INDEX

Explanations

references to specific scientific or technical terms related to medical conditions, measurements, or research findings

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

units of measurement

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

 AssemblyCulture

-0.80

exitRule

-0.56

parsedMessage

-0.43

 disambiguazione

-0.43

 nahilalakip

-0.42

 Smol

-0.42

 Biôgrafia

-0.42

InitVars

-0.41

oredCriteria

-0.40

انيف

-0.39

POSITIVE LOGITS

túi

0.42

 ModelExpression

0.42

HandlerContext

0.39

 numerus

0.39

édrale

0.39

ษัท

0.38

のでは

0.37

ady

0.36

いざ

0.36

zzar

0.36

Activations Density 0.134%

references to specific scientific or technical terms related to medical conditions, measurements, or research findings

units of measurement

No Comments

No Known Activations