INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

wart

-0.93

sov

-0.73

dayName

-0.72

ãĢı

-0.70

itious

-0.70

ItemThumbnailImage

-0.70

src

-0.69

-|

-0.69

paren

-0.69

href

-0.68

POSITIVE LOGITS

 Upton

0.77

hare

0.70

 dead

0.67

 jaws

0.67

NCT

0.63

 Gore

0.62

 catastrophe

0.61

 thriller

0.60

 defect

0.59

ufact

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.