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

Sham

-0.77

ãĤ¡

-0.76

pace

-0.74

æ©

-0.68

vt

-0.66

 Kund

-0.65

Wan

-0.65

kamp

-0.65

tur

-0.64

å¤

-0.64

POSITIVE LOGITS

 recess

0.79

 deduction

0.77

utory

0.74

clair

0.71

 prejudice

0.71

pling

0.67

ples

0.66

 sensitivity

0.64

obin

0.64

ourke

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.