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

xit

-0.72

 hashing

-0.63

fro

-0.63

caster

-0.62

 realize

-0.62

 modeling

-0.62

 refin

-0.61

tsky

-0.61

 apprentice

-0.60

 ende

-0.60

POSITIVE LOGITS

ëĭ

0.74

Gra

0.70

mith

0.70

ļéĨĴ

0.65

Kak

0.64

Jump

0.64

collection

0.63

ĨĴ

0.62

Ide

0.62

 Wolves

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.