| { | |
| "best_metric": 0.9238303497851478, | |
| "best_model_checkpoint": "./saved_models/llama_prompt_sbdh_gpt4_v2_0/checkpoint-120", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.6714589595794678, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.5644, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_acc_macro": 0.7160291923788676, | |
| "eval_acc_micro": 0.7509433962263584, | |
| "eval_auc_macro": 0.9741292675539216, | |
| "eval_auc_micro": 0.9750886034759222, | |
| "eval_f1_at_5": 0.4217622259434855, | |
| "eval_f1_at_8": 0.2920473967500829, | |
| "eval_f1_macro": 0.7989797861861626, | |
| "eval_f1_micro": 0.8577586206895812, | |
| "eval_loss": 0.11446325480937958, | |
| "eval_prec_at_5": 0.27009132420091325, | |
| "eval_prec_at_8": 0.1716609589041096, | |
| "eval_prec_macro": 0.8115680364358134, | |
| "eval_prec_micro": 0.8931777378814278, | |
| "eval_rec_at_5": 0.9619482496194824, | |
| "eval_rec_at_8": 0.9777397260273972, | |
| "eval_rec_macro": 0.7915340851938971, | |
| "eval_rec_micro": 0.8250414593697492, | |
| "eval_runtime": 30.8714, | |
| "eval_samples_per_second": 28.376, | |
| "eval_steps_per_second": 3.563, | |
| "eval_threshold": -0.625, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.35452723503112793, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.0736, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_acc_macro": 0.8391717260025603, | |
| "eval_acc_micro": 0.8443432042520239, | |
| "eval_auc_macro": 0.9939078443751225, | |
| "eval_auc_micro": 0.99449507245633, | |
| "eval_f1_at_5": 0.42918048633589306, | |
| "eval_f1_at_8": 0.2922793659426448, | |
| "eval_f1_macro": 0.9106835825010626, | |
| "eval_f1_micro": 0.9156031288595375, | |
| "eval_loss": 0.06005650386214256, | |
| "eval_prec_at_5": 0.27488584474885847, | |
| "eval_prec_at_8": 0.17180365296803654, | |
| "eval_prec_macro": 0.9048916396910188, | |
| "eval_prec_micro": 0.9092395748159517, | |
| "eval_rec_at_5": 0.978310502283105, | |
| "eval_rec_at_8": 0.978310502283105, | |
| "eval_rec_macro": 0.919126994992955, | |
| "eval_rec_micro": 0.9220563847428754, | |
| "eval_runtime": 31.0816, | |
| "eval_samples_per_second": 28.184, | |
| "eval_steps_per_second": 3.539, | |
| "eval_threshold": -1.375, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.5112707614898682, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0498, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_acc_macro": 0.8538277777274056, | |
| "eval_acc_micro": 0.8602316602315938, | |
| "eval_auc_macro": 0.9946434924695814, | |
| "eval_auc_micro": 0.9959011091116522, | |
| "eval_f1_at_5": 0.4288472848093916, | |
| "eval_f1_at_8": 0.29274329072668936, | |
| "eval_f1_macro": 0.9190983946030712, | |
| "eval_f1_micro": 0.924865089248574, | |
| "eval_loss": 0.047555435448884964, | |
| "eval_prec_at_5": 0.27465753424657535, | |
| "eval_prec_at_8": 0.1720890410958904, | |
| "eval_prec_macro": 0.919149383334144, | |
| "eval_prec_micro": 0.9260182876142206, | |
| "eval_rec_at_5": 0.9777397260273972, | |
| "eval_rec_at_8": 0.9794520547945206, | |
| "eval_rec_macro": 0.9214974849786018, | |
| "eval_rec_micro": 0.9237147595355785, | |
| "eval_runtime": 31.1362, | |
| "eval_samples_per_second": 28.134, | |
| "eval_steps_per_second": 3.533, | |
| "eval_threshold": 0.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.3539921045303345, | |
| "learning_rate": 9.411764705882353e-05, | |
| "loss": 0.0409, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_acc_macro": 0.8544346372841939, | |
| "eval_acc_micro": 0.8602484472049021, | |
| "eval_auc_macro": 0.9958044036634556, | |
| "eval_auc_micro": 0.9960722303661861, | |
| "eval_f1_at_5": 0.4288472848093916, | |
| "eval_f1_at_8": 0.29274329072668936, | |
| "eval_f1_macro": 0.919964042577448, | |
| "eval_f1_micro": 0.9248747913187875, | |
| "eval_loss": 0.05106280744075775, | |
| "eval_prec_at_5": 0.27465753424657535, | |
| "eval_prec_at_8": 0.1720890410958904, | |
| "eval_prec_macro": 0.9270437825158037, | |
| "eval_prec_micro": 0.9310924369747117, | |
| "eval_rec_at_5": 0.9777397260273972, | |
| "eval_rec_at_8": 0.9794520547945206, | |
| "eval_rec_macro": 0.914455594415295, | |
| "eval_rec_micro": 0.9187396351574694, | |
| "eval_runtime": 31.1411, | |
| "eval_samples_per_second": 28.13, | |
| "eval_steps_per_second": 3.532, | |
| "eval_threshold": -0.125, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.4515529274940491, | |
| "learning_rate": 8.823529411764706e-05, | |
| "loss": 0.0306, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_acc_macro": 0.8608096832911526, | |
| "eval_acc_micro": 0.8670212765956787, | |
| "eval_auc_macro": 0.9959155249262238, | |
| "eval_auc_micro": 0.9964546135288169, | |
| "eval_f1_at_5": 0.4288472848093916, | |
| "eval_f1_at_8": 0.29274329072668936, | |
| "eval_f1_macro": 0.9238303497851478, | |
| "eval_f1_micro": 0.9287749287748531, | |
| "eval_loss": 0.05085020139813423, | |
| "eval_prec_at_5": 0.27465753424657535, | |
| "eval_prec_at_8": 0.1720890410958904, | |
| "eval_prec_macro": 0.9063834789244836, | |
| "eval_prec_micro": 0.912070343724947, | |
| "eval_rec_at_5": 0.9777397260273972, | |
| "eval_rec_at_8": 0.9794520547945206, | |
| "eval_rec_macro": 0.9425487929333234, | |
| "eval_rec_micro": 0.9461028192370691, | |
| "eval_runtime": 31.1075, | |
| "eval_samples_per_second": 28.16, | |
| "eval_steps_per_second": 3.536, | |
| "eval_threshold": -0.375, | |
| "step": 120 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 480, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.619710176329728e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |