Add files using upload-large-folder tool

revision_runs/heldout_v3/trainonly/runs/base_2/bce_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.9004047322540473,0.7460169792175293,0.845690732483238,0.853121089515289

revision_runs/heldout_v3/trainonly/runs/base_2/bce_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8994396014943961,0.7460169792175293,0.845690732483238,0.8499103231168018

revision_runs/heldout_v3/trainonly/runs/base_2/bce_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8972914072229141,0.7460169792175293,0.845690732483238,0.8548046649655083

revision_runs/heldout_v3/trainonly/runs/base_2/bce_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only,100.0,0.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8953300124533002,0.7460169792175293,0.845690732483238,0.8524716891489157

revision_runs/heldout_v3/trainonly/runs/base_2/confidnet_lr0.0003_ep30_wd0.0/model_roberta/dataset_20newsgroups/head_seed_1/LoraConfig_20newsgroups_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
+roberta-base,LoraConfig,20newsgroups,train,True,2,0.7460169792175293,0.8618929016189291,0.813800388068002,0.0003,30,1.0,400,0.0

revision_runs/heldout_v3/trainonly/runs/base_2/confidnet_lr0.0003_ep30_wd0.0/model_roberta/dataset_20newsgroups/head_seed_2/LoraConfig_20newsgroups_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
+roberta-base,LoraConfig,20newsgroups,train,True,2,0.7460169792175293,0.8545143212951433,0.8069932432778154,0.0003,30,1.0,400,0.0

revision_runs/heldout_v3/trainonly/runs/base_2/confidnet_lr0.0003_ep30_wd0.0/model_roberta/dataset_20newsgroups/head_seed_3/LoraConfig_20newsgroups_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
+roberta-base,LoraConfig,20newsgroups,train,True,2,0.7460169792175293,0.8570983810709838,0.8138716495364702,0.0003,30,1.0,400,0.0

revision_runs/heldout_v3/trainonly/runs/base_2/confidnet_lr0.0003_ep30_wd0.0/model_roberta/dataset_20newsgroups/head_seed_4/LoraConfig_20newsgroups_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
+roberta-base,LoraConfig,20newsgroups,train,True,2,0.7460169792175293,0.8636986301369862,0.8187495714776251,0.0003,30,1.0,400,0.0

revision_runs/heldout_v3/trainonly/runs/base_2/l2sp_lr0.0004_b0.01/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune_l2sp_only,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8813200498132004,0.7460169792175293,0.845690732483238,0.8546518156277889

revision_runs/heldout_v3/trainonly/runs/base_2/l2sp_lr0.0004_b0.01/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune_l2sp_only,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8868306351183064,0.7460169792175293,0.845690732483238,0.8559000076936337

revision_runs/heldout_v3/trainonly/runs/base_2/l2sp_lr0.0004_b0.01/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune_l2sp_only,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.886052303860523,0.7460169792175293,0.845690732483238,0.8511847498224743

revision_runs/heldout_v3/trainonly/runs/base_2/l2sp_lr0.0004_b0.01/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune_l2sp_only,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8895392278953923,0.7460169792175293,0.845690732483238,0.8511785167697492

revision_runs/heldout_v3/trainonly/runs/base_2/loo_bce/model_electra/dataset_20newsgroups/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,google/electra-base-discriminator,LoraConfig,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.0,cls_head,0.0001,entropy,0,0.7748344540596008,0.8991117814647226,0.9199206748226356,0.7205257415771484,0.8302089884876662,0.8548269473429895

revision_runs/heldout_v3/trainonly/runs/base_2/loo_bce/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8982565379825653,0.7460169792175293,0.845690732483238,0.8526584481540721

revision_runs/heldout_v3/trainonly/runs/base_2/loo_bce/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.9004047322540473,0.7460169792175293,0.845690732483238,0.853121089515289

revision_runs/heldout_v3/trainonly/runs/base_2/loo_bce/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8994396014943961,0.7460169792175293,0.845690732483238,0.8499103231168018

revision_runs/heldout_v3/trainonly/runs/base_2/loo_bce/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8972914072229141,0.7460169792175293,0.845690732483238,0.8548046649655083

revision_runs/heldout_v3/trainonly/runs/base_2/loo_bce/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8953300124533002,0.7460169792175293,0.845690732483238,0.8524716891489157

revision_runs/heldout_v3/trainonly/runs/base_2/loo_nol2sp/model_electra/dataset_20newsgroups/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,google/electra-base-discriminator,LoraConfig,20newsgroups,train,2,entropy_response_finetune,100.0,1.0,0.0,cls_head,0.0001,entropy,0,0.7748344540596008,0.8991117814647226,0.9200603318250378,0.7205257415771484,0.8302089884876662,0.8541545811892415

revision_runs/heldout_v3/trainonly/runs/base_2/loo_nol2sp/model_electra/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,1.0,0.0,cls_head,0.0001,entropy,0,0.7748344540596008,0.8991117814647226,0.9207306854365678,0.7205257415771484,0.8302089884876662,0.8531521157299629

revision_runs/heldout_v3/trainonly/runs/base_2/loo_nol2sp/model_electra/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,1.0,0.0,cls_head,0.0001,entropy,0,0.7748344540596008,0.8991117814647226,0.9183006535947713,0.7205257415771484,0.8302089884876662,0.8517363915007525

revision_runs/heldout_v3/trainonly/runs/base_2/loo_nol2sp/model_electra/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,1.0,0.0,cls_head,0.0001,entropy,0,0.7748344540596008,0.8991117814647226,0.9208982738394503,0.7205257415771484,0.8302089884876662,0.8546156347671338

revision_runs/heldout_v3/trainonly/runs/base_2/loo_nol2sp/model_electra/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,1.0,0.0,cls_head,0.0001,entropy,0,0.7748344540596008,0.8991117814647226,0.9201161946259986,0.7205257415771484,0.8302089884876662,0.8526059768895471

revision_runs/heldout_v3/trainonly/runs/base_2/loo_nol2sp/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,1.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8989414694894148,0.7460169792175293,0.845690732483238,0.8556804553886928

revision_runs/heldout_v3/trainonly/runs/base_2/loo_nol2sp/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,1.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8987546699875467,0.7460169792175293,0.845690732483238,0.8543652812637131

revision_runs/heldout_v3/trainonly/runs/base_2/loo_nol2sp/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,1.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8985990037359901,0.7460169792175293,0.845690732483238,0.8504731584748073

revision_runs/heldout_v3/trainonly/runs/base_2/loo_nol2sp/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,1.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8990971357409714,0.7460169792175293,0.845690732483238,0.8544885468586485

revision_runs/heldout_v3/trainonly/runs/base_2/loo_nol2sp/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,1.0,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8967621419676214,0.7460169792175293,0.845690732483238,0.8561542604264318

revision_runs/heldout_v3/trainonly/runs/base_2/loo_noreg/model_electra/dataset_20newsgroups/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,google/electra-base-discriminator,LoraConfig,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.01,cls_head,0.0001,entropy,0,0.7748344540596008,0.8991117814647226,0.9150606111390425,0.7205257415771484,0.8302089884876662,0.8453756553731737

revision_runs/heldout_v3/trainonly/runs/base_2/loo_noreg/model_electra/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.01,cls_head,0.0001,entropy,0,0.7748344540596008,0.8991117814647226,0.9157030333500921,0.7205257415771484,0.8302089884876662,0.8437157486956001

revision_runs/heldout_v3/trainonly/runs/base_2/loo_noreg/model_electra/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.01,cls_head,0.0001,entropy,0,0.7748344540596008,0.8991117814647226,0.9142226691246299,0.7205257415771484,0.8302089884876662,0.8451678442484507

revision_runs/heldout_v3/trainonly/runs/base_2/loo_noreg/model_electra/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.01,cls_head,0.0001,entropy,0,0.7748344540596008,0.8991117814647226,0.9140830121222279,0.7205257415771484,0.8302089884876662,0.8454603904905839

revision_runs/heldout_v3/trainonly/runs/base_2/loo_noreg/model_electra/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.01,cls_head,0.0001,entropy,0,0.7748344540596008,0.8991117814647226,0.9165689067649853,0.7205257415771484,0.8302089884876662,0.8442502889791388

revision_runs/heldout_v3/trainonly/runs/base_2/loo_noreg/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.886052303860523,0.7460169792175293,0.845690732483238,0.8511847498224743

revision_runs/heldout_v3/trainonly/runs/base_2/loo_noreg/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune,100.0,0.0,0.01,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8895392278953923,0.7460169792175293,0.845690732483238,0.8511785167697492

revision_runs/heldout_v3/trainonly/runs/base_2/loo_randcls/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_cls_random_reg,100.0,1.0,0.01,cls_random,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.7681506849315068,0.7460169792175293,0.845690732483238,0.7453853780211583

revision_runs/heldout_v3/trainonly/runs/base_2/loo_randcls/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_cls_random_reg,100.0,1.0,0.01,cls_random,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.7433686176836862,0.7460169792175293,0.845690732483238,0.7288921623269269

revision_runs/heldout_v3/trainonly/runs/base_2/loo_randcls/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_cls_random_reg,100.0,1.0,0.01,cls_random,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.7672166874221669,0.7460169792175293,0.845690732483238,0.7468105143598837

revision_runs/heldout_v3/trainonly/runs/base_2/randcls_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.7952054794520548,0.7460169792175293,0.845690732483238,0.7760134827442586

revision_runs/heldout_v3/trainonly/runs/base_2/randcls_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.7939290161892901,0.7460169792175293,0.845690732483238,0.7714870770676037

revision_runs/heldout_v3/trainonly/runs/base_2/randcls_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.7988480697384807,0.7460169792175293,0.845690732483238,0.7724225001295453

revision_runs/heldout_v3/trainonly/runs/base_2/randcls_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only_cls_random,100.0,0.0,0.0,cls_random,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8041095890410959,0.7460169792175293,0.845690732483238,0.7770075151079431

revision_runs/heldout_v3/trainonly/runs/base_2/randlin_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.7568804483188045,0.7460169792175293,0.845690732483238,0.7407405443427278

revision_runs/heldout_v3/trainonly/runs/base_2/randlin_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.7459215442092154,0.7460169792175293,0.845690732483238,0.7370670435523861

revision_runs/heldout_v3/trainonly/runs/base_2/randlin_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.7515877957658779,0.7460169792175293,0.845690732483238,0.7378373372324335

revision_runs/heldout_v3/trainonly/runs/base_2/randlin_lr0.0004/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,ablation_entropy_finetune_bce_only_linear_random,100.0,0.0,0.0,linear_random,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.7465753424657534,0.7460169792175293,0.845690732483238,0.7402637623245827

revision_runs/heldout_v3/trainonly/runs/base_2/reg_lr0.0004_a0.1/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune_reg_only,100.0,0.1,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.9004669987546701,0.7460169792175293,0.845690732483238,0.8533756213400002

revision_runs/heldout_v3/trainonly/runs/base_2/reg_lr0.0004_a0.1/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune_reg_only,100.0,0.1,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8993150684931508,0.7460169792175293,0.845690732483238,0.8499763748695596

revision_runs/heldout_v3/trainonly/runs/base_2/reg_lr0.0004_a0.1/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune_reg_only,100.0,0.1,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8974782067247821,0.7460169792175293,0.845690732483238,0.8527837604230364

revision_runs/heldout_v3/trainonly/runs/base_2/reg_lr0.0004_a0.1/model_roberta/dataset_20newsgroups/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,20newsgroups,train,2,entropy_response_finetune_reg_only,100.0,0.1,0.0,cls_head,0.0004,entropy,0,0.8057395219802856,0.8654420921544209,0.8951743462017434,0.7460169792175293,0.845690732483238,0.8526006761280687