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

jriwal

-0.72

 resur

-0.71

ABE

-0.70

 mainline

-0.65

ļéĨĴ

-0.63

¥ŀ

-0.63

 captcha

-0.63

peg

-0.63

 curtains

-0.63

 privat

-0.62

POSITIVE LOGITS

rats

0.79

rix

0.72

olves

0.72

entin

0.72

matter

0.69

bell

0.68

elta

0.68

viol

0.67

ENSE

0.66

intent

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.