INDEX

Explanations

words related to people's names or titles

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

咤

-0.50

utiva

-0.47

GraphicsUnit

-0.47

PhysRevD

-0.46

Copia

-0.46

pañas

-0.44

eezy

-0.44

lcccccc

-0.44

Chham

-0.43

latimes

-0.43

POSITIVE LOGITS

 affor

0.99

 exé

0.95

 soigne

0.93

 increa

0.91

 reconnaît

0.86

 prétend

0.86

yaa

0.85

 préfère

0.84

GOR

0.84

 ricor

0.83

Activations Density 0.274%

words related to people's names or titles

No Comments

No Known Activations