INDEX

Explanations

profane language and related expressions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Fichier

-0.60

 Bourgoin

-0.56

 nyeri

-0.56

lã

-0.55

 kemer

-0.55

 mena

-0.54

 meras

-0.53

República

-0.52

gico

-0.52

ípio

-0.52

POSITIVE LOGITS

 shewn

1.05

 pamph

1.03

 shit

0.98

 viciss

0.98

 disagre

0.94

 paradiso

0.92

 tolerably

0.92

 SHIT

0.90

 impractica

0.90

 Shit

0.89

Activations Density 0.102%

profane language and related expressions

No Comments

No Known Activations