Marta Samoilenko

Retrogradi's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

updated a dataset 2 days ago

Retrogradi/gender_stereoset_rephrased

Preview • Updated 2 days ago • 28

updated a collection 3 months ago

Bias Mitigation via Mechanistic Interpretability

Collection

10 items • Updated Apr 28

updated a model 3 months ago

Retrogradi/gemma2b_best_model_wino_cmp_sft_snr_mlp_1layers_ul1.0_lr5e-06

Updated Apr 28

published a model 3 months ago

Retrogradi/gemma2b_best_model_wino_cmp_sft_snr_mlp_1layers_ul1.0_lr5e-06

Updated Apr 28

updated a model 3 months ago

Retrogradi/llama3.2_wino_cmp_dpo_snr_mlp_1layers_beta0.3_lr5e-06

Updated Apr 28

published a model 3 months ago

Retrogradi/llama3.2_wino_cmp_dpo_snr_mlp_1layers_beta0.3_lr5e-06

Updated Apr 28

updated a model 3 months ago

Retrogradi/gemma2b_wino_dpo_mlp_impact_only_0.5_beta0.5_lr1e-06

Updated Apr 27

published a model 3 months ago

Retrogradi/gemma2b_wino_dpo_mlp_impact_only_0.5_beta0.5_lr1e-06

Updated Apr 27

updated a collection 3 months ago

Bias Mitigation via Mechanistic Interpretability

Collection

10 items • Updated Apr 28

updated a model 3 months ago

Retrogradi/llama3.2_wino_dpo_mlp_from_attn_10.0_beta0.5_lr1e-05

Updated Apr 27

published a model 3 months ago

Retrogradi/llama3.2_wino_dpo_mlp_from_attn_10.0_beta0.5_lr1e-05

Updated Apr 27

updated a collection 3 months ago

Bias Mitigation via Mechanistic Interpretability

Collection

10 items • Updated Apr 28

updated 2 models 3 months ago

Retrogradi/gemma2b_best_model_wino_sft_attn_0.5_ul1.0_lr1e-05

Updated Apr 27

Retrogradi/llama3.2_best_model_wino_sft_attn_1.0_ul1.0_lr1e-06

Updated Apr 27

published a model 3 months ago

Retrogradi/gemma2b_best_model_wino_sft_attn_0.5_ul1.0_lr1e-05

Updated Apr 27