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

 homebrew

-0.61

quire

-0.60

okÃ©

-0.59

vable

-0.57

 Randy

-0.57

 Vend

-0.56

cloth

-0.56

 Eisen

-0.56

 unbeliev

-0.56

 Drake

-0.56

POSITIVE LOGITS

tips

0.69

shr

0.69

 Â®

0.68

zzi

0.67

ITNESS

0.65

ridges

0.64

opa

0.62

aceutical

0.62

helle

0.62

letters

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.