INDEX

Explanations

instances where possessive pronouns "your" or "you" are followed by an action or directive

oai_token-act-pair · gpt-3.5-turbo

the word "your."

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

ALE

-0.66

ZA

-0.64

IX

-0.60

NetMessage

-0.58

ISM

-0.58

ãĤ´ãĥ³

-0.55

ãģ®éŃĶ

-0.55

EVA

-0.55

cknowled

-0.54

 Reconstruction

-0.54

POSITIVE LOGITS

 your

2.83

your

2.45

 YOUR

2.32

 Your

2.13

 yours

2.12

Your

2.10

 yourself

2.09

 yourselves

1.99

 Yourself

1.43

thy

1.41

Activations Density 0.107%

instances where possessive pronouns "your" or "you" are followed by an action or directive

the word "your."

No Comments

No Known Activations