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

cry

-0.84

odi

-0.79

veyard

-0.73

 Residents

-0.70

Cu

-0.69

interstitial

-0.69

gallery

-0.66

bub

-0.65

thro

-0.65

 Slovakia

-0.64

POSITIVE LOGITS

lege

0.81

restling

0.73

ifa

0.71

ython

0.71

<[

0.71

 histor

0.64

Jiu

0.64

andowski

0.63

ahn

0.63

 historically

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.