INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Adin

-0.75

 Resident

-0.66

 ILCS

-0.65

risome

-0.63

Yon

-0.62

 Halo

-0.62

gee

-0.62

 CODE

-0.62

MLG

-0.62

pee

-0.61

POSITIVE LOGITS

acid

0.71

iers

0.68

uter

0.66

fired

0.66

asel

0.65

ometimes

0.65

eting

0.64

tube

0.63

street

0.62

iesel

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.