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

 Dragonbound

-0.78

[+

-0.68

 pest

-0.63

athi

-0.63

Mek

-0.62

 neigh

-0.62

Gon

-0.61

WHERE

-0.60

asa

-0.60

iri

-0.59

POSITIVE LOGITS

mint

0.89

itely

0.78

ãĥ¼ãĥĨ

0.66

balls

0.66

rats

0.65

abal

0.64

GD

0.63

untled

0.63

 Balance

0.63

weed

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.