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

rots

-0.70

iggins

-0.67

 IPCC

-0.67

heit

-0.64

FSA

-0.64

 Matthews

-0.64

ancial

-0.61

Downloadha

-0.60

Mac

-0.60

doors

-0.60

POSITIVE LOGITS

 BaseType

0.73

eering

0.70

¥µ

0.68

Jew

0.66

civ

0.66

 Puzz

0.64

Merit

0.64

%%%%

0.62

iazep

0.62

uin

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.