INDEX

Explanations

references to quantifiable entities or units of measurement in a data context

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

number of entities

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

 viewDidLoad

-0.52

 zijne

-0.50

randomUUID

-0.49

insegna

-0.48

 femininas

-0.47

 very

-0.47

lant

-0.47

Captor

-0.47

 împre

-0.47

 Gestão

-0.46

POSITIVE LOGITS

 numberOfRows

0.50

MessageOf

0.42

chartInstance

0.40

Архівовано

0.37

 onData

0.35

numRows

0.35

 available

0.35

Demografie

0.34

ścioła

0.34

 than

0.34

Activations Density 0.082%

references to quantifiable entities or units of measurement in a data context

number of entities

No Comments

No Known Activations