| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9296875, | |
| "eval_steps": 25, | |
| "global_step": 750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.43993993993994e-05, | |
| "loss": 1.1099, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.0085899829864502, | |
| "eval_runtime": 168.9968, | |
| "eval_samples_per_second": 1.456, | |
| "eval_steps_per_second": 0.183, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.3773773773773775e-05, | |
| "loss": 0.966, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.955312192440033, | |
| "eval_runtime": 169.2723, | |
| "eval_samples_per_second": 1.453, | |
| "eval_steps_per_second": 0.183, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.9251, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 0.9321236610412598, | |
| "eval_runtime": 169.3712, | |
| "eval_samples_per_second": 1.452, | |
| "eval_steps_per_second": 0.183, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 2.2547547547547548e-05, | |
| "loss": 0.9126, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_loss": 0.9122704267501831, | |
| "eval_runtime": 169.5064, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 2.1996996996997e-05, | |
| "loss": 0.884, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_loss": 0.9045689105987549, | |
| "eval_runtime": 169.4144, | |
| "eval_samples_per_second": 1.452, | |
| "eval_steps_per_second": 0.183, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.142142142142142e-05, | |
| "loss": 0.9249, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_loss": 0.9013924598693848, | |
| "eval_runtime": 169.5232, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 2.0845845845845847e-05, | |
| "loss": 0.8931, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 0.9005089402198792, | |
| "eval_runtime": 169.4928, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 2.0245245245245244e-05, | |
| "loss": 0.8724, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_loss": 0.8974775671958923, | |
| "eval_runtime": 169.6039, | |
| "eval_samples_per_second": 1.45, | |
| "eval_steps_per_second": 0.183, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.961961961961962e-05, | |
| "loss": 0.8692, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_loss": 0.8939986824989319, | |
| "eval_runtime": 169.6629, | |
| "eval_samples_per_second": 1.45, | |
| "eval_steps_per_second": 0.183, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.9019019019019018e-05, | |
| "loss": 0.8791, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_loss": 0.8978664875030518, | |
| "eval_runtime": 169.6122, | |
| "eval_samples_per_second": 1.45, | |
| "eval_steps_per_second": 0.183, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.8443443443443443e-05, | |
| "loss": 0.8534, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 0.8919960856437683, | |
| "eval_runtime": 169.6602, | |
| "eval_samples_per_second": 1.45, | |
| "eval_steps_per_second": 0.183, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.781781781781782e-05, | |
| "loss": 0.8173, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 0.8853669166564941, | |
| "eval_runtime": 169.6012, | |
| "eval_samples_per_second": 1.45, | |
| "eval_steps_per_second": 0.183, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.7192192192192195e-05, | |
| "loss": 0.7995, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 0.8796117305755615, | |
| "eval_runtime": 169.6226, | |
| "eval_samples_per_second": 1.45, | |
| "eval_steps_per_second": 0.183, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.6566566566566568e-05, | |
| "loss": 0.8031, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_loss": 0.8701273798942566, | |
| "eval_runtime": 169.668, | |
| "eval_samples_per_second": 1.45, | |
| "eval_steps_per_second": 0.183, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 1.594094094094094e-05, | |
| "loss": 0.7914, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 0.8678767681121826, | |
| "eval_runtime": 169.5954, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 1.5315315315315316e-05, | |
| "loss": 0.7792, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_loss": 0.8583410382270813, | |
| "eval_runtime": 169.6512, | |
| "eval_samples_per_second": 1.45, | |
| "eval_steps_per_second": 0.183, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 1.468968968968969e-05, | |
| "loss": 0.7739, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_loss": 0.8527249693870544, | |
| "eval_runtime": 169.6094, | |
| "eval_samples_per_second": 1.45, | |
| "eval_steps_per_second": 0.183, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 1.4064064064064064e-05, | |
| "loss": 0.7829, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_loss": 0.8476611971855164, | |
| "eval_runtime": 169.7218, | |
| "eval_samples_per_second": 1.449, | |
| "eval_steps_per_second": 0.183, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.3438438438438438e-05, | |
| "loss": 0.7881, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_loss": 0.8437426686286926, | |
| "eval_runtime": 169.563, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.2812812812812813e-05, | |
| "loss": 0.791, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_loss": 0.8393863439559937, | |
| "eval_runtime": 169.5296, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.2187187187187189e-05, | |
| "loss": 0.6864, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_loss": 0.863290548324585, | |
| "eval_runtime": 169.551, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 1.1561561561561563e-05, | |
| "loss": 0.6342, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "eval_loss": 0.8623055219650269, | |
| "eval_runtime": 169.5813, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.0935935935935937e-05, | |
| "loss": 0.6208, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_loss": 0.8574148416519165, | |
| "eval_runtime": 169.5698, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 1.031031031031031e-05, | |
| "loss": 0.5885, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "eval_loss": 0.8582069873809814, | |
| "eval_runtime": 169.5406, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 9.684684684684685e-06, | |
| "loss": 0.607, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "eval_loss": 0.8599240779876709, | |
| "eval_runtime": 169.6355, | |
| "eval_samples_per_second": 1.45, | |
| "eval_steps_per_second": 0.183, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 9.05905905905906e-06, | |
| "loss": 0.6149, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "eval_loss": 0.8572269678115845, | |
| "eval_runtime": 169.5236, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 8.433433433433434e-06, | |
| "loss": 0.5893, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "eval_loss": 0.8562537431716919, | |
| "eval_runtime": 169.5361, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 7.807807807807808e-06, | |
| "loss": 0.6176, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "eval_loss": 0.8548978567123413, | |
| "eval_runtime": 169.5309, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 7.182182182182183e-06, | |
| "loss": 0.5953, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_loss": 0.8483610153198242, | |
| "eval_runtime": 169.5898, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 6.556556556556556e-06, | |
| "loss": 0.6088, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_loss": 0.8442074060440063, | |
| "eval_runtime": 169.5261, | |
| "eval_samples_per_second": 1.451, | |
| "eval_steps_per_second": 0.183, | |
| "step": 750 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 25, | |
| "total_flos": 6.766615596367872e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |