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

xtap

-0.79

SPONSORED

-0.79

uble

-0.71

interstitial

-0.66

verson

-0.66

TMZ

-0.65

[+

-0.65

 Celebrity

-0.64

 showc

-0.64

 overshadowed

-0.62

POSITIVE LOGITS

hiro

0.71

 Orient

0.70

 Western

0.70

nect

0.69

enment

0.68

 Mith

0.66

 Raid

0.64

Mit

0.63

Sov

0.62

qt

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.