INDEX

Explanations

the pronoun "I" at a high activation level

oai_token-act-pair · gpt-3.5-turbo

first-person singular pronouns

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

rules

-0.62

 apex

-0.62

 excess

-0.62

 Rapids

-0.61

Ħ¢

-0.60

 Clair

-0.60

ses

-0.60

 Shelby

-0.58

 Georgetown

-0.56

 Slav

-0.56

POSITIVE LOGITS

'm

1.35

've

1.18

nex

1.14

EEE

1.06

 suppose

1.06

'll

1.03

stanbul

1.03

 dunno

1.03

WI

1.02

zzy

0.98

Activations Density 0.141%

the pronoun "I" at a high activation level

first-person singular pronouns

No Comments

No Known Activations