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

allo

-0.69

 Tanaka

-0.69

PLIED

-0.68

 appreciated

-0.65

SourceFile

-0.65

 variability

-0.64

 outstanding

-0.64

 Canaver

-0.64

 Origin

-0.63

 Tenn

-0.62

POSITIVE LOGITS

pless

0.78

olph

0.72

ipher

0.72

NetMessage

0.71

merce

0.70

nown

0.69

ettings

0.68

ixt

0.66

liga

0.66

edd

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.