INDEX

Explanations

phrases related to admiration, respect, sympathy, message content, and judgment

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 maksi

-0.96

 territo

-0.95

uhr

-0.95

ivi

-0.94

 embra

-0.93

 accla

-0.91

 dises

-0.89

 saar

-0.88

 sena

-0.88

 endom

-0.87

POSITIVE LOGITS

 TextAppearance

0.52

hasPermission

0.51

 nodig

0.51

ffilm

0.50

Personensuche

0.49

styleType

0.48

 beginnetje

0.48

 بيها

0.48

LayoutConstraint

0.46

woordig

0.46

Activations Density 0.336%

phrases related to admiration, respect, sympathy, message content, and judgment

No Comments

No Known Activations