INDEX

Explanations

the term "Fr" followed by a single character or specific word

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Fran" or its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eers

-0.87

 Twain

-0.80

manship

-0.74

 Izan

-0.73

 Occupations

-0.71

phrine

-0.69

ership

-0.67

WARE

-0.66

anwhile

-0.65

eer

-0.65

POSITIVE LOGITS

aternity

1.11

inge

1.08

aternal

1.08

umpy

1.02

atell

1.00

antic

0.98

antz

0.97

unker

0.97

atern

0.95

isco

0.95

Activations Density 0.009%

the term "Fr" followed by a single character or specific word

mentions of the name "Fran" or its variations in different contexts

No Comments

No Known Activations