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

umbn

-0.64

alth

-0.63

emia

-0.62

uty

-0.61

ifter

-0.61

Ys

-0.61

wagen

-0.61

Ruk

-0.61

ugen

-0.60

POSITIVE LOGITS

Queue

0.75

bone

0.69

upon

0.69

hook

0.66

 Spoon

0.64

ÄŁ

0.63

bidden

0.63

Place

0.62

 implicitly

0.62

osphere

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.