INDEX

Explanations

important nouns and entities, particularly those indicating roles or descriptions in various contexts

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

اريخ

-0.60

 Condiciones

-0.60

 vectorielles

-0.58

forRoot

-0.56

حيح

-0.56

 privées

-0.53

chown

-0.52

 supérieures

-0.52

 ouvertes

-0.52

 Less

-0.51

POSITIVE LOGITS

 gynhyrchwyd

0.74

AutoresizingMask

0.72

$}}

0.64

StructEnd

0.62

lava

0.61

 protoimpl

0.60

]}

0.60

 فريبيس

0.57

ErrUnexpectedEOF

0.57

expandindo

0.57

Activations Density 0.043%

important nouns and entities, particularly those indicating roles or descriptions in various contexts

No Comments

No Known Activations