
| "loss": 0.0577, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_gen_len": 43.2325347292128, | |
| "eval_loss": 0.06985782831907272, | |
| "eval_runtime": 1536.9981, | |
| "eval_samples_per_second": 3.232, | |
| "eval_steps_per_second": 0.404, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 6413, | |
| "total_flos": 4.5543286951870464e+17, | |
| "train_loss": 0.108196587787856, | |
| "train_runtime": 33928.5621, | |
| "train_samples_per_second": 3.024, | |
| "train_steps_per_second": 0.189 | |
| } | |
| ], | |
| "max_steps": 6413, | |
| "num_train_epochs": 1, | |
| "total_flos": 4.5543286951870464e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |