INDEX

Explanations

terms related to health and medical research

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

terms related to healthcare treatment and research

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

technical terms followed by context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 militaires

-0.51

 religieuse

-0.48

 agrícolas

-0.47

 industriels

-0.44

 Púb

-0.43

 quelcon

-0.42

macam

-0.42

 venons

-0.42

 pubblica

-0.41

 prisonniers

-0.41

POSITIVE LOGITS

 ModelExpression

1.18

OGND

1.14

SourceChecksum

1.02

GEBURTSDATUM

0.99

NameInMap

0.98

+#+#

0.97

 Biôgrafia

0.94

Kjelder

0.91

Personensuche

0.91

AccessorTable

0.89

Activations Density 0.802%

terms related to health and medical research

terms related to healthcare treatment and research

technical terms followed by context

No Comments

No Known Activations