INDEX

Explanations

phrases indicating various degrees of effort, condition, or emphasis in different contexts

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

specifically foreign nationalities

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

IntoConstraints

-0.67

 كومونز

-0.62

OrNil

-0.60

nace

-0.58

 shields

-0.57

cristo

-0.57

ilist

-0.56

 ſeines

-0.56

 ſua

-0.56

URLException

-0.54

POSITIVE LOGITS

 cherchés

0.54

 Italijanski

0.37

 régimen

0.34

 griega

0.33

 extracto

0.29

Panamoan

0.28

ๆ

0.28

WebElementEntity

0.28

 hacerlo

0.27

 تكبرها

0.26

Activations Density 0.053%

phrases indicating various degrees of effort, condition, or emphasis in different contexts

specifically foreign nationalities

No Comments

No Known Activations