| { | |
| "best_metric": 2.123983860015869, | |
| "best_model_checkpoint": "/mnt/bn/qingyi-bn-lq/llama/saved_llamaHC3_human/checkpoint-160", | |
| "epoch": 2.8959276018099547, | |
| "global_step": 160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 2.3469691276550293, | |
| "eval_runtime": 40.7893, | |
| "eval_samples_per_second": 49.032, | |
| "eval_steps_per_second": 0.785, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 2.254398822784424, | |
| "eval_runtime": 40.7321, | |
| "eval_samples_per_second": 49.101, | |
| "eval_steps_per_second": 0.786, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 2.216655731201172, | |
| "eval_runtime": 40.7948, | |
| "eval_samples_per_second": 49.026, | |
| "eval_steps_per_second": 0.784, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.000271875, | |
| "loss": 2.265, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.192472219467163, | |
| "eval_runtime": 40.9073, | |
| "eval_samples_per_second": 48.891, | |
| "eval_steps_per_second": 0.782, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 2.1772661209106445, | |
| "eval_runtime": 40.8472, | |
| "eval_samples_per_second": 48.963, | |
| "eval_steps_per_second": 0.783, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_loss": 2.1680707931518555, | |
| "eval_runtime": 40.7376, | |
| "eval_samples_per_second": 49.095, | |
| "eval_steps_per_second": 0.786, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 2.15969181060791, | |
| "eval_runtime": 40.7043, | |
| "eval_samples_per_second": 49.135, | |
| "eval_steps_per_second": 0.786, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00023624999999999997, | |
| "loss": 2.1528, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_loss": 2.1535723209381104, | |
| "eval_runtime": 40.7472, | |
| "eval_samples_per_second": 49.083, | |
| "eval_steps_per_second": 0.785, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_loss": 2.149122476577759, | |
| "eval_runtime": 40.7594, | |
| "eval_samples_per_second": 49.068, | |
| "eval_steps_per_second": 0.785, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 2.1453487873077393, | |
| "eval_runtime": 40.7319, | |
| "eval_samples_per_second": 49.102, | |
| "eval_steps_per_second": 0.786, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.142251968383789, | |
| "eval_runtime": 40.8482, | |
| "eval_samples_per_second": 48.962, | |
| "eval_steps_per_second": 0.783, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.00019874999999999998, | |
| "loss": 2.1332, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 2.139620065689087, | |
| "eval_runtime": 40.7619, | |
| "eval_samples_per_second": 49.065, | |
| "eval_steps_per_second": 0.785, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_loss": 2.137423276901245, | |
| "eval_runtime": 40.783, | |
| "eval_samples_per_second": 49.04, | |
| "eval_steps_per_second": 0.785, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 2.135718822479248, | |
| "eval_runtime": 40.7802, | |
| "eval_samples_per_second": 49.043, | |
| "eval_steps_per_second": 0.785, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 2.134214401245117, | |
| "eval_runtime": 40.7213, | |
| "eval_samples_per_second": 49.114, | |
| "eval_steps_per_second": 0.786, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00016125, | |
| "loss": 2.1152, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_loss": 2.1330068111419678, | |
| "eval_runtime": 40.8447, | |
| "eval_samples_per_second": 48.966, | |
| "eval_steps_per_second": 0.783, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_loss": 2.131664276123047, | |
| "eval_runtime": 40.9633, | |
| "eval_samples_per_second": 48.824, | |
| "eval_steps_per_second": 0.781, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_loss": 2.1305532455444336, | |
| "eval_runtime": 40.7803, | |
| "eval_samples_per_second": 49.043, | |
| "eval_steps_per_second": 0.785, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "eval_loss": 2.1295671463012695, | |
| "eval_runtime": 40.8088, | |
| "eval_samples_per_second": 49.009, | |
| "eval_steps_per_second": 0.784, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.00012374999999999997, | |
| "loss": 2.1138, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_loss": 2.1289596557617188, | |
| "eval_runtime": 40.7573, | |
| "eval_samples_per_second": 49.071, | |
| "eval_steps_per_second": 0.785, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_loss": 2.1280689239501953, | |
| "eval_runtime": 40.8322, | |
| "eval_samples_per_second": 48.981, | |
| "eval_steps_per_second": 0.784, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_loss": 2.127443552017212, | |
| "eval_runtime": 40.9036, | |
| "eval_samples_per_second": 48.895, | |
| "eval_steps_per_second": 0.782, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_loss": 2.126908779144287, | |
| "eval_runtime": 41.0214, | |
| "eval_samples_per_second": 48.755, | |
| "eval_steps_per_second": 0.78, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 8.624999999999998e-05, | |
| "loss": 2.1101, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_loss": 2.126199245452881, | |
| "eval_runtime": 40.9707, | |
| "eval_samples_per_second": 48.815, | |
| "eval_steps_per_second": 0.781, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 2.125767230987549, | |
| "eval_runtime": 40.7834, | |
| "eval_samples_per_second": 49.04, | |
| "eval_steps_per_second": 0.785, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_loss": 2.125452995300293, | |
| "eval_runtime": 40.8009, | |
| "eval_samples_per_second": 49.019, | |
| "eval_steps_per_second": 0.784, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "eval_loss": 2.1250967979431152, | |
| "eval_runtime": 40.7689, | |
| "eval_samples_per_second": 49.057, | |
| "eval_steps_per_second": 0.785, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 4.875e-05, | |
| "loss": 2.108, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 2.124704599380493, | |
| "eval_runtime": 40.7682, | |
| "eval_samples_per_second": 49.058, | |
| "eval_steps_per_second": 0.785, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 2.124401807785034, | |
| "eval_runtime": 40.7695, | |
| "eval_samples_per_second": 49.056, | |
| "eval_steps_per_second": 0.785, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "eval_loss": 2.1242740154266357, | |
| "eval_runtime": 40.8982, | |
| "eval_samples_per_second": 48.902, | |
| "eval_steps_per_second": 0.782, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "eval_loss": 2.124105930328369, | |
| "eval_runtime": 40.7925, | |
| "eval_samples_per_second": 49.029, | |
| "eval_steps_per_second": 0.784, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.1249999999999999e-05, | |
| "loss": 2.0996, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_loss": 2.123983860015869, | |
| "eval_runtime": 40.7694, | |
| "eval_samples_per_second": 49.056, | |
| "eval_steps_per_second": 0.785, | |
| "step": 160 | |
| } | |
| ], | |
| "max_steps": 165, | |
| "num_train_epochs": 3, | |
| "total_flos": 2.354051216792617e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |