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

olean

-0.79

uyomi

-0.71

iership

-0.69

hai

-0.68

 guiIcon

-0.67

harm

-0.65

ilic

-0.64

 holiest

-0.63

atomic

-0.62

opausal

-0.61

POSITIVE LOGITS

luaj

0.67

atos

0.65

olver

0.64

DOC

0.63

ourcing

0.62

 captcha

0.61

igator

0.61

 Predators

0.60

document

0.59

DET

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.