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

 Cosponsors

-0.74

icators

-0.70

ablishment

-0.69

ettlement

-0.67

reddits

-0.66

emon

-0.66

iHUD

-0.64

amac

-0.64

uits

-0.63

McM

-0.63

POSITIVE LOGITS

bia

0.81

etheless

0.74

KI

0.73

stru

0.67

fe

0.67

Score

0.66

ngth

0.64

RIP

0.64

antically

0.62

icket

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.