| { | |
| "best_metric": 1.2222630977630615, | |
| "best_model_checkpoint": "./outputs/checkpoint-4000", | |
| "epoch": 2.914754098360656, | |
| "eval_steps": 100, | |
| "global_step": 4000, | |
| "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": 2.1822, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.0121946334838867, | |
| "eval_runtime": 144.2461, | |
| "eval_samples_per_second": 43.495, | |
| "eval_steps_per_second": 5.442, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9624, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 1.9195032119750977, | |
| "eval_runtime": 144.1813, | |
| "eval_samples_per_second": 43.515, | |
| "eval_steps_per_second": 5.445, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8883, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 1.8597155809402466, | |
| "eval_runtime": 144.1877, | |
| "eval_samples_per_second": 43.513, | |
| "eval_steps_per_second": 5.444, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8371, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.8130855560302734, | |
| "eval_runtime": 144.1652, | |
| "eval_samples_per_second": 43.52, | |
| "eval_steps_per_second": 5.445, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7855, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.772993564605713, | |
| "eval_runtime": 144.0971, | |
| "eval_samples_per_second": 43.54, | |
| "eval_steps_per_second": 5.448, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7573, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.7435261011123657, | |
| "eval_runtime": 144.1427, | |
| "eval_samples_per_second": 43.526, | |
| "eval_steps_per_second": 5.446, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7188, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.7126474380493164, | |
| "eval_runtime": 144.1825, | |
| "eval_samples_per_second": 43.514, | |
| "eval_steps_per_second": 5.444, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6978, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.6834746599197388, | |
| "eval_runtime": 144.2172, | |
| "eval_samples_per_second": 43.504, | |
| "eval_steps_per_second": 5.443, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6627, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 1.6580406427383423, | |
| "eval_runtime": 144.1973, | |
| "eval_samples_per_second": 43.51, | |
| "eval_steps_per_second": 5.444, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6478, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 1.6355735063552856, | |
| "eval_runtime": 144.1901, | |
| "eval_samples_per_second": 43.512, | |
| "eval_steps_per_second": 5.444, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6278, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.6143633127212524, | |
| "eval_runtime": 144.1411, | |
| "eval_samples_per_second": 43.527, | |
| "eval_steps_per_second": 5.446, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5927, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 1.5936414003372192, | |
| "eval_runtime": 144.3064, | |
| "eval_samples_per_second": 43.477, | |
| "eval_steps_per_second": 5.44, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6005, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 1.573447346687317, | |
| "eval_runtime": 144.216, | |
| "eval_samples_per_second": 43.504, | |
| "eval_steps_per_second": 5.443, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5531, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.5534614324569702, | |
| "eval_runtime": 144.2247, | |
| "eval_samples_per_second": 43.502, | |
| "eval_steps_per_second": 5.443, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 1.525, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 1.5355420112609863, | |
| "eval_runtime": 144.287, | |
| "eval_samples_per_second": 43.483, | |
| "eval_steps_per_second": 5.441, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5066, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 1.5180158615112305, | |
| "eval_runtime": 144.1973, | |
| "eval_samples_per_second": 43.51, | |
| "eval_steps_per_second": 5.444, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5049, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 1.5018609762191772, | |
| "eval_runtime": 144.11, | |
| "eval_samples_per_second": 43.536, | |
| "eval_steps_per_second": 5.447, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4781, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 1.4871549606323242, | |
| "eval_runtime": 148.0177, | |
| "eval_samples_per_second": 42.387, | |
| "eval_steps_per_second": 5.303, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4858, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 1.498619794845581, | |
| "eval_runtime": 133.9486, | |
| "eval_samples_per_second": 46.839, | |
| "eval_steps_per_second": 5.86, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4665, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 1.475897192955017, | |
| "eval_runtime": 133.9594, | |
| "eval_samples_per_second": 46.835, | |
| "eval_steps_per_second": 5.86, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4389, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 1.458662509918213, | |
| "eval_runtime": 136.2607, | |
| "eval_samples_per_second": 46.044, | |
| "eval_steps_per_second": 5.761, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4404, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.4420702457427979, | |
| "eval_runtime": 133.798, | |
| "eval_samples_per_second": 46.892, | |
| "eval_steps_per_second": 5.867, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4162, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 1.4285529851913452, | |
| "eval_runtime": 133.9432, | |
| "eval_samples_per_second": 46.841, | |
| "eval_steps_per_second": 5.861, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4165, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 1.4121776819229126, | |
| "eval_runtime": 133.8876, | |
| "eval_samples_per_second": 46.86, | |
| "eval_steps_per_second": 5.863, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3912, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 1.3997336626052856, | |
| "eval_runtime": 133.7984, | |
| "eval_samples_per_second": 46.891, | |
| "eval_steps_per_second": 5.867, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3742, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 1.3850334882736206, | |
| "eval_runtime": 133.6811, | |
| "eval_samples_per_second": 46.933, | |
| "eval_steps_per_second": 5.872, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3701, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 1.3720322847366333, | |
| "eval_runtime": 133.6791, | |
| "eval_samples_per_second": 46.933, | |
| "eval_steps_per_second": 5.872, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3339, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 1.3605430126190186, | |
| "eval_runtime": 133.753, | |
| "eval_samples_per_second": 46.907, | |
| "eval_steps_per_second": 5.869, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3132, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 1.3472238779067993, | |
| "eval_runtime": 133.6686, | |
| "eval_samples_per_second": 46.937, | |
| "eval_steps_per_second": 5.873, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3122, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 1.3349665403366089, | |
| "eval_runtime": 133.7348, | |
| "eval_samples_per_second": 46.914, | |
| "eval_steps_per_second": 5.87, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2997, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 1.3228119611740112, | |
| "eval_runtime": 133.8158, | |
| "eval_samples_per_second": 46.885, | |
| "eval_steps_per_second": 5.866, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2979, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 1.3118646144866943, | |
| "eval_runtime": 133.8382, | |
| "eval_samples_per_second": 46.878, | |
| "eval_steps_per_second": 5.865, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2821, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 1.2987654209136963, | |
| "eval_runtime": 133.6905, | |
| "eval_samples_per_second": 46.929, | |
| "eval_steps_per_second": 5.872, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2616, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 1.288960576057434, | |
| "eval_runtime": 133.7641, | |
| "eval_samples_per_second": 46.903, | |
| "eval_steps_per_second": 5.869, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2611, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 1.278650164604187, | |
| "eval_runtime": 133.8143, | |
| "eval_samples_per_second": 46.886, | |
| "eval_steps_per_second": 5.866, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2484, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 1.265230655670166, | |
| "eval_runtime": 133.7663, | |
| "eval_samples_per_second": 46.903, | |
| "eval_steps_per_second": 5.868, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2333, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 1.2536190748214722, | |
| "eval_runtime": 133.8137, | |
| "eval_samples_per_second": 46.886, | |
| "eval_steps_per_second": 5.866, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2116, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 1.245966911315918, | |
| "eval_runtime": 133.7977, | |
| "eval_samples_per_second": 46.892, | |
| "eval_steps_per_second": 5.867, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2198, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 1.2307173013687134, | |
| "eval_runtime": 133.8147, | |
| "eval_samples_per_second": 46.886, | |
| "eval_steps_per_second": 5.866, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2025, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_loss": 1.2222630977630615, | |
| "eval_runtime": 133.8615, | |
| "eval_samples_per_second": 46.869, | |
| "eval_steps_per_second": 5.864, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 2.3841325139747635e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |