INDEX

Explanations

occurrences of the word "our."

oai_token-act-pair · gpt-3.5-turbo

references to a collective identity or ownership

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

edit

-0.70

Trivia

-0.65

ysis

-0.57

puff

-0.55

ãĥĹ

-0.55

achy

-0.55

.[

-0.53

ullivan

-0.53

ynes

-0.52

ibliography

-0.50

POSITIVE LOGITS

our

2.94

 ourselves

2.45

 ours

2.30

OUR

2.15

Our

2.08

Our

1.91

us

1.67

our

1.44

we

1.39

my

1.34

Activations Density 0.103%

occurrences of the word "our."

references to a collective identity or ownership

No Comments

No Known Activations