INDEX

Explanations

time-related terms indicating duration such as "months," "weeks," "years," and "hours."

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

 Ename

-1.06

 indestru

-1.03

 depic

-1.02

 unden

-1.01

 madonna

-1.00

 Manufact

-1.00

 affor

-0.99

 embodi

-0.99

 Décembre

-0.98

 pymysql

-0.97

POSITIVE LOGITS

 months

0.85

 weeks

0.77

 month

0.70

 longer

0.68

 years

0.67

months

0.66

 decades

0.64

decade

0.64

 days

0.59

 indefinitely

0.59

Activations Density 0.211%

time-related terms indicating duration such as "months," "weeks," "years," and "hours."

No Comments

No Known Activations