INDEX

Explanations

mentions of people holding specific positions or affiliations in various organizations

oai_token-act-pair · gpt-3.5-turbo

articles frequently associated with notable figures and their roles or attributes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

andon

-0.66

mares

-0.64

oops

-0.63

fixes

-0.63

ifts

-0.63

words

-0.62

osity

-0.62

anism

-0.62

idi

-0.62

girls

-0.60

POSITIVE LOGITS

 member

1.25

 contributor

1.12

 supporter

1.09

 proponent

1.07

 fixture

1.07

 participant

1.06

 frequent

1.05

 trustee

1.03

 collaborator

0.98

 shareholder

0.98

Activations Density 0.159%

mentions of people holding specific positions or affiliations in various organizations

articles frequently associated with notable figures and their roles or attributes

No Comments

No Known Activations