INDEX

Explanations

symbols and special characters often used in formatting or structuring data

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

separator followed by non-English words

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

 betweenstory

-0.51

 autorytatywna

-0.48

multer

-0.45

 kasarigan

-0.44

LLocation

-0.44

stdc

-0.43

 onPostExecute

-0.41

存于互联网档案馆

-0.41

yelitis

-0.41

 newBuilder

-0.40

POSITIVE LOGITS

 مشين

0.43

 maternidade

0.40

 Ministério

0.39

 piú

0.39

 lunak

0.39

 personaggio

0.38

 municipalité

0.38

 tradição

0.38

 signore

0.37

 démocratie

0.37

Activations Density 0.198%

symbols and special characters often used in formatting or structuring data

separator followed by non-English words

No Comments

No Known Activations