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

 crane

-0.92

 mature

-0.75

Pis

-0.73

 night

-0.71

aceous

-0.69

 rocket

-0.69

pal

-0.68

 daylight

-0.67

Ń·

-0.66

 recess

-0.66

POSITIVE LOGITS

ebook

0.84

ODUCT

0.79

EV

0.79

DEV

0.76

BuyableInstoreAndOnline

0.72

SOURCE

0.72

RFC

0.71

tap

0.71

gaard

0.71

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.