INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Canaver

-1.00

INTON

-0.83

 Williamson

-0.79

 Feld

-0.70

 Coulter

-0.69

 Akin

-0.68

 McCull

-0.68

Discussion

-0.67

Rush

-0.67

 Chiefs

-0.66

POSITIVE LOGITS

kered

0.79

isine

0.72

 pengu

0.69

ophobia

0.69

elling

0.68

è¯

0.68

à¸

0.66

 snowball

0.64

vable

0.64

urai

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.