🔄 In a Training Loop

Aditya Iyer

thedarkknight7

·

AI & ML interests

Interpretability, Alignment, Safety

Recent Activity

published a bucket 19 days ago

thedarkknight7/shared_circuits_math_data

published a bucket 19 days ago

thedarkknight7/shared_circuits_math_jsonl

published a bucket 19 days ago

thedarkknight7/shared_circuits_physics_data

View all activity

Organizations

None yet

buckets 6

thedarkknight7/shared_circuits_math_data

thedarkknight7/shared_circuits_math_jsonl

thedarkknight7/shared_circuits_physics_jsonl

thedarkknight7/shared_circuits_physics_data

thedarkknight7/sae-for-monosemanticity-model-weights

thedarkknight7/GPT-2-Activations

spaces 2

SAE Feature Explorer

Launch a Streamlit web app in your browser

Adis Portfolio

Chat with Aditya’s portfolio to learn about his career

models 0

None public yet

datasets 16

thedarkknight7/SAE_monosemanticity_features_32x_0.0001_sampling

Viewer • Updated Mar 27 • 174M • 19

thedarkknight7/SAE_monosemanticity_features_32x_0.01_sampling

Viewer • Updated Mar 27 • 174M • 228

thedarkknight7/SAE_monosemanticity_features_16x_0.0001_sampling

Viewer • Updated Mar 27 • 174M • 47

thedarkknight7/SAE_monosemanticity_features_16x_0.01_sampling

Viewer • Updated Mar 27 • 174M • 267

thedarkknight7/SAE_monosemanticity_features_8x_0.0001_sampling

Viewer • Updated Mar 27 • 174M • 20

thedarkknight7/SAE_monosemanticity_features_8x_0.01_sampling

Viewer • Updated Mar 27 • 174M • 143

thedarkknight7/SAE_monosemanticity_features_4x_0.0001_sampling

Viewer • Updated Mar 27 • 174M • 244

thedarkknight7/SAE_monosemanticity_features_4x_0.01_sampling

Viewer • Updated Mar 27 • 174M • 265

thedarkknight7/SAE_monosemanticity_features_32x_0.0001

Viewer • Updated Mar 19 • 174M • 234

thedarkknight7/SAE_monosemanticity_features_32x_0.01

Viewer • Updated Mar 19 • 174M • 20

View 16 datasets