| { | |
| "best_metric": 2.6527161598205566, | |
| "best_model_checkpoint": "./results/checkpoint-2000", | |
| "epoch": 0.9691096305269533, | |
| "eval_steps": 250, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04845548152634767, | |
| "grad_norm": 0.2021484375, | |
| "learning_rate": 0.0019926230341909047, | |
| "loss": 5.5548, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.09691096305269534, | |
| "grad_norm": 0.484375, | |
| "learning_rate": 0.0019623193935821215, | |
| "loss": 4.0801, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.12113870381586916, | |
| "eval_loss": 3.651378631591797, | |
| "eval_runtime": 1225.4443, | |
| "eval_samples_per_second": 47.902, | |
| "eval_steps_per_second": 1.497, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.145366444579043, | |
| "grad_norm": 0.6796875, | |
| "learning_rate": 0.0019092830690827923, | |
| "loss": 3.6511, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.19382192610539067, | |
| "grad_norm": 0.474609375, | |
| "learning_rate": 0.0018347669274724923, | |
| "loss": 3.4945, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.24227740763173833, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 0.0017405312490731885, | |
| "loss": 3.3662, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.24227740763173833, | |
| "eval_loss": 3.2901480197906494, | |
| "eval_runtime": 1223.1792, | |
| "eval_samples_per_second": 47.991, | |
| "eval_steps_per_second": 1.5, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.290732889158086, | |
| "grad_norm": 0.90234375, | |
| "learning_rate": 0.0016288021449895208, | |
| "loss": 3.2248, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.3391883706844337, | |
| "grad_norm": 0.5859375, | |
| "learning_rate": 0.001502218970119089, | |
| "loss": 3.1488, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.3634161114476075, | |
| "eval_loss": 3.054584503173828, | |
| "eval_runtime": 1225.6755, | |
| "eval_samples_per_second": 47.893, | |
| "eval_steps_per_second": 1.497, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.38764385221078135, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 0.001363771974184993, | |
| "loss": 3.0595, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.436099333737129, | |
| "grad_norm": 0.498046875, | |
| "learning_rate": 0.001216731663648138, | |
| "loss": 2.9756, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.48455481526347666, | |
| "grad_norm": 0.423828125, | |
| "learning_rate": 0.0010645715431691336, | |
| "loss": 2.9173, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.48455481526347666, | |
| "eval_loss": 2.87882399559021, | |
| "eval_runtime": 1223.6, | |
| "eval_samples_per_second": 47.974, | |
| "eval_steps_per_second": 1.5, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5330102967898244, | |
| "grad_norm": 0.431640625, | |
| "learning_rate": 0.0009108860616831929, | |
| "loss": 2.853, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.581465778316172, | |
| "grad_norm": 0.546875, | |
| "learning_rate": 0.0007593057014319566, | |
| "loss": 2.7923, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.6056935190793459, | |
| "eval_loss": 2.7545697689056396, | |
| "eval_runtime": 1225.0225, | |
| "eval_samples_per_second": 47.918, | |
| "eval_steps_per_second": 1.498, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.6299212598425197, | |
| "grad_norm": 0.392578125, | |
| "learning_rate": 0.0006134112157875245, | |
| "loss": 2.7647, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.6783767413688674, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 0.00047664904181190284, | |
| "loss": 2.7262, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.726832222895215, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.0003522498857445232, | |
| "loss": 2.692, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.726832222895215, | |
| "eval_loss": 2.6806631088256836, | |
| "eval_runtime": 1223.5768, | |
| "eval_samples_per_second": 47.975, | |
| "eval_steps_per_second": 1.5, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7752877044215627, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 0.0002431524046570296, | |
| "loss": 2.6635, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.8237431859479104, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.00015193378712876149, | |
| "loss": 2.6561, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.8479709267110842, | |
| "eval_loss": 2.655076503753662, | |
| "eval_runtime": 1225.3417, | |
| "eval_samples_per_second": 47.906, | |
| "eval_steps_per_second": 1.498, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.872198667474258, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 8.074887282213439e-05, | |
| "loss": 2.6481, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.9206541490006057, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 3.127924912435132e-05, | |
| "loss": 2.6565, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.9691096305269533, | |
| "grad_norm": 0.416015625, | |
| "learning_rate": 4.693527335575154e-06, | |
| "loss": 2.6513, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9691096305269533, | |
| "eval_loss": 2.6527161598205566, | |
| "eval_runtime": 1223.4873, | |
| "eval_samples_per_second": 47.978, | |
| "eval_steps_per_second": 1.5, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2063, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 250, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.678589479911424e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |