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

pmwiki

-0.80

 Flavoring

-0.72

urities

-0.71

 dearly

-0.69

 Solitaire

-0.69

aimon

-0.69

tera

-0.66

pron

-0.66

Elf

-0.64

laus

-0.64

POSITIVE LOGITS

library

0.76

 Canal

0.72

SHARE

0.71

ository

0.69

@@

0.68

GOP

0.67

credit

0.67

 reconc

0.64

realDonaldTrump

0.62

RECT

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.