INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

kaya

-0.89

igun

-0.70

Posted

-0.70

ï¸

-0.68

\<

-0.66

hetical

-0.63

istical

-0.63

istani

-0.62

hip

-0.61

ÅŁ

-0.61

POSITIVE LOGITS

 millenn

0.81

maxwell

0.68

Ã©r

0.68

ãĤ¦ãĤ¹

0.66

éŃĶ

0.65

 tiss

0.65

 Eaton

0.64

 mathemat

0.64

 Afric

0.64

 mosqu

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.