INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Ļ½

-0.70

nesday

-0.69

ulz

-0.64

worm

-0.63

ledge

-0.63

tein

-0.62

 occurrence

-0.61

ial

-0.60

 Golden

-0.60

rica

-0.59

POSITIVE LOGITS

OTUS

0.76

isSpecialOrderable

0.71

Tex

0.67

HUD

0.66

âĸº

0.64

 Nasa

0.64

wcsstore

0.63

NOW

0.63

bed

0.63

Leave

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.