INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anski

-0.67

 athlet

-0.67

ichick

-0.63

angler

-0.63

utor

-0.62

 intimidating

-0.62

kay

-0.59

eele

-0.59

hare

-0.59

tery

-0.59

POSITIVE LOGITS

ationally

0.69

ETA

0.66

inational

0.66

enced

0.65

idia

0.64

etus

0.64

Ł

0.63

NL

0.62

DCS

0.62

ENCE

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.