INDEX

Explanations

specific patterns in alphanumeric sequences that may correspond to coding or categorization

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

 '\\;'

-0.92

 GenerationType

-0.75

 صوتيه

-0.73

queryInterface

-0.69

 useAppContext

-0.68

getDescriptor

-0.68

 متعلقه

-0.68

Numerology

-0.68

setVerticalGroup

-0.66

spender

-0.66

POSITIVE LOGITS

 geworden

0.56

non

0.56

Ữ

0.49

uns

0.48

 itself

0.48

lui

0.48

 himself

0.47

 пото

0.47

 series

0.47

Activations Density 0.054%

specific patterns in alphanumeric sequences that may correspond to coding or categorization

No Comments

No Known Activations