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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

NAS

-0.72

Boost

-0.69

 istg

-0.67

 feeds

-0.65

nz

-0.64

 Reborn

-0.64

Ì¶

-0.63

ciplinary

-0.63

duct

-0.63

sers

-0.62

POSITIVE LOGITS

elia

0.88

wl

0.73

 precedent

0.73

 princes

0.70

 BaseType

0.66

 unnamed

0.66

 impression

0.65

 unspecified

0.65

 royalty

0.64

 discretion

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.