INDEX

Explanations

threatening language and accusations in a confrontational context

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

 ?...

-1.24

fuf

-1.22

 emphat

-1.19

 reluct

-1.16

fta

-1.16

 inev

-1.14

 strick

-1.14

 gend

-1.12

 guarante

-1.11

fte

-1.11

POSITIVE LOGITS

 ***!

0.60

avril

0.59

testify

0.56

 YOUR

0.54

YOU

0.54

oredCriteria

0.53

 your

0.53

 please

0.52

your

0.52

you

0.52

Activations Density 0.505%

threatening language and accusations in a confrontational context

No Comments

No Known Activations