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

arie

-0.76

bsite

-0.72

apter

-0.70

anish

-0.69

taker

-0.67

 Journals

-0.66

 Tournament

-0.66

journal

-0.66

icipated

-0.65

aru

-0.65

POSITIVE LOGITS

 grooming

0.70

DIV

0.67

 relent

0.64

WHERE

0.64

BuyableInstoreAndOnline

0.64

Weak

0.62

CONT

0.62

YL

0.60

Orange

0.59

Change

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.