| { | |
| "best_metric": 1.8560636043548584, | |
| "best_model_checkpoint": "/content/models/TinyLlama-TinyLlama-1-1B-Chat-v1-0/checkpoint-564", | |
| "epoch": 5.0, | |
| "eval_steps": 141, | |
| "global_step": 2810, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 5.238644123077393, | |
| "learning_rate": 4.7491103202846974e-05, | |
| "loss": 2.103, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 1.9888125658035278, | |
| "eval_runtime": 4.1833, | |
| "eval_samples_per_second": 56.654, | |
| "eval_steps_per_second": 7.171, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 4.7757039070129395, | |
| "learning_rate": 4.498220640569395e-05, | |
| "loss": 1.908, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 1.9252748489379883, | |
| "eval_runtime": 4.1854, | |
| "eval_samples_per_second": 56.625, | |
| "eval_steps_per_second": 7.168, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 3.813783645629883, | |
| "learning_rate": 4.247330960854093e-05, | |
| "loss": 1.8483, | |
| "step": 423 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 1.9029674530029297, | |
| "eval_runtime": 4.1786, | |
| "eval_samples_per_second": 56.717, | |
| "eval_steps_per_second": 7.179, | |
| "step": 423 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.197076797485352, | |
| "learning_rate": 3.99644128113879e-05, | |
| "loss": 1.8007, | |
| "step": 564 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.8560636043548584, | |
| "eval_runtime": 4.1817, | |
| "eval_samples_per_second": 56.675, | |
| "eval_steps_per_second": 7.174, | |
| "step": 564 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 5.801040172576904, | |
| "learning_rate": 3.7455516014234874e-05, | |
| "loss": 1.4215, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 1.9399982690811157, | |
| "eval_runtime": 4.1763, | |
| "eval_samples_per_second": 56.749, | |
| "eval_steps_per_second": 7.183, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 3.8744606971740723, | |
| "learning_rate": 3.494661921708185e-05, | |
| "loss": 1.4322, | |
| "step": 846 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_loss": 1.8941714763641357, | |
| "eval_runtime": 4.1838, | |
| "eval_samples_per_second": 56.648, | |
| "eval_steps_per_second": 7.171, | |
| "step": 846 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 5.074963092803955, | |
| "learning_rate": 3.243772241992883e-05, | |
| "loss": 1.4325, | |
| "step": 987 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_loss": 1.8861587047576904, | |
| "eval_runtime": 4.1802, | |
| "eval_samples_per_second": 56.696, | |
| "eval_steps_per_second": 7.177, | |
| "step": 987 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 4.776240348815918, | |
| "learning_rate": 2.9928825622775803e-05, | |
| "loss": 1.4241, | |
| "step": 1128 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 1.9016921520233154, | |
| "eval_runtime": 4.1862, | |
| "eval_samples_per_second": 56.614, | |
| "eval_steps_per_second": 7.166, | |
| "step": 1128 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 6.694357395172119, | |
| "learning_rate": 2.7419928825622775e-05, | |
| "loss": 0.7606, | |
| "step": 1269 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 2.2999789714813232, | |
| "eval_runtime": 4.1771, | |
| "eval_samples_per_second": 56.738, | |
| "eval_steps_per_second": 7.182, | |
| "step": 1269 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 24.188169479370117, | |
| "learning_rate": 2.4911032028469753e-05, | |
| "loss": 0.7631, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_loss": 2.3000378608703613, | |
| "eval_runtime": 4.1828, | |
| "eval_samples_per_second": 56.661, | |
| "eval_steps_per_second": 7.172, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 5.172097206115723, | |
| "learning_rate": 2.2402135231316725e-05, | |
| "loss": 0.7264, | |
| "step": 1551 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_loss": 2.3029186725616455, | |
| "eval_runtime": 4.1769, | |
| "eval_samples_per_second": 56.741, | |
| "eval_steps_per_second": 7.182, | |
| "step": 1551 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 3.475619077682495, | |
| "learning_rate": 1.9893238434163703e-05, | |
| "loss": 0.7021, | |
| "step": 1692 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "eval_loss": 2.3999218940734863, | |
| "eval_runtime": 4.1882, | |
| "eval_samples_per_second": 56.587, | |
| "eval_steps_per_second": 7.163, | |
| "step": 1692 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "grad_norm": 4.135960102081299, | |
| "learning_rate": 1.7384341637010675e-05, | |
| "loss": 0.2238, | |
| "step": 1833 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "eval_loss": 2.7608110904693604, | |
| "eval_runtime": 4.1782, | |
| "eval_samples_per_second": 56.723, | |
| "eval_steps_per_second": 7.18, | |
| "step": 1833 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 3.5472917556762695, | |
| "learning_rate": 1.4875444839857652e-05, | |
| "loss": 0.3114, | |
| "step": 1974 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "eval_loss": 2.845644474029541, | |
| "eval_runtime": 4.1819, | |
| "eval_samples_per_second": 56.673, | |
| "eval_steps_per_second": 7.174, | |
| "step": 1974 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 4.326556205749512, | |
| "learning_rate": 1.2366548042704627e-05, | |
| "loss": 0.2282, | |
| "step": 2115 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "eval_loss": 2.8092262744903564, | |
| "eval_runtime": 4.1782, | |
| "eval_samples_per_second": 56.723, | |
| "eval_steps_per_second": 7.18, | |
| "step": 2115 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 2.737431049346924, | |
| "learning_rate": 9.857651245551602e-06, | |
| "loss": 0.2086, | |
| "step": 2256 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "eval_loss": 2.9099254608154297, | |
| "eval_runtime": 4.1802, | |
| "eval_samples_per_second": 56.696, | |
| "eval_steps_per_second": 7.177, | |
| "step": 2256 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "grad_norm": 3.1757376194000244, | |
| "learning_rate": 7.348754448398577e-06, | |
| "loss": 0.0954, | |
| "step": 2397 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "eval_loss": 3.1383473873138428, | |
| "eval_runtime": 4.1771, | |
| "eval_samples_per_second": 56.738, | |
| "eval_steps_per_second": 7.182, | |
| "step": 2397 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "grad_norm": 2.6402652263641357, | |
| "learning_rate": 4.839857651245551e-06, | |
| "loss": 0.1015, | |
| "step": 2538 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "eval_loss": 3.1500625610351562, | |
| "eval_runtime": 4.1764, | |
| "eval_samples_per_second": 56.747, | |
| "eval_steps_per_second": 7.183, | |
| "step": 2538 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "grad_norm": 3.826103687286377, | |
| "learning_rate": 2.330960854092527e-06, | |
| "loss": 0.0769, | |
| "step": 2679 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "eval_loss": 3.1649973392486572, | |
| "eval_runtime": 4.1784, | |
| "eval_samples_per_second": 56.72, | |
| "eval_steps_per_second": 7.18, | |
| "step": 2679 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 2810, | |
| "total_flos": 6693708929433600.0, | |
| "train_loss": 0.8862062440648197, | |
| "train_runtime": 1654.6028, | |
| "train_samples_per_second": 5.092, | |
| "train_steps_per_second": 1.698 | |
| } | |
| ], | |
| "logging_steps": 141, | |
| "max_steps": 2810, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 141, | |
| "total_flos": 6693708929433600.0, | |
| "train_batch_size": 3, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |