INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æĹ

-0.78

sat

-0.76

inx

-0.72

åŃ

-0.71

spell

-0.71

Georg

-0.70

èª

-0.69

Alexander

-0.69

éĸ

-0.67

ottage

-0.67

POSITIVE LOGITS

 corrid

0.73

 Rica

0.64

 staples

0.64

 Trinidad

0.64

 offline

0.62

 apex

0.60

 alike

0.59

 landslide

0.58

 Soda

0.58

unl

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.