INDEX

Explanations

occurrences of logical concepts or boolean values in coding contexts

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

logical True

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

 Schluss

-0.47

 limitada

-0.41

 Schall

-0.40

XMLSchema

-0.37

 Persönlichkeit

-0.36

 extranjera

-0.36

 antaranya

-0.36

 Nutzung

-0.36

 Europäischen

-0.36

 conjunta

-0.36

POSITIVE LOGITS

disposing

0.75

الحياه

0.70

ंदीखरीदारी

0.69

 defaultstate

0.68

 esternos

0.64

 nahilalakip

0.60

FunctionFlags

0.59

 kasarigan

0.59

 tartalomajánló

0.59

Obrázky

0.59

Activations Density 0.000%

occurrences of logical concepts or boolean values in coding contexts

logical True

No Comments

No Known Activations