INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 reinvest

-0.67

lished

-0.67

 churn

-0.67

RIS

-0.66

ellar

-0.62

ADRA

-0.62

itched

-0.61

ADE

-0.60

ITION

-0.59

POSITIVE LOGITS

ydia

0.85

issues

0.79

worldly

0.74

wise

0.71

woman

0.69

wagon

0.69

uo

0.68

yk

0.67

xit

0.67

force

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.