| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.64, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.8396336436271667, | |
| "learning_rate": 0.00019987329060020616, | |
| "loss": 2.8923, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.7428969740867615, | |
| "learning_rate": 0.00019949348350626456, | |
| "loss": 2.0944, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.5937550067901611, | |
| "learning_rate": 0.00019886154122075343, | |
| "loss": 2.0323, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.4460415542125702, | |
| "learning_rate": 0.00019797906520422677, | |
| "loss": 1.9436, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.4390580952167511, | |
| "learning_rate": 0.00019684829181681234, | |
| "loss": 2.0292, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.43469828367233276, | |
| "learning_rate": 0.00019547208665085457, | |
| "loss": 1.9654, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.48699721693992615, | |
| "learning_rate": 0.0001938539372689649, | |
| "loss": 1.9041, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.5205313563346863, | |
| "learning_rate": 0.00019199794436588243, | |
| "loss": 1.8992, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.5224802494049072, | |
| "learning_rate": 0.00018990881137654258, | |
| "loss": 1.8393, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.5004997849464417, | |
| "learning_rate": 0.0001875918325566888, | |
| "loss": 1.615, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.5422343611717224, | |
| "learning_rate": 0.00018505287956623297, | |
| "loss": 1.9873, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.47483357787132263, | |
| "learning_rate": 0.00018229838658936564, | |
| "loss": 1.9264, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.4191422164440155, | |
| "learning_rate": 0.00017933533402912354, | |
| "loss": 1.828, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.41457584500312805, | |
| "learning_rate": 0.00017617123081773591, | |
| "loss": 1.8083, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.524440348148346, | |
| "learning_rate": 0.00017281409538757883, | |
| "loss": 1.8268, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.5396173000335693, | |
| "learning_rate": 0.00016927243535095997, | |
| "loss": 1.7587, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.5633752942085266, | |
| "learning_rate": 0.0001655552259402295, | |
| "loss": 1.7939, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.5931205153465271, | |
| "learning_rate": 0.00016167188726285434, | |
| "loss": 1.848, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.677586019039154, | |
| "learning_rate": 0.00015763226042909455, | |
| "loss": 1.7407, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.5012004971504211, | |
| "learning_rate": 0.0001534465826127801, | |
| "loss": 1.8559, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.5996416807174683, | |
| "learning_rate": 0.00014912546110838775, | |
| "loss": 1.8363, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.5592181086540222, | |
| "learning_rate": 0.00014467984645016258, | |
| "loss": 1.6798, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.6498942971229553, | |
| "learning_rate": 0.00014012100466140578, | |
| "loss": 1.6868, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.5339984893798828, | |
| "learning_rate": 0.00013546048870425356, | |
| "loss": 1.8213, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.5681377649307251, | |
| "learning_rate": 0.00013071010920229909, | |
| "loss": 1.7561, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.5253719687461853, | |
| "learning_rate": 0.00012588190451025207, | |
| "loss": 1.7834, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.5420334935188293, | |
| "learning_rate": 0.00012098811020648475, | |
| "loss": 1.9342, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.6371601819992065, | |
| "learning_rate": 0.00011604112808577603, | |
| "loss": 1.8081, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.5578604340553284, | |
| "learning_rate": 0.000111053494730832, | |
| "loss": 1.7449, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.768509030342102, | |
| "learning_rate": 0.00010603784974222861, | |
| "loss": 1.8202, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.627295970916748, | |
| "learning_rate": 0.00010100690370728755, | |
| "loss": 1.7587, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.6519939303398132, | |
| "learning_rate": 9.597340598905852e-05, | |
| "loss": 1.8827, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.6148818135261536, | |
| "learning_rate": 9.095011241703623e-05, | |
| "loss": 1.8022, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.5009266138076782, | |
| "learning_rate": 8.594975296149076e-05, | |
| "loss": 1.7515, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.5753130316734314, | |
| "learning_rate": 8.098499947332934e-05, | |
| "loss": 1.7027, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.5501881241798401, | |
| "learning_rate": 7.606843357124426e-05, | |
| "loss": 1.7445, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.6279743909835815, | |
| "learning_rate": 7.121251475752539e-05, | |
| "loss": 1.7094, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.5455606579780579, | |
| "learning_rate": 6.642954884333955e-05, | |
| "loss": 1.7047, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.5445593595504761, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 1.6086, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.6656392812728882, | |
| "learning_rate": 5.713074385969457e-05, | |
| "loss": 1.7098, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 312, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 9.200724451688448e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |