INDEX

Explanations

specific terms associated with specific brands or organizations

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

Eſ

-0.75

 photolibrary

-0.71

 Monfieur

-0.71

 myſelf

-0.70

 iſt

-0.67

 havaianas

-0.65

 dieß

-0.64

 itſelf

-0.64

 abbildung

-0.63

 ſche

-0.62

POSITIVE LOGITS

interopRequire

0.67

CppMethod

0.64

astify

0.63

InstrumentedTest

0.61



0.55

endpush

0.55

NewGuid

0.53

nocache

0.53

ESSOR

0.52

Revenir

0.52

Activations Density 0.161%

specific terms associated with specific brands or organizations

No Comments

No Known Activations