| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 192, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.1568627450980392, |
| "grad_norm": 16.678112030029297, |
| "learning_rate": 4.5e-06, |
| "loss": 0.408135986328125, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.3137254901960784, |
| "grad_norm": 8.07027816772461, |
| "learning_rate": 9.5e-06, |
| "loss": 0.23105964660644532, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.47058823529411764, |
| "grad_norm": 3.686859130859375, |
| "learning_rate": 9.932595378142233e-06, |
| "loss": 0.18245965242385864, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.6274509803921569, |
| "grad_norm": 3.441028356552124, |
| "learning_rate": 9.701923994199784e-06, |
| "loss": 0.1605141878128052, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.7843137254901961, |
| "grad_norm": 3.1651456356048584, |
| "learning_rate": 9.314825963771724e-06, |
| "loss": 0.158633816242218, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.9411764705882353, |
| "grad_norm": 2.6688830852508545, |
| "learning_rate": 8.784179514331683e-06, |
| "loss": 0.15139737129211425, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.0941176470588236, |
| "grad_norm": 2.1281120777130127, |
| "learning_rate": 8.127638534938227e-06, |
| "loss": 0.1387779712677002, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.2509803921568627, |
| "grad_norm": 4.023155689239502, |
| "learning_rate": 7.367045255243217e-06, |
| "loss": 0.12184641361236573, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.4078431372549018, |
| "grad_norm": 3.4635119438171387, |
| "learning_rate": 6.527703584078219e-06, |
| "loss": 0.11027407646179199, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.5647058823529412, |
| "grad_norm": 5.118541717529297, |
| "learning_rate": 5.63753728266054e-06, |
| "loss": 0.11028397083282471, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.7215686274509805, |
| "grad_norm": 2.325312852859497, |
| "learning_rate": 4.726160978856782e-06, |
| "loss": 0.10918575525283813, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.8784313725490196, |
| "grad_norm": 3.4077281951904297, |
| "learning_rate": 3.823894928601822e-06, |
| "loss": 0.0961656391620636, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.0313725490196077, |
| "grad_norm": 2.6073880195617676, |
| "learning_rate": 2.9607563020271446e-06, |
| "loss": 0.08220374584197998, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.1882352941176473, |
| "grad_norm": 2.4839987754821777, |
| "learning_rate": 2.1654605528435774e-06, |
| "loss": 0.05620207190513611, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.3450980392156864, |
| "grad_norm": 3.4678547382354736, |
| "learning_rate": 1.4644660940672628e-06, |
| "loss": 0.06362113952636719, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.5019607843137255, |
| "grad_norm": 3.457308769226074, |
| "learning_rate": 8.810940624342784e-07, |
| "loss": 0.063688462972641, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.6588235294117646, |
| "grad_norm": 2.5227317810058594, |
| "learning_rate": 4.347524557502919e-07, |
| "loss": 0.05568122863769531, |
| "step": 170 |
| }, |
| { |
| "epoch": 2.8156862745098037, |
| "grad_norm": 2.379140853881836, |
| "learning_rate": 1.4029045507537696e-07, |
| "loss": 0.05975309610366821, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.9725490196078432, |
| "grad_norm": 3.0802645683288574, |
| "learning_rate": 7.5044125708712e-09, |
| "loss": 0.052491509914398195, |
| "step": 190 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 192, |
| "total_flos": 2.653029551570944e+16, |
| "train_loss": 0.1262091943838944, |
| "train_runtime": 2059.2539, |
| "train_samples_per_second": 5.944, |
| "train_steps_per_second": 0.093 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 192, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 2000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.653029551570944e+16, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
