INDEX

Explanations

phrases that indicate relationships or connections between entities, especially in financial or material contexts

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

expiration its other shares

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

hloro

-0.34

yra

-0.34

 panorama

-0.33

DIT

-0.33

cinta

-0.33

 focus

-0.33

希望

-0.32

 Disclosure

-0.32

стоин

-0.32

XCTAssertEqual

-0.32

POSITIVE LOGITS

 noDo

0.73

DockStyle

0.73

الحياه

0.64

новништво

0.64

0.60

tagHelperRunner

0.57

webElementGuid

0.54

 autorytatywna

0.53

expandindo

0.53

ContentAsync

0.52

Activations Density 0.079%

phrases that indicate relationships or connections between entities, especially in financial or material contexts

expiration its other shares

No Comments

No Known Activations