Add files using upload-large-folder tool

revision_runs/heldout_v3/trainonly/runs/base_1/bce_lr0.0001/model_roberta/dataset_SST5/head_seed_1/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,0.0001,entropy,0,0.5384615659713745,0.6454934915142527,0.6368429724831108,0.5579185485839844,0.6134931485811955,0.6100946257017651

revision_runs/heldout_v3/trainonly/runs/base_1/bce_lr0.0001/model_roberta/dataset_SST5/head_seed_2/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,0.0001,entropy,0,0.5384615659713745,0.6454934915142527,0.6347009391992091,0.5579185485839844,0.6134931485811955,0.6073352974039569

revision_runs/heldout_v3/trainonly/runs/base_1/bce_lr0.0001/model_roberta/dataset_SST5/head_seed_3/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,0.0001,entropy,0,0.5384615659713745,0.6454934915142527,0.632146976437634,0.5579185485839844,0.6134931485811955,0.6057539134065667

revision_runs/heldout_v3/trainonly/runs/base_1/bce_lr0.0001/model_roberta/dataset_SST5/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,0.0001,entropy,0,0.5384615659713745,0.6454934915142527,0.637584445542923,0.5579185485839844,0.6134931485811955,0.6135952536896885

revision_runs/heldout_v3/trainonly/runs/base_1/confidnet_lr0.0003_ep50_wd0.0/model_roberta/dataset_SST5/head_seed_1/LoraConfig_SST5_seed_1_normalization_True_confidnet.csv

@@ -0,0 +1,2 @@
+model,adapter,dataset,train_on_dataset,normalization,seed,test-acc,valid-confidnet-roc-auc,confidnet_roc_auc,best_lr,best_epoch,best_weighting,best_hidden,best_weight_decay
+roberta-base,LoraConfig,SST5,train,True,1,0.5579185485839844,0.5389685285879058,0.5767278384182508,0.0003,50,1.0,400,0.0

revision_runs/heldout_v3/trainonly/runs/base_1/l2sp_lr0.0004_b0.01/model_roberta/dataset_SST5/head_seed_1/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune_l2sp_only,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.6293458559894546,0.5579185485839844,0.6134931485811955,0.6092703137283224

revision_runs/heldout_v3/trainonly/runs/base_1/l2sp_lr0.0004_b0.01/model_roberta/dataset_SST5/head_seed_2/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune_l2sp_only,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.6525786785302357,0.5579185485839844,0.6134931485811955,0.6035540878983864

revision_runs/heldout_v3/trainonly/runs/base_1/l2sp_lr0.0004_b0.01/model_roberta/dataset_SST5/head_seed_3/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune_l2sp_only,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.6322293623331684,0.5579185485839844,0.6134931485811955,0.6058780167701414

revision_runs/heldout_v3/trainonly/runs/base_1/l2sp_lr0.0004_b0.01/model_roberta/dataset_SST5/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune_l2sp_only,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.6549678695007414,0.5579185485839844,0.6134931485811955,0.6151467532650806

revision_runs/heldout_v3/trainonly/runs/base_1/loo_noreg/model_roberta/dataset_SST5/head_seed_0/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.6472235953204811,0.5579185485839844,0.6134931485811955,0.6064346141298528

revision_runs/heldout_v3/trainonly/runs/base_1/loo_noreg/model_roberta/dataset_SST5/head_seed_1/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.6293458559894546,0.5579185485839844,0.6134931485811955,0.6092703137283224

revision_runs/heldout_v3/trainonly/runs/base_1/loo_noreg/model_roberta/dataset_SST5/head_seed_3/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.6322293623331684,0.5579185485839844,0.6134931485811955,0.6058780167701414

revision_runs/heldout_v3/trainonly/runs/base_1/loo_noreg/model_roberta/dataset_SST5/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.6549678695007414,0.5579185485839844,0.6134931485811955,0.6151467532650806

revision_runs/heldout_v3/trainonly/runs/base_1/randcls_lr0.0004/model_roberta/dataset_SST5/head_seed_1/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.5904597132970836,0.5579185485839844,0.6134931485811955,0.5645491063312638

revision_runs/heldout_v3/trainonly/runs/base_1/randcls_lr0.0004/model_roberta/dataset_SST5/head_seed_2/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.5736529906080079,0.5579185485839844,0.6134931485811955,0.5422021996594836

revision_runs/heldout_v3/trainonly/runs/base_1/randcls_lr0.0004/model_roberta/dataset_SST5/head_seed_3/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.5795023891909705,0.5579185485839844,0.6134931485811955,0.5365449125507101

revision_runs/heldout_v3/trainonly/runs/base_1/randcls_lr0.0004/model_roberta/dataset_SST5/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.5988630746416214,0.5579185485839844,0.6134931485811955,0.5650845355587266

revision_runs/heldout_v3/trainonly/runs/base_1/randlin_lr0.0004/model_roberta/dataset_SST5/head_seed_1/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.5847750865051903,0.5579185485839844,0.6134931485811955,0.5538894990291714

revision_runs/heldout_v3/trainonly/runs/base_1/randlin_lr0.0004/model_roberta/dataset_SST5/head_seed_2/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.572829131652661,0.5579185485839844,0.6134931485811955,0.5623916170875803

revision_runs/heldout_v3/trainonly/runs/base_1/randlin_lr0.0004/model_roberta/dataset_SST5/head_seed_3/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.5782666007579502,0.5579185485839844,0.6134931485811955,0.5705176895025157

revision_runs/heldout_v3/trainonly/runs/base_1/randlin_lr0.0004/model_roberta/dataset_SST5/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.5384615659713745,0.6454934915142527,0.5784313725490197,0.5579185485839844,0.6134931485811955,0.5705957210488436

revision_runs/heldout_v3/trainonly/runs/base_1/reg_lr0.0001_a0.01/model_roberta/dataset_SST5/head_seed_1/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune_reg_only,100.0,0.01,0.0,cls_head,0.0001,entropy,0,0.5384615659713745,0.6454934915142527,0.6366782006920415,0.5579185485839844,0.6134931485811955,0.6102340863377553

revision_runs/heldout_v3/trainonly/runs/base_1/reg_lr0.0001_a0.01/model_roberta/dataset_SST5/head_seed_2/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune_reg_only,100.0,0.01,0.0,cls_head,0.0001,entropy,0,0.5384615659713745,0.6454934915142527,0.6347833250947438,0.5579185485839844,0.6134931485811955,0.6074041975991188

revision_runs/heldout_v3/trainonly/runs/base_1/reg_lr0.0001_a0.01/model_roberta/dataset_SST5/head_seed_3/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune_reg_only,100.0,0.01,0.0,cls_head,0.0001,entropy,0,0.5384615659713745,0.6454934915142527,0.6316526610644257,0.5579185485839844,0.6134931485811955,0.60580289065373

revision_runs/heldout_v3/trainonly/runs/base_1/reg_lr0.0001_a0.01/model_roberta/dataset_SST5/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,roberta-base,LoraConfig,SST5,train,1,entropy_response_finetune_reg_only,100.0,0.01,0.0,cls_head,0.0001,entropy,0,0.5384615659713745,0.6454934915142527,0.6370901301697149,0.5579185485839844,0.6134931485811955,0.6136948684296816

revision_runs/heldout_v3/trainonly/runs/base_2/bce_lr1e-05/model_electra/dataset_toxigen/head_seed_1/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,1e-05,entropy,0,0.8467966318130493,0.7159688995215311,0.7242224880382775,0.8180851340293884,0.7450779093377136,0.743545578293371

revision_runs/heldout_v3/trainonly/runs/base_2/bce_lr1e-05/model_electra/dataset_toxigen/head_seed_2/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,1e-05,entropy,0,0.8467966318130493,0.7159688995215311,0.7227272727272727,0.8180851340293884,0.7450779093377136,0.7430094525433655

revision_runs/heldout_v3/trainonly/runs/base_2/bce_lr1e-05/model_electra/dataset_toxigen/head_seed_3/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,1e-05,entropy,0,0.8467966318130493,0.7159688995215311,0.7235047846889953,0.8180851340293884,0.7450779093377136,0.7429219994068397

revision_runs/heldout_v3/trainonly/runs/base_2/bce_lr1e-05/model_electra/dataset_toxigen/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,1e-05,entropy,0,0.8467966318130493,0.7159688995215311,0.7236244019138756,0.8180851340293884,0.7450779093377136,0.7433174396763473

revision_runs/heldout_v3/trainonly/runs/base_2/confidnet_lr0.0003_ep50_wd0.0/model_electra/dataset_toxigen/head_seed_1/LoraConfig_toxigen_seed_2_normalization_True_confidnet.csv

@@ -0,0 +1,2 @@
+model,adapter,dataset,train_on_dataset,normalization,seed,test-acc,valid-confidnet-roc-auc,confidnet_roc_auc,best_lr,best_epoch,best_weighting,best_hidden,best_weight_decay
+google/electra-base-discriminator,LoraConfig,toxigen,train,True,2,0.8180851340293884,0.7260167464114833,0.722488383942083,0.0003,50,1.0,400,0.0

revision_runs/heldout_v3/trainonly/runs/base_2/confidnet_lr0.0003_ep50_wd0.0/model_electra/dataset_toxigen/head_seed_2/LoraConfig_toxigen_seed_2_normalization_True_confidnet.csv

@@ -0,0 +1,2 @@
+model,adapter,dataset,train_on_dataset,normalization,seed,test-acc,valid-confidnet-roc-auc,confidnet_roc_auc,best_lr,best_epoch,best_weighting,best_hidden,best_weight_decay
+google/electra-base-discriminator,LoraConfig,toxigen,train,True,2,0.8180851340293884,0.7272428229665071,0.7218381888835657,0.0003,50,1.0,400,0.0

revision_runs/heldout_v3/trainonly/runs/base_2/confidnet_lr0.0003_ep50_wd0.0/model_electra/dataset_toxigen/head_seed_3/LoraConfig_toxigen_seed_2_normalization_True_confidnet.csv

@@ -0,0 +1,2 @@
+model,adapter,dataset,train_on_dataset,normalization,seed,test-acc,valid-confidnet-roc-auc,confidnet_roc_auc,best_lr,best_epoch,best_weighting,best_hidden,best_weight_decay
+google/electra-base-discriminator,LoraConfig,toxigen,train,True,2,0.8180851340293884,0.727811004784689,0.7211955984456156,0.0003,50,1.0,400,0.0

revision_runs/heldout_v3/trainonly/runs/base_2/confidnet_lr0.0003_ep50_wd0.0/model_electra/dataset_toxigen/head_seed_4/LoraConfig_toxigen_seed_2_normalization_True_confidnet.csv

@@ -0,0 +1,2 @@
+model,adapter,dataset,train_on_dataset,normalization,seed,test-acc,valid-confidnet-roc-auc,confidnet_roc_auc,best_lr,best_epoch,best_weighting,best_hidden,best_weight_decay
+google/electra-base-discriminator,LoraConfig,toxigen,train,True,2,0.8180851340293884,0.7343899521531101,0.7204655548711397,0.0003,50,1.0,400,0.0

revision_runs/heldout_v3/trainonly/runs/base_2/l2sp_lr0.0001_b0.1/model_electra/dataset_toxigen/head_seed_1/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,entropy_response_finetune_l2sp_only,100.0,0.0,0.1,cls_head,0.0001,entropy,0,0.8467966318130493,0.7159688995215311,0.7217703349282296,0.8180851340293884,0.7450779093377136,0.745518977330626

revision_runs/heldout_v3/trainonly/runs/base_2/l2sp_lr0.0001_b0.1/model_electra/dataset_toxigen/head_seed_2/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,entropy_response_finetune_l2sp_only,100.0,0.0,0.1,cls_head,0.0001,entropy,0,0.8467966318130493,0.7159688995215311,0.7219497607655502,0.8180851340293884,0.7450779093377136,0.7460132776675107

revision_runs/heldout_v3/trainonly/runs/base_2/l2sp_lr0.0001_b0.1/model_electra/dataset_toxigen/head_seed_3/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,entropy_response_finetune_l2sp_only,100.0,0.0,0.1,cls_head,0.0001,entropy,0,0.8467966318130493,0.7159688995215311,0.7252990430622008,0.8180851340293884,0.7450779093377136,0.746332671731344

revision_runs/heldout_v3/trainonly/runs/base_2/l2sp_lr0.0001_b0.1/model_electra/dataset_toxigen/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,entropy_response_finetune_l2sp_only,100.0,0.0,0.1,cls_head,0.0001,entropy,0,0.8467966318130493,0.7159688995215311,0.7215311004784688,0.8180851340293884,0.7450779093377136,0.74568627898311

revision_runs/heldout_v3/trainonly/runs/base_2/loo_randlin/model_electra/dataset_toxigen/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_linear_random_reg,100.0,0.1,1.0,linear_random,0.0001,entropy,0,0.8467966318130493,0.7159688995215311,0.5611842105263157,0.8180851340293884,0.7450779093377136,0.49985170989893457

revision_runs/heldout_v3/trainonly/runs/base_2/randcls_lr0.0004/model_electra/dataset_toxigen/head_seed_1/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.8467966318130493,0.7159688995215311,0.601255980861244,0.8180851340293884,0.7450779093377136,0.48465007338458843

revision_runs/heldout_v3/trainonly/runs/base_2/randcls_lr0.0004/model_electra/dataset_toxigen/head_seed_2/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.8467966318130493,0.7159688995215311,0.6842105263157895,0.8180851340293884,0.7450779093377136,0.5446010996281341

revision_runs/heldout_v3/trainonly/runs/base_2/randcls_lr0.0004/model_electra/dataset_toxigen/head_seed_3/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.8467966318130493,0.7159688995215311,0.722188995215311,0.8180851340293884,0.7450779093377136,0.7162944204898897

revision_runs/heldout_v3/trainonly/runs/base_2/randcls_lr0.0004/model_electra/dataset_toxigen/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.8467966318130493,0.7159688995215311,0.65311004784689,0.8180851340293884,0.7450779093377136,0.5219887603708012

revision_runs/heldout_v3/trainonly/runs/base_2/randlin_lr0.0004/model_electra/dataset_toxigen/head_seed_1/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.8467966318130493,0.7159688995215311,0.5757177033492823,0.8180851340293884,0.7450779093377136,0.5301333089985475

revision_runs/heldout_v3/trainonly/runs/base_2/randlin_lr0.0004/model_electra/dataset_toxigen/head_seed_2/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.8467966318130493,0.7159688995215311,0.5643540669856459,0.8180851340293884,0.7450779093377136,0.4470452246785146

revision_runs/heldout_v3/trainonly/runs/base_2/randlin_lr0.0004/model_electra/dataset_toxigen/head_seed_3/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.8467966318130493,0.7159688995215311,0.5341507177033492,0.8180851340293884,0.7450779093377136,0.44310603122457204

revision_runs/heldout_v3/trainonly/runs/base_2/randlin_lr0.0004/model_electra/dataset_toxigen/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.8467966318130493,0.7159688995215311,0.5584928229665072,0.8180851340293884,0.7450779093377136,0.4947109863953338

revision_runs/heldout_v3/trainonly/runs/base_2/reg_lr0.0001_a0.1/model_electra/dataset_toxigen/head_seed_1/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,entropy_response_finetune_reg_only,100.0,0.1,0.0,cls_head,0.0001,entropy,0,0.8467966318130493,0.7159688995215311,0.7257775119617225,0.8180851340293884,0.7450779093377136,0.7445836090008289

revision_runs/heldout_v3/trainonly/runs/base_2/reg_lr0.0001_a0.1/model_electra/dataset_toxigen/head_seed_2/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,entropy_response_finetune_reg_only,100.0,0.1,0.0,cls_head,0.0001,entropy,0,0.8467966318130493,0.7159688995215311,0.7249700956937798,0.8180851340293884,0.7450779093377136,0.7460893238731853

revision_runs/heldout_v3/trainonly/runs/base_2/reg_lr0.0001_a0.1/model_electra/dataset_toxigen/head_seed_3/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,entropy_response_finetune_reg_only,100.0,0.1,0.0,cls_head,0.0001,entropy,0,0.8467966318130493,0.7159688995215311,0.7249700956937799,0.8180851340293884,0.7450779093377136,0.7433668697100358

revision_runs/heldout_v3/trainonly/runs/base_2/reg_lr0.0001_a0.1/model_electra/dataset_toxigen/head_seed_4/best_run_metric_df.csv

@@ -0,0 +1,2 @@
+,model,adapter,dataset,train_on_dataset,seed,grid,lam,reg_alpha,l2sp_alpha,load_weights,lr,head_type,trial_number,valid-acc,valid-base-max-prob-roc-auc,valid-fine-tune-max-prob-roc-auc,test-acc,test-base-max-prob-roc-auc,test-fine-tune-max-prob-roc-auc
+0,google/electra-base-discriminator,LoraConfig,toxigen,train,2,entropy_response_finetune_reg_only,100.0,0.1,0.0,cls_head,0.0001,entropy,0,0.8467966318130493,0.7159688995215311,0.7222488038277513,0.8180851340293884,0.7450779093377136,0.7455113727100585