| { | |
| "best_metric": 0.2832432985305786, | |
| "best_model_checkpoint": "/content/best_model/checkpoint-65000", | |
| "epoch": 2.0, | |
| "global_step": 69122, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.6385376580538756e-05, | |
| "loss": 0.3587, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.883381724357605, | |
| "eval_loss": 0.313699334859848, | |
| "eval_runtime": 130.2101, | |
| "eval_samples_per_second": 222.195, | |
| "eval_steps_per_second": 3.479, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.2770029802378405e-05, | |
| "loss": 0.3318, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.8830706477165222, | |
| "eval_loss": 0.31002119183540344, | |
| "eval_runtime": 130.4095, | |
| "eval_samples_per_second": 221.855, | |
| "eval_steps_per_second": 3.474, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.915395966551894e-05, | |
| "loss": 0.3286, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.8863887786865234, | |
| "eval_loss": 0.303303599357605, | |
| "eval_runtime": 130.6696, | |
| "eval_samples_per_second": 221.413, | |
| "eval_steps_per_second": 3.467, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 3.5537889528659474e-05, | |
| "loss": 0.3236, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_accuracy": 0.8862159252166748, | |
| "eval_loss": 0.3037019670009613, | |
| "eval_runtime": 130.8546, | |
| "eval_samples_per_second": 221.1, | |
| "eval_steps_per_second": 3.462, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 3.1922542750499116e-05, | |
| "loss": 0.3182, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_accuracy": 0.8876330852508545, | |
| "eval_loss": 0.2938973009586334, | |
| "eval_runtime": 130.8776, | |
| "eval_samples_per_second": 221.061, | |
| "eval_steps_per_second": 3.461, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.830647261363965e-05, | |
| "loss": 0.3129, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.8885317444801331, | |
| "eval_loss": 0.2910305857658386, | |
| "eval_runtime": 131.4925, | |
| "eval_samples_per_second": 220.028, | |
| "eval_steps_per_second": 3.445, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2.469184919417841e-05, | |
| "loss": 0.3078, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_accuracy": 0.8887391090393066, | |
| "eval_loss": 0.29142218828201294, | |
| "eval_runtime": 131.659, | |
| "eval_samples_per_second": 219.75, | |
| "eval_steps_per_second": 3.441, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 2.107794913341628e-05, | |
| "loss": 0.2791, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_accuracy": 0.8873911499977112, | |
| "eval_loss": 0.2975434362888336, | |
| "eval_runtime": 132.2605, | |
| "eval_samples_per_second": 218.75, | |
| "eval_steps_per_second": 3.425, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.7462602355255925e-05, | |
| "loss": 0.2723, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.8906055688858032, | |
| "eval_loss": 0.29133233428001404, | |
| "eval_runtime": 132.58, | |
| "eval_samples_per_second": 218.223, | |
| "eval_steps_per_second": 3.417, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.384580885969735e-05, | |
| "loss": 0.2724, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_accuracy": 0.8903636336326599, | |
| "eval_loss": 0.2879472076892853, | |
| "eval_runtime": 132.7015, | |
| "eval_samples_per_second": 218.023, | |
| "eval_steps_per_second": 3.414, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.0229738722837882e-05, | |
| "loss": 0.27, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_accuracy": 0.8910548686981201, | |
| "eval_loss": 0.28737255930900574, | |
| "eval_runtime": 132.683, | |
| "eval_samples_per_second": 218.054, | |
| "eval_steps_per_second": 3.414, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 6.615115303376639e-06, | |
| "loss": 0.2681, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_accuracy": 0.8927831053733826, | |
| "eval_loss": 0.28480109572410583, | |
| "eval_runtime": 132.2639, | |
| "eval_samples_per_second": 218.744, | |
| "eval_steps_per_second": 3.425, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 2.9997685252162845e-06, | |
| "loss": 0.2672, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "eval_accuracy": 0.893439769744873, | |
| "eval_loss": 0.2832432985305786, | |
| "eval_runtime": 132.917, | |
| "eval_samples_per_second": 217.67, | |
| "eval_steps_per_second": 3.408, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 69122, | |
| "total_flos": 2.955371035238809e+17, | |
| "train_loss": 0.29886255871073064, | |
| "train_runtime": 56655.4623, | |
| "train_samples_per_second": 78.081, | |
| "train_steps_per_second": 1.22 | |
| } | |
| ], | |
| "max_steps": 69122, | |
| "num_train_epochs": 2, | |
| "total_flos": 2.955371035238809e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |