INDEX

Explanations

terms associated with suspicion and suspicion of wrongdoing

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

 مشين

-0.76

ftagPool

-0.69

 Argo

-0.69

 utafitiHapana

-0.67

 Barak

-0.66

 referenties

-0.66

 Davi

-0.66

ArrowToggle

-0.65

 שוליים

-0.64

 ModelExpression

-0.64

POSITIVE LOGITS

picion

0.93

Sus

0.90

 suspected

0.88

 suspects

0.84

 suspect

0.84

 suspicion

0.82

sus

0.79

Sus

0.77

SUS

0.74

SPJ

0.73

Activations Density 0.006%

terms associated with suspicion and suspicion of wrongdoing

No Comments

No Known Activations