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

nard

-0.76

sein

-0.75

heimer

-0.70

phe

-0.65

hra

-0.65

amn

-0.64

hammer

-0.64

hari

-0.63

HCR

-0.62

channelAvailability

-0.62

POSITIVE LOGITS

 similar

0.72

lot

0.64

 prosecut

0.64

actory

0.63

 stewards

0.63

aires

0.62

minist

0.62

 heirs

0.61

loving

0.58

llor

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.