| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 50, |
| "global_step": 1000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": false, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.05, |
| "eval_accuracy": 0.14088188976377952, |
| "eval_loss": 49.78975296020508, |
| "eval_runtime": 13.6275, |
| "eval_samples_per_second": 9.173, |
| "eval_steps_per_second": 1.174, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.1, |
| "grad_norm": 13452.2255859375, |
| "learning_rate": 0.000264, |
| "loss": 1505.6178125, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.1, |
| "eval_accuracy": 0.1445748031496063, |
| "eval_loss": 49.24732208251953, |
| "eval_runtime": 8.7677, |
| "eval_samples_per_second": 14.257, |
| "eval_steps_per_second": 1.825, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.15, |
| "eval_accuracy": 0.1491968503937008, |
| "eval_loss": 48.760948181152344, |
| "eval_runtime": 8.1271, |
| "eval_samples_per_second": 15.381, |
| "eval_steps_per_second": 1.969, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.2, |
| "grad_norm": 8017.09326171875, |
| "learning_rate": 0.0003967291041791484, |
| "loss": 1469.6371875, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.2, |
| "eval_accuracy": 0.15311023622047243, |
| "eval_loss": 48.39834213256836, |
| "eval_runtime": 8.3927, |
| "eval_samples_per_second": 14.894, |
| "eval_steps_per_second": 1.906, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.25, |
| "eval_accuracy": 0.15462204724409448, |
| "eval_loss": 48.1512336730957, |
| "eval_runtime": 10.9865, |
| "eval_samples_per_second": 11.378, |
| "eval_steps_per_second": 1.456, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.3, |
| "grad_norm": 9418.4111328125, |
| "learning_rate": 0.0003704314025795668, |
| "loss": 1449.61, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.3, |
| "eval_accuracy": 0.15623622047244096, |
| "eval_loss": 47.92970275878906, |
| "eval_runtime": 8.5559, |
| "eval_samples_per_second": 14.61, |
| "eval_steps_per_second": 1.87, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.35, |
| "eval_accuracy": 0.15805511811023623, |
| "eval_loss": 47.6953125, |
| "eval_runtime": 10.9375, |
| "eval_samples_per_second": 11.429, |
| "eval_steps_per_second": 1.463, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.4, |
| "grad_norm": 3742.375, |
| "learning_rate": 0.0003211159956686115, |
| "loss": 1437.1715625, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_accuracy": 0.16048031496062992, |
| "eval_loss": 47.554046630859375, |
| "eval_runtime": 8.2976, |
| "eval_samples_per_second": 15.065, |
| "eval_steps_per_second": 1.928, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.45, |
| "eval_accuracy": 0.16221259842519686, |
| "eval_loss": 47.40172576904297, |
| "eval_runtime": 7.9017, |
| "eval_samples_per_second": 15.819, |
| "eval_steps_per_second": 2.025, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 12861.1259765625, |
| "learning_rate": 0.00025544320241571017, |
| "loss": 1426.95953125, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_accuracy": 0.1636220472440945, |
| "eval_loss": 47.27663803100586, |
| "eval_runtime": 8.5898, |
| "eval_samples_per_second": 14.552, |
| "eval_steps_per_second": 1.863, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.55, |
| "eval_accuracy": 0.1651023622047244, |
| "eval_loss": 47.143985748291016, |
| "eval_runtime": 10.7415, |
| "eval_samples_per_second": 11.637, |
| "eval_steps_per_second": 1.49, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.6, |
| "grad_norm": 9027.2001953125, |
| "learning_rate": 0.0001822824974551769, |
| "loss": 1423.1790625, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.6, |
| "eval_accuracy": 0.16569291338582678, |
| "eval_loss": 47.07837677001953, |
| "eval_runtime": 8.4354, |
| "eval_samples_per_second": 14.818, |
| "eval_steps_per_second": 1.897, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.65, |
| "eval_accuracy": 0.16648031496062993, |
| "eval_loss": 47.00508117675781, |
| "eval_runtime": 10.7099, |
| "eval_samples_per_second": 11.671, |
| "eval_steps_per_second": 1.494, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.7, |
| "grad_norm": 5090.822265625, |
| "learning_rate": 0.0001115146393893927, |
| "loss": 1419.58796875, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.7, |
| "eval_accuracy": 0.16728346456692914, |
| "eval_loss": 46.935848236083984, |
| "eval_runtime": 8.4034, |
| "eval_samples_per_second": 14.875, |
| "eval_steps_per_second": 1.904, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.75, |
| "eval_accuracy": 0.1675748031496063, |
| "eval_loss": 46.88325119018555, |
| "eval_runtime": 7.7053, |
| "eval_samples_per_second": 16.223, |
| "eval_steps_per_second": 2.076, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.8, |
| "grad_norm": 8640.748046875, |
| "learning_rate": 5.269721958838045e-05, |
| "loss": 1415.78859375, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_accuracy": 0.1682755905511811, |
| "eval_loss": 46.84225845336914, |
| "eval_runtime": 8.5769, |
| "eval_samples_per_second": 14.574, |
| "eval_steps_per_second": 1.865, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.85, |
| "eval_accuracy": 0.16845669291338583, |
| "eval_loss": 46.823673248291016, |
| "eval_runtime": 10.5888, |
| "eval_samples_per_second": 11.805, |
| "eval_steps_per_second": 1.511, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.9, |
| "grad_norm": 14275.54296875, |
| "learning_rate": 1.3773856514840955e-05, |
| "loss": 1415.929375, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.9, |
| "eval_accuracy": 0.1687007874015748, |
| "eval_loss": 46.804019927978516, |
| "eval_runtime": 8.7116, |
| "eval_samples_per_second": 14.349, |
| "eval_steps_per_second": 1.837, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.95, |
| "eval_accuracy": 0.16866141732283466, |
| "eval_loss": 46.80295181274414, |
| "eval_runtime": 8.0554, |
| "eval_samples_per_second": 15.518, |
| "eval_steps_per_second": 1.986, |
| "step": 950 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 11099.6376953125, |
| "learning_rate": 1.3660336561915898e-09, |
| "loss": 1415.8740625, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.16873228346456692, |
| "eval_loss": 46.800193786621094, |
| "eval_runtime": 8.4996, |
| "eval_samples_per_second": 14.707, |
| "eval_steps_per_second": 1.882, |
| "step": 1000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 1000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9223372036854775807, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
