| { | |
| "best_metric": 1.8635917901992798, | |
| "best_model_checkpoint": "./outputs/checkpoint-2100", | |
| "epoch": 2.9829545454545454, | |
| "eval_steps": 100, | |
| "global_step": 2100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4609, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 2.123192071914673, | |
| "eval_runtime": 56.1237, | |
| "eval_samples_per_second": 26.584, | |
| "eval_steps_per_second": 3.332, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2635, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 2.083603858947754, | |
| "eval_runtime": 53.713, | |
| "eval_samples_per_second": 27.777, | |
| "eval_steps_per_second": 3.481, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2344, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_loss": 2.054030656814575, | |
| "eval_runtime": 53.5493, | |
| "eval_samples_per_second": 27.862, | |
| "eval_steps_per_second": 3.492, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2009, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 2.038727045059204, | |
| "eval_runtime": 53.5913, | |
| "eval_samples_per_second": 27.84, | |
| "eval_steps_per_second": 3.489, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1811, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 2.0150375366210938, | |
| "eval_runtime": 53.7081, | |
| "eval_samples_per_second": 27.78, | |
| "eval_steps_per_second": 3.482, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1648, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 1.9949842691421509, | |
| "eval_runtime": 53.6059, | |
| "eval_samples_per_second": 27.833, | |
| "eval_steps_per_second": 3.488, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1446, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_loss": 1.9850085973739624, | |
| "eval_runtime": 53.5892, | |
| "eval_samples_per_second": 27.841, | |
| "eval_steps_per_second": 3.49, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1122, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 1.9744948148727417, | |
| "eval_runtime": 53.6175, | |
| "eval_samples_per_second": 27.827, | |
| "eval_steps_per_second": 3.488, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0852, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": 1.9583721160888672, | |
| "eval_runtime": 53.7484, | |
| "eval_samples_per_second": 27.759, | |
| "eval_steps_per_second": 3.479, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0848, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_loss": 1.9483006000518799, | |
| "eval_runtime": 53.6139, | |
| "eval_samples_per_second": 27.829, | |
| "eval_steps_per_second": 3.488, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0002, | |
| "loss": 2.091, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_loss": 1.936788558959961, | |
| "eval_runtime": 53.6874, | |
| "eval_samples_per_second": 27.791, | |
| "eval_steps_per_second": 3.483, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0684, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 1.929431438446045, | |
| "eval_runtime": 53.7712, | |
| "eval_samples_per_second": 27.747, | |
| "eval_steps_per_second": 3.478, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0524, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_loss": 1.9179173707962036, | |
| "eval_runtime": 53.6365, | |
| "eval_samples_per_second": 27.817, | |
| "eval_steps_per_second": 3.486, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0562, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_loss": 1.9093526601791382, | |
| "eval_runtime": 53.6029, | |
| "eval_samples_per_second": 27.834, | |
| "eval_steps_per_second": 3.489, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0075, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_loss": 1.9054597616195679, | |
| "eval_runtime": 53.6616, | |
| "eval_samples_per_second": 27.804, | |
| "eval_steps_per_second": 3.485, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0119, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_loss": 1.8929505348205566, | |
| "eval_runtime": 53.7348, | |
| "eval_samples_per_second": 27.766, | |
| "eval_steps_per_second": 3.48, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9964, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "eval_loss": 1.8908874988555908, | |
| "eval_runtime": 53.6745, | |
| "eval_samples_per_second": 27.797, | |
| "eval_steps_per_second": 3.484, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9869, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_loss": 1.8806322813034058, | |
| "eval_runtime": 53.6914, | |
| "eval_samples_per_second": 27.788, | |
| "eval_steps_per_second": 3.483, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0004, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 1.877378225326538, | |
| "eval_runtime": 53.6054, | |
| "eval_samples_per_second": 27.833, | |
| "eval_steps_per_second": 3.488, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0002, | |
| "loss": 1.991, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 1.8717212677001953, | |
| "eval_runtime": 53.7679, | |
| "eval_samples_per_second": 27.749, | |
| "eval_steps_per_second": 3.478, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9898, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_loss": 1.8635917901992798, | |
| "eval_runtime": 53.6308, | |
| "eval_samples_per_second": 27.82, | |
| "eval_steps_per_second": 3.487, | |
| "step": 2100 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2112, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 6.361597443704832e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |