INDEX

Explanations

pronouns and possessive words

oai_token-act-pair · gpt-3.5-turbo

references to individuals and their roles in various scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

venge

-0.80

 righteous

-0.69

 GOODMAN

-0.68

 triumphant

-0.65

STON

-0.64

liber

-0.64

irming

-0.63

ovy

-0.63

 congratulations

-0.63

knit

-0.63

POSITIVE LOGITS

 lacked

1.84

 lacks

1.62

 failed

1.54

 cannot

1.51

 forgot

1.46

 underestimated

1.42

 refused

1.39

 incorrectly

1.38

failed

1.38

 couldn

1.37

Activations Density 0.730%

pronouns and possessive words

references to individuals and their roles in various scenarios

No Comments

No Known Activations