INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ocese

-0.79

pez

-0.78

ornings

-0.74

gyn

-0.73

isexual

-0.71

oeuv

-0.70

oji

-0.70

quished

-0.69

luster

-0.64

)=

-0.63

POSITIVE LOGITS

Rah

0.68

heid

0.66

Sed

0.63

scription

0.63

iterator

0.63

Sph

0.62

Coh

0.62

endar

0.61

 Cater

0.61

interstitial

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.