INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

mitt

-0.67

ãĤ°

-0.65

eps

-0.65

ãĤª

-0.62

alog

-0.62

ales

-0.62

psy

-0.60

imgur

-0.60

posts

-0.60

FAQ

-0.60

POSITIVE LOGITS

yip

0.89

 metic

0.86

ividual

0.85

ĪĴ

0.76

 ignition

0.72

iership

0.70

iosyncr

0.70

illac

0.68

 Colleges

0.67

atchewan

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.