INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

SPONSORED

-0.70

chant

-0.67

ench

-0.64

EStreamFrame

-0.64

utter

-0.63

Jiu

-0.62

ICAN

-0.59

HER

-0.59

 grappling

-0.58

anche

-0.58

POSITIVE LOGITS

atson

0.72

zbek

0.70

idem

0.70

ulhu

0.69

alys

0.69

avier

0.69

ensibly

0.68

incerity

0.68

umerable

0.67

redd

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.