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

 explorer

-0.69

 Podesta

-0.63

ECO

-0.62

 embark

-0.62

Tokens

-0.62

 NEWS

-0.60

aline

-0.60

HI

-0.58

hr

-0.58

WER

-0.57

POSITIVE LOGITS

Ĥª

0.78

earances

0.77

terness

0.74

isphere

0.72

Registered

0.69

assing

0.69

 constituted

0.68

ilt

0.66

ledged

0.65

ccording

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.