| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.8022053756030324, | |
| "eval_steps": 291, | |
| "global_step": 1164, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06891798759476224, | |
| "grad_norm": 22.43581199645996, | |
| "learning_rate": 9.890136982846565e-06, | |
| "loss": 0.5932, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13783597518952448, | |
| "grad_norm": 17.24200439453125, | |
| "learning_rate": 9.552035901252971e-06, | |
| "loss": 0.5145, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2005513439007581, | |
| "eval_accuracy": 0.7659483067229899, | |
| "eval_loss": 0.47258511185646057, | |
| "eval_runtime": 1145.9546, | |
| "eval_samples_per_second": 73.129, | |
| "eval_steps_per_second": 9.142, | |
| "step": 291 | |
| }, | |
| { | |
| "epoch": 0.2067539627842867, | |
| "grad_norm": 13.582720756530762, | |
| "learning_rate": 9.001378813765828e-06, | |
| "loss": 0.4774, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.27567195037904896, | |
| "grad_norm": 8.677756309509277, | |
| "learning_rate": 8.263878494553072e-06, | |
| "loss": 0.4627, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.34458993797381116, | |
| "grad_norm": 9.698088645935059, | |
| "learning_rate": 7.373972308696536e-06, | |
| "loss": 0.4356, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4011026878015162, | |
| "eval_accuracy": 0.773012577265459, | |
| "eval_loss": 0.46410173177719116, | |
| "eval_runtime": 1144.776, | |
| "eval_samples_per_second": 73.204, | |
| "eval_steps_per_second": 9.151, | |
| "step": 582 | |
| }, | |
| { | |
| "epoch": 0.4135079255685734, | |
| "grad_norm": 6.973531246185303, | |
| "learning_rate": 6.3732141693438686e-06, | |
| "loss": 0.4279, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.4824259131633356, | |
| "grad_norm": 9.015981674194336, | |
| "learning_rate": 5.308334189697317e-06, | |
| "loss": 0.4089, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.5513439007580979, | |
| "grad_norm": 11.981879234313965, | |
| "learning_rate": 4.2290566338312015e-06, | |
| "loss": 0.3823, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.6016540317022743, | |
| "eval_accuracy": 0.7859716951862724, | |
| "eval_loss": 0.4455520510673523, | |
| "eval_runtime": 1145.8146, | |
| "eval_samples_per_second": 73.137, | |
| "eval_steps_per_second": 9.143, | |
| "step": 873 | |
| }, | |
| { | |
| "epoch": 0.6202618883528601, | |
| "grad_norm": 9.888021469116211, | |
| "learning_rate": 3.185778056496399e-06, | |
| "loss": 0.3752, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.6891798759476223, | |
| "grad_norm": 13.089375495910645, | |
| "learning_rate": 2.227214050505248e-06, | |
| "loss": 0.3627, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7580978635423845, | |
| "grad_norm": 10.356831550598145, | |
| "learning_rate": 1.3981244861502275e-06, | |
| "loss": 0.3616, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8022053756030324, | |
| "eval_accuracy": 0.7928092408295745, | |
| "eval_loss": 0.43128883838653564, | |
| "eval_runtime": 1145.9685, | |
| "eval_samples_per_second": 73.128, | |
| "eval_steps_per_second": 9.142, | |
| "step": 1164 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1451, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 291, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |