| { | |
| "best_metric": 1.3689783811569214, | |
| "best_model_checkpoint": "./outputs/checkpoint-2000", | |
| "epoch": 2.6666666666666665, | |
| "eval_steps": 100, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4574, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 1.5645889043807983, | |
| "eval_runtime": 93.6488, | |
| "eval_samples_per_second": 15.494, | |
| "eval_steps_per_second": 1.943, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2971, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 1.5321934223175049, | |
| "eval_runtime": 93.3383, | |
| "eval_samples_per_second": 15.546, | |
| "eval_steps_per_second": 1.95, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2756, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 1.5133259296417236, | |
| "eval_runtime": 93.2652, | |
| "eval_samples_per_second": 15.558, | |
| "eval_steps_per_second": 1.951, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2661, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 1.500829815864563, | |
| "eval_runtime": 93.3122, | |
| "eval_samples_per_second": 15.55, | |
| "eval_steps_per_second": 1.95, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2493, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 1.4804445505142212, | |
| "eval_runtime": 93.3662, | |
| "eval_samples_per_second": 15.541, | |
| "eval_steps_per_second": 1.949, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2327, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.4673335552215576, | |
| "eval_runtime": 93.4173, | |
| "eval_samples_per_second": 15.532, | |
| "eval_steps_per_second": 1.948, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2216, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 1.456375241279602, | |
| "eval_runtime": 93.4236, | |
| "eval_samples_per_second": 15.531, | |
| "eval_steps_per_second": 1.948, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2086, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 1.4467251300811768, | |
| "eval_runtime": 93.3142, | |
| "eval_samples_per_second": 15.55, | |
| "eval_steps_per_second": 1.95, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1922, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 1.4370195865631104, | |
| "eval_runtime": 93.2407, | |
| "eval_samples_per_second": 15.562, | |
| "eval_steps_per_second": 1.952, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1859, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 1.4297349452972412, | |
| "eval_runtime": 93.4112, | |
| "eval_samples_per_second": 15.533, | |
| "eval_steps_per_second": 1.948, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1846, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_loss": 1.4200090169906616, | |
| "eval_runtime": 93.3115, | |
| "eval_samples_per_second": 15.55, | |
| "eval_steps_per_second": 1.95, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1697, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.4181303977966309, | |
| "eval_runtime": 93.16, | |
| "eval_samples_per_second": 15.575, | |
| "eval_steps_per_second": 1.954, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1844, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_loss": 1.407381296157837, | |
| "eval_runtime": 94.8759, | |
| "eval_samples_per_second": 15.294, | |
| "eval_steps_per_second": 1.918, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1693, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_loss": 1.3997007608413696, | |
| "eval_runtime": 93.3773, | |
| "eval_samples_per_second": 15.539, | |
| "eval_steps_per_second": 1.949, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1695, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.3967318534851074, | |
| "eval_runtime": 93.2877, | |
| "eval_samples_per_second": 15.554, | |
| "eval_steps_per_second": 1.951, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1462, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_loss": 1.3913403749465942, | |
| "eval_runtime": 93.297, | |
| "eval_samples_per_second": 15.552, | |
| "eval_steps_per_second": 1.951, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1354, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_loss": 1.3872549533843994, | |
| "eval_runtime": 93.3779, | |
| "eval_samples_per_second": 15.539, | |
| "eval_steps_per_second": 1.949, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1306, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 1.3809529542922974, | |
| "eval_runtime": 93.2845, | |
| "eval_samples_per_second": 15.555, | |
| "eval_steps_per_second": 1.951, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1369, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 1.3753933906555176, | |
| "eval_runtime": 93.318, | |
| "eval_samples_per_second": 15.549, | |
| "eval_steps_per_second": 1.95, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1357, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 1.3689783811569214, | |
| "eval_runtime": 93.3265, | |
| "eval_samples_per_second": 15.548, | |
| "eval_steps_per_second": 1.95, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 1.1694767356919808e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |