INDEX

Explanations

phrases related to belief and credibility in testimonies

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

!*\

-0.71

dafx

-0.67

-0.66

WebVitals

-0.59

 كمان

-0.58

IContainer

-0.58

RectangleBorder

-0.57

 nakalista

-0.57

Hochspringen

-0.57

AndroidJUnit

-0.56

POSITIVE LOGITS

 trustworthy

0.70

 trustworthiness

0.62

 confiable

0.58

 credible

0.56

 verlä

0.56

 trusted

0.55

zionalità

0.53

 confiance

0.52

 veracity

0.52

trusted

0.51

Activations Density 0.133%

phrases related to belief and credibility in testimonies

No Comments

No Known Activations