INDEX

Explanations

structured data and markup in a document

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

list syntax and structure

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

zzleHttp

-0.87

الحياه

-0.78

⤹

-0.70

endphp

-0.69

GTCX

-0.69

Rhestr

-0.68

DNEY

-0.67

PreferredItem

-0.67

 تضيفلها

-0.67

dafx

-0.66

POSITIVE LOGITS

træ

0.39

 Kerajaan

0.35

 negru

0.35

 serez

0.34

 dejan

0.33

jske

0.33

 culturelle

0.32

 précédemment

0.32

 collè

0.32

 recevrez

0.32

Activations Density 0.013%

structured data and markup in a document

list syntax and structure

No Comments

No Known Activations