INDEX

Explanations

specific names of publishers and important terms related to educational content

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

publisher names and prefaces

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

obacterium

-0.59

nith

-0.59

outine

-0.56

ctools

-0.56

 Fisch

-0.55

Bim

-0.55

document

-0.54

choon

-0.54

stanz

-0.54

Bim

-0.52

POSITIVE LOGITS

})*/

0.55

 ervaren

0.48

 getLayout

0.47

StoreMessageInfo

0.47

PREFACE

0.47

 piernas

0.46

 alturas

0.46

 tewas

0.46

 getAge

0.46

 leyendas

0.45

Activations Density 0.005%

specific names of publishers and important terms related to educational content

publisher names and prefaces

No Comments

No Known Activations