| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 2862, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9968553459119497, | |
| "grad_norm": 0.6337465047836304, | |
| "learning_rate": 1.7784765897973445e-05, | |
| "loss": 0.3867, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5703225806451613, | |
| "eval_loss": 0.19408121705055237, | |
| "eval_runtime": 1.3792, | |
| "eval_samples_per_second": 2247.653, | |
| "eval_steps_per_second": 47.128, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.5281593799591064, | |
| "learning_rate": 1.556953179594689e-05, | |
| "loss": 0.1621, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8274193548387097, | |
| "eval_loss": 0.0953042134642601, | |
| "eval_runtime": 1.5805, | |
| "eval_samples_per_second": 1961.446, | |
| "eval_steps_per_second": 41.127, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 0.4498799741268158, | |
| "learning_rate": 1.3354297693920338e-05, | |
| "loss": 0.1018, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8806451612903226, | |
| "eval_loss": 0.06378939002752304, | |
| "eval_runtime": 1.5977, | |
| "eval_samples_per_second": 1940.328, | |
| "eval_steps_per_second": 40.684, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.3946130573749542, | |
| "learning_rate": 1.1139063591893781e-05, | |
| "loss": 0.0769, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8980645161290323, | |
| "eval_loss": 0.04853447526693344, | |
| "eval_runtime": 1.6017, | |
| "eval_samples_per_second": 1935.5, | |
| "eval_steps_per_second": 40.583, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.2938636541366577, | |
| "learning_rate": 8.923829489867226e-06, | |
| "loss": 0.0634, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9067741935483871, | |
| "eval_loss": 0.04064928740262985, | |
| "eval_runtime": 1.4207, | |
| "eval_samples_per_second": 2182.039, | |
| "eval_steps_per_second": 45.752, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.3579629063606262, | |
| "learning_rate": 6.708595387840672e-06, | |
| "loss": 0.0555, | |
| "step": 1902 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9135483870967742, | |
| "eval_loss": 0.03552273288369179, | |
| "eval_runtime": 1.617, | |
| "eval_samples_per_second": 1917.16, | |
| "eval_steps_per_second": 40.199, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.977987421383648, | |
| "grad_norm": 0.2372261881828308, | |
| "learning_rate": 4.4933612858141165e-06, | |
| "loss": 0.0508, | |
| "step": 2219 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9193548387096774, | |
| "eval_loss": 0.033018212765455246, | |
| "eval_runtime": 1.8153, | |
| "eval_samples_per_second": 1707.728, | |
| "eval_steps_per_second": 35.807, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.9748427672955975, | |
| "grad_norm": 0.26516783237457275, | |
| "learning_rate": 2.2781271837875614e-06, | |
| "loss": 0.0479, | |
| "step": 2536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9219354838709677, | |
| "eval_loss": 0.03124028816819191, | |
| "eval_runtime": 1.622, | |
| "eval_samples_per_second": 1911.172, | |
| "eval_steps_per_second": 40.073, | |
| "step": 2544 | |
| }, | |
| { | |
| "epoch": 8.971698113207546, | |
| "grad_norm": 0.26064929366111755, | |
| "learning_rate": 6.289308176100629e-08, | |
| "loss": 0.0462, | |
| "step": 2853 | |
| } | |
| ], | |
| "logging_steps": 317, | |
| "max_steps": 2862, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 1000000000.0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 742838622388212.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.26432685644138476, | |
| "num_train_epochs": 9, | |
| "temperature": 15 | |
| } | |
| } | |