INDEX

Explanations

text written in a specific formatting style with unique characters interspersed

oai_token-act-pair · gpt-3.5-turbo

phrases that express strong emotions or opinions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

itably

-0.57

ussian

-0.56

anwhile

-0.54

 scatter

-0.54

 dust

-0.53

 scramble

-0.52

 Samar

-0.51

 Mayo

-0.51

 staggered

-0.50

 stagger

-0.50

POSITIVE LOGITS

catentry

0.84

¹

0.78

£

0.72

atures

0.68

bryce

0.68

¬

0.67

į

0.65

ı

0.64

drivers

0.63

cipline

0.63

Activations Density 0.645%

text written in a specific formatting style with unique characters interspersed

phrases that express strong emotions or opinions

No Comments

No Known Activations