INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

FG

-0.64

din

-0.62

aptic

-0.61

chest

-0.60

ousel

-0.60

peat

-0.60

DCS

-0.60

 Starg

-0.60

 mutants

-0.60

unia

-0.59

POSITIVE LOGITS

 Â¥

0.69

ulia

0.67

kt

0.66

 Municip

0.66

 lawy

0.65

cms

0.63

atown

0.63

andem

0.62

 spread

0.61

cation

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.