| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5698005698005698, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.011396011396011397, | |
| "eval_loss": 10.161840438842773, | |
| "eval_runtime": 1.2617, | |
| "eval_samples_per_second": 117.305, | |
| "eval_steps_per_second": 15.059, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03418803418803419, | |
| "grad_norm": 23.6888427734375, | |
| "learning_rate": 3e-05, | |
| "loss": 10.3451, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05698005698005698, | |
| "eval_loss": 9.563305854797363, | |
| "eval_runtime": 1.2584, | |
| "eval_samples_per_second": 117.605, | |
| "eval_steps_per_second": 15.098, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06837606837606838, | |
| "grad_norm": 20.72124481201172, | |
| "learning_rate": 6e-05, | |
| "loss": 9.8666, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.10256410256410256, | |
| "grad_norm": 20.2584285736084, | |
| "learning_rate": 9e-05, | |
| "loss": 8.914, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.11396011396011396, | |
| "eval_loss": 7.412201881408691, | |
| "eval_runtime": 1.2606, | |
| "eval_samples_per_second": 117.406, | |
| "eval_steps_per_second": 15.072, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13675213675213677, | |
| "grad_norm": 18.148941040039062, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 7.354, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.17094017094017094, | |
| "grad_norm": 17.978164672851562, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 5.8882, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.17094017094017094, | |
| "eval_loss": 5.262752056121826, | |
| "eval_runtime": 1.2623, | |
| "eval_samples_per_second": 117.251, | |
| "eval_steps_per_second": 15.052, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.20512820512820512, | |
| "grad_norm": 17.917509078979492, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 5.0732, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.22792022792022792, | |
| "eval_loss": 3.6343350410461426, | |
| "eval_runtime": 1.2596, | |
| "eval_samples_per_second": 117.498, | |
| "eval_steps_per_second": 15.084, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23931623931623933, | |
| "grad_norm": 15.94485855102539, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 3.8526, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.27350427350427353, | |
| "grad_norm": 8.799951553344727, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 3.201, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2849002849002849, | |
| "eval_loss": 2.803992509841919, | |
| "eval_runtime": 1.26, | |
| "eval_samples_per_second": 117.457, | |
| "eval_steps_per_second": 15.079, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 6.905522346496582, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 2.8209, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3418803418803419, | |
| "grad_norm": 7.833083629608154, | |
| "learning_rate": 5e-05, | |
| "loss": 2.7369, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3418803418803419, | |
| "eval_loss": 2.528679847717285, | |
| "eval_runtime": 1.2636, | |
| "eval_samples_per_second": 117.126, | |
| "eval_steps_per_second": 15.036, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.37606837606837606, | |
| "grad_norm": 4.130680084228516, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 2.4977, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.39886039886039887, | |
| "eval_loss": 2.408794641494751, | |
| "eval_runtime": 1.2622, | |
| "eval_samples_per_second": 117.256, | |
| "eval_steps_per_second": 15.053, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.41025641025641024, | |
| "grad_norm": 4.275054454803467, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 2.4953, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 3.2638161182403564, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 2.3914, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.45584045584045585, | |
| "eval_loss": 2.357823610305786, | |
| "eval_runtime": 1.262, | |
| "eval_samples_per_second": 117.27, | |
| "eval_steps_per_second": 15.055, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.47863247863247865, | |
| "grad_norm": 2.7736527919769287, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 2.3498, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.5128205128205128, | |
| "grad_norm": 4.598160266876221, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 2.3898, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5128205128205128, | |
| "eval_loss": 2.3354244232177734, | |
| "eval_runtime": 1.2683, | |
| "eval_samples_per_second": 116.688, | |
| "eval_steps_per_second": 14.98, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5470085470085471, | |
| "grad_norm": 6.029376029968262, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 2.4233, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.5698005698005698, | |
| "eval_loss": 2.3306732177734375, | |
| "eval_runtime": 1.259, | |
| "eval_samples_per_second": 117.55, | |
| "eval_steps_per_second": 15.091, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 41046198190080.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |