INDEX

Explanations

specific medical or technical terms and concepts

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

foreign language terms and technical types

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

InitVars

-0.52

amental

-0.47

VYMaps

-0.47

хьтан

-0.47

OGND

-0.46

 تضيفلها

-0.46

 Exactos

-0.45

Autoritní

-0.43

arms

-0.42

出版年

-0.42

POSITIVE LOGITS

 betweenstory

0.40

gående

0.39

MLLoader

0.38

quelize

0.38

 dép

0.38

 quedando

0.37

Brainz

0.36

 мер

0.36

kej

0.35

Przyp

0.34

Activations Density 0.169%

specific medical or technical terms and concepts

foreign language terms and technical types

No Comments

No Known Activations