| { | |
| "best_metric": 2.033215284347534, | |
| "best_model_checkpoint": "./outputs/checkpoint-3600", | |
| "epoch": 2.6229508196721314, | |
| "eval_steps": 100, | |
| "global_step": 3600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.7392, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.641645669937134, | |
| "eval_runtime": 216.5632, | |
| "eval_samples_per_second": 28.971, | |
| "eval_steps_per_second": 3.625, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6048, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.5937132835388184, | |
| "eval_runtime": 206.5927, | |
| "eval_samples_per_second": 30.369, | |
| "eval_steps_per_second": 3.8, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5626, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.556633949279785, | |
| "eval_runtime": 206.6295, | |
| "eval_samples_per_second": 30.364, | |
| "eval_steps_per_second": 3.799, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5359, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.52408504486084, | |
| "eval_runtime": 206.6465, | |
| "eval_samples_per_second": 30.361, | |
| "eval_steps_per_second": 3.799, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4941, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.4982783794403076, | |
| "eval_runtime": 206.6237, | |
| "eval_samples_per_second": 30.364, | |
| "eval_steps_per_second": 3.799, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4681, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.4732720851898193, | |
| "eval_runtime": 206.7564, | |
| "eval_samples_per_second": 30.345, | |
| "eval_steps_per_second": 3.797, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4509, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.4523911476135254, | |
| "eval_runtime": 206.8041, | |
| "eval_samples_per_second": 30.338, | |
| "eval_steps_per_second": 3.796, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 2.429, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 2.4297590255737305, | |
| "eval_runtime": 206.8909, | |
| "eval_samples_per_second": 30.325, | |
| "eval_steps_per_second": 3.794, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3922, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 2.4085500240325928, | |
| "eval_runtime": 206.8407, | |
| "eval_samples_per_second": 30.333, | |
| "eval_steps_per_second": 3.795, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3898, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 2.3908660411834717, | |
| "eval_runtime": 206.8251, | |
| "eval_samples_per_second": 30.335, | |
| "eval_steps_per_second": 3.795, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3825, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 2.36782169342041, | |
| "eval_runtime": 206.8621, | |
| "eval_samples_per_second": 30.329, | |
| "eval_steps_per_second": 3.795, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3384, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 2.348783254623413, | |
| "eval_runtime": 206.6622, | |
| "eval_samples_per_second": 30.359, | |
| "eval_steps_per_second": 3.798, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3505, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 2.334380865097046, | |
| "eval_runtime": 206.6361, | |
| "eval_samples_per_second": 30.363, | |
| "eval_steps_per_second": 3.799, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3013, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 2.3094112873077393, | |
| "eval_runtime": 206.8997, | |
| "eval_samples_per_second": 30.324, | |
| "eval_steps_per_second": 3.794, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2619, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 2.293931484222412, | |
| "eval_runtime": 207.4615, | |
| "eval_samples_per_second": 30.242, | |
| "eval_steps_per_second": 3.784, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2542, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 2.275839328765869, | |
| "eval_runtime": 207.0504, | |
| "eval_samples_per_second": 30.302, | |
| "eval_steps_per_second": 3.791, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2529, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 2.2586843967437744, | |
| "eval_runtime": 207.0741, | |
| "eval_samples_per_second": 30.298, | |
| "eval_steps_per_second": 3.791, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2154, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 2.2427449226379395, | |
| "eval_runtime": 207.0755, | |
| "eval_samples_per_second": 30.298, | |
| "eval_steps_per_second": 3.791, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1995, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 2.229428291320801, | |
| "eval_runtime": 207.3553, | |
| "eval_samples_per_second": 30.257, | |
| "eval_steps_per_second": 3.786, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1939, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 2.214716911315918, | |
| "eval_runtime": 207.5218, | |
| "eval_samples_per_second": 30.233, | |
| "eval_steps_per_second": 3.783, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002, | |
| "loss": 2.178, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 2.201956272125244, | |
| "eval_runtime": 207.8863, | |
| "eval_samples_per_second": 30.18, | |
| "eval_steps_per_second": 3.776, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1611, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 2.1847541332244873, | |
| "eval_runtime": 207.4083, | |
| "eval_samples_per_second": 30.25, | |
| "eval_steps_per_second": 3.785, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1507, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 2.1721720695495605, | |
| "eval_runtime": 207.158, | |
| "eval_samples_per_second": 30.286, | |
| "eval_steps_per_second": 3.789, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1581, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 2.1605775356292725, | |
| "eval_runtime": 207.453, | |
| "eval_samples_per_second": 30.243, | |
| "eval_steps_per_second": 3.784, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002, | |
| "loss": 2.126, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 2.146674633026123, | |
| "eval_runtime": 207.6555, | |
| "eval_samples_per_second": 30.214, | |
| "eval_steps_per_second": 3.78, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 2.127, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 2.1346609592437744, | |
| "eval_runtime": 208.0228, | |
| "eval_samples_per_second": 30.16, | |
| "eval_steps_per_second": 3.774, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1099, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 2.12068247795105, | |
| "eval_runtime": 207.6654, | |
| "eval_samples_per_second": 30.212, | |
| "eval_steps_per_second": 3.78, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0663, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 2.1133551597595215, | |
| "eval_runtime": 312.5947, | |
| "eval_samples_per_second": 20.071, | |
| "eval_steps_per_second": 2.511, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0394, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 2.099728584289551, | |
| "eval_runtime": 207.7011, | |
| "eval_samples_per_second": 30.207, | |
| "eval_steps_per_second": 3.779, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0583, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 2.09155011177063, | |
| "eval_runtime": 207.9624, | |
| "eval_samples_per_second": 30.169, | |
| "eval_steps_per_second": 3.775, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0272, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 2.081061840057373, | |
| "eval_runtime": 207.7848, | |
| "eval_samples_per_second": 30.195, | |
| "eval_steps_per_second": 3.778, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0447, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 2.0687415599823, | |
| "eval_runtime": 207.7228, | |
| "eval_samples_per_second": 30.204, | |
| "eval_steps_per_second": 3.779, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0104, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 2.0587158203125, | |
| "eval_runtime": 207.6455, | |
| "eval_samples_per_second": 30.215, | |
| "eval_steps_per_second": 3.78, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0034, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 2.0520451068878174, | |
| "eval_runtime": 207.9034, | |
| "eval_samples_per_second": 30.177, | |
| "eval_steps_per_second": 3.776, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0004, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 2.0399410724639893, | |
| "eval_runtime": 207.2867, | |
| "eval_samples_per_second": 30.267, | |
| "eval_steps_per_second": 3.787, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9977, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 2.033215284347534, | |
| "eval_runtime": 207.459, | |
| "eval_samples_per_second": 30.242, | |
| "eval_steps_per_second": 3.784, | |
| "step": 3600 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 1.0503452935471104e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |