INDEX

Explanations

questions and inquiries about specific details or features

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Bath

-0.41

ALC

-0.40

 rumour

-0.39

Leg

-0.39

 Market

-0.39

 journey

-0.39

 market

-0.38

offerta

-0.38

 ceremony

-0.38

dh

-0.38

POSITIVE LOGITS

 what

0.72

what

0.58

how

0.58

 كومونز

0.54

What

0.52

 whats

0.52

 miten

0.51

 وتسجيلات

0.50

Welke

0.50

 welke

0.49

Activations Density 0.262%

questions and inquiries about specific details or features

No Comments

No Known Activations