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

agons

-0.79

CLS

-0.77

ijk

-0.72

oppy

-0.71

inness

-0.70

achel

-0.67

alias

-0.66

Alc

-0.63

asm

-0.61

atel

-0.61

POSITIVE LOGITS

ð

0.80

hower

0.73

HER

0.71

ulhu

0.70

terday

0.70

yip

0.66

rss

0.66

heet

0.66

DonaldTrump

0.65

 Calculator

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.