INDEX

Explanations

the word "you" and phrases related to interactions or instructions involving the reader

oai_token-act-pair · gpt-3.5-turbo

references to the pronoun "you."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 speculation

-0.61

 cent

-0.59

dec

-0.59

 Central

-0.58

 base

-0.58

 foot

-0.58

fr

-0.57

st

-0.57

 median

-0.57

 critical

-0.56

POSITIVE LOGITS

you

3.56

your

2.10

YOU

2.06

You

1.68

yout

1.65

they

1.62

everyone

1.48

yeah

1.48

there

1.44

yo

1.39

Activations Density 0.012%

the word "you" and phrases related to interactions or instructions involving the reader

references to the pronoun "you."

No Comments

No Known Activations