| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 669, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.07, |
| "learning_rate": 1.1194029850746268e-07, |
| "loss": 2.3842, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.13, |
| "learning_rate": 2.2388059701492537e-07, |
| "loss": 2.1216, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.2, |
| "learning_rate": 3.3582089552238805e-07, |
| "loss": 1.209, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.27, |
| "learning_rate": 4.4776119402985074e-07, |
| "loss": 0.3572, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.34, |
| "learning_rate": 4.997821619858613e-07, |
| "loss": 0.2376, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.4, |
| "learning_rate": 4.982013320674148e-07, |
| "loss": 0.2259, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.47, |
| "learning_rate": 4.951004040601898e-07, |
| "loss": 0.2179, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.54, |
| "learning_rate": 4.904983694607182e-07, |
| "loss": 0.2312, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.61, |
| "learning_rate": 4.844234132265139e-07, |
| "loss": 0.2625, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.67, |
| "learning_rate": 4.769127411585451e-07, |
| "loss": 0.2149, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.74, |
| "learning_rate": 4.680123520359519e-07, |
| "loss": 0.2285, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.81, |
| "learning_rate": 4.577767558985552e-07, |
| "loss": 0.1952, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.87, |
| "learning_rate": 4.4626864020252767e-07, |
| "loss": 0.216, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.94, |
| "learning_rate": 4.3355848589383875e-07, |
| "loss": 0.2209, |
| "step": 210 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.1885647028684616, |
| "eval_runtime": 84.9949, |
| "eval_samples_per_second": 4.424, |
| "eval_steps_per_second": 0.553, |
| "step": 223 |
| }, |
| { |
| "epoch": 1.01, |
| "learning_rate": 4.197241357508159e-07, |
| "loss": 0.2079, |
| "step": 225 |
| }, |
| { |
| "epoch": 1.08, |
| "learning_rate": 4.048503176394893e-07, |
| "loss": 0.1639, |
| "step": 240 |
| }, |
| { |
| "epoch": 1.14, |
| "learning_rate": 3.8902812560152067e-07, |
| "loss": 0.2003, |
| "step": 255 |
| }, |
| { |
| "epoch": 1.21, |
| "learning_rate": 3.7235446195277136e-07, |
| "loss": 0.2023, |
| "step": 270 |
| }, |
| { |
| "epoch": 1.28, |
| "learning_rate": 3.5493144380935153e-07, |
| "loss": 0.2093, |
| "step": 285 |
| }, |
| { |
| "epoch": 1.35, |
| "learning_rate": 3.3686577767585543e-07, |
| "loss": 0.1862, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.41, |
| "learning_rate": 3.182681059260903e-07, |
| "loss": 0.2173, |
| "step": 315 |
| }, |
| { |
| "epoch": 1.48, |
| "learning_rate": 2.992523291787476e-07, |
| "loss": 0.1854, |
| "step": 330 |
| }, |
| { |
| "epoch": 1.55, |
| "learning_rate": 2.79934908718098e-07, |
| "loss": 0.1846, |
| "step": 345 |
| }, |
| { |
| "epoch": 1.61, |
| "learning_rate": 2.604341532320033e-07, |
| "loss": 0.2093, |
| "step": 360 |
| }, |
| { |
| "epoch": 1.68, |
| "learning_rate": 2.4086949423558525e-07, |
| "loss": 0.215, |
| "step": 375 |
| }, |
| { |
| "epoch": 1.75, |
| "learning_rate": 2.21360754618191e-07, |
| "loss": 0.2203, |
| "step": 390 |
| }, |
| { |
| "epoch": 1.82, |
| "learning_rate": 2.020274147934019e-07, |
| "loss": 0.2012, |
| "step": 405 |
| }, |
| { |
| "epoch": 1.88, |
| "learning_rate": 1.8298788094652156e-07, |
| "loss": 0.1801, |
| "step": 420 |
| }, |
| { |
| "epoch": 1.95, |
| "learning_rate": 1.6435875986112683e-07, |
| "loss": 0.232, |
| "step": 435 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.1808764636516571, |
| "eval_runtime": 84.9305, |
| "eval_samples_per_second": 4.427, |
| "eval_steps_per_second": 0.553, |
| "step": 446 |
| }, |
| { |
| "epoch": 2.02, |
| "learning_rate": 1.4625414476597541e-07, |
| "loss": 0.1974, |
| "step": 450 |
| }, |
| { |
| "epoch": 2.09, |
| "learning_rate": 1.287849165760687e-07, |
| "loss": 0.2059, |
| "step": 465 |
| }, |
| { |
| "epoch": 2.15, |
| "learning_rate": 1.120580648073885e-07, |
| "loss": 0.1932, |
| "step": 480 |
| }, |
| { |
| "epoch": 2.22, |
| "learning_rate": 9.617603232433475e-08, |
| "loss": 0.1538, |
| "step": 495 |
| }, |
| { |
| "epoch": 2.29, |
| "learning_rate": 8.123608793292986e-08, |
| "loss": 0.2281, |
| "step": 510 |
| }, |
| { |
| "epoch": 2.35, |
| "learning_rate": 6.732973066231562e-08, |
| "loss": 0.1934, |
| "step": 525 |
| }, |
| { |
| "epoch": 2.42, |
| "learning_rate": 5.454212938299255e-08, |
| "loss": 0.2006, |
| "step": 540 |
| }, |
| { |
| "epoch": 2.49, |
| "learning_rate": 4.295160119383712e-08, |
| "loss": 0.1785, |
| "step": 555 |
| }, |
| { |
| "epoch": 2.56, |
| "learning_rate": 3.2629131772490434e-08, |
| "loss": 0.1655, |
| "step": 570 |
| }, |
| { |
| "epoch": 2.62, |
| "learning_rate": 2.3637940626713342e-08, |
| "loss": 0.1975, |
| "step": 585 |
| }, |
| { |
| "epoch": 2.69, |
| "learning_rate": 1.6033093909304853e-08, |
| "loss": 0.1877, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.76, |
| "learning_rate": 9.861167167883044e-09, |
| "loss": 0.1972, |
| "step": 615 |
| }, |
| { |
| "epoch": 2.83, |
| "learning_rate": 5.1599600950022195e-09, |
| "loss": 0.2058, |
| "step": 630 |
| }, |
| { |
| "epoch": 2.89, |
| "learning_rate": 1.9582650256064203e-09, |
| "loss": 0.1942, |
| "step": 645 |
| }, |
| { |
| "epoch": 2.96, |
| "learning_rate": 2.75690599646522e-10, |
| "loss": 0.1667, |
| "step": 660 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.1806206852197647, |
| "eval_runtime": 84.9152, |
| "eval_samples_per_second": 4.428, |
| "eval_steps_per_second": 0.553, |
| "step": 669 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 669, |
| "total_flos": 331458329182208.0, |
| "train_loss": 0.3205718675180045, |
| "train_runtime": 57740.8386, |
| "train_samples_per_second": 0.37, |
| "train_steps_per_second": 0.012 |
| } |
| ], |
| "logging_steps": 15, |
| "max_steps": 669, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "total_flos": 331458329182208.0, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
