INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Directory

-0.79

nered

-0.68

eland

-0.66

ety

-0.66

 Zhou

-0.65

otle

-0.65

edom

-0.63

Sky

-0.63

crim

-0.63

leigh

-0.63

POSITIVE LOGITS

ģĸ

0.72

acers

0.69

 vaccinations

0.64

ABE

0.63

*:

0.63

 privat

0.62

 Nutr

0.62

KS

0.61

 tiss

0.61

rot

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.