INDEX

Explanations

words related to legal and government contexts

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

before named entities or locations

np_acts-logits-general · gemini-2.0-flash

logistic regression analyses

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-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 createState

-0.58

таратура

-0.56

 برانيه

-0.54

 uintptr

-0.51

}))

-0.50

ंदीखरीदारी

-0.47

ally

-0.47

ledem

-0.47

 références

-0.47

 MonoBehaviour

-0.45

POSITIVE LOGITS

$_"

0.78

TagMode

0.72

UrlResolution

0.71

SuppressMessage

0.67

\<^

0.67

 photolibrary

0.65

newtheorem

0.64

cherichia

0.64

ogni

0.63

bonucleic

0.63

Activations Density 1.108%

words related to legal and government contexts

before named entities or locations

logistic regression analyses

No Comments

No Known Activations