INDEX

Explanations

phrases related to rules, regulations, and restrictions

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

 impractica

-0.74

 unlaw

-0.73

 belliger

-0.66

 desertion

-0.66

 disgra

-0.66

 Kün

-0.65

 revan

-0.64

 cytoplas

-0.63

 impra

-0.63

 unemp

-0.60

POSITIVE LOGITS

 restrictions

1.08

 restriction

1.06

 restrict

1.00

 restricted

0.97

restricted

0.94

 restricting

0.94

restrict

0.93

 Restriction

0.91

 Restrictions

0.89

restriction

0.88

Activations Density 0.078%

phrases related to rules, regulations, and restrictions

No Comments

No Known Activations