INDEX

Explanations

technical or programming-related instructions and parameters

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

Geometric shapes and markers

np_acts-logits-general · gemini-2.0-flash

filled, open, solid, dashed markers and lines

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IonicModule

-0.69

IUrlHelper

-0.67

RenderAtEndOf

-0.66

ViewFeatures

-0.65

 المعيارى

-0.65

windowFixed

-0.64

addCriterion

-0.60

interopRequire

-0.60

mergeFrom

-0.60

TagHelper

-0.59

POSITIVE LOGITS

 solid

0.65

 filled

0.54

filled

0.52

线

0.52

solid

0.51

lin

0.51

 пун

0.50

 dashed

0.50

 SOLID

0.50

Solid

0.50

Activations Density 0.299%

technical or programming-related instructions and parameters

Geometric shapes and markers

filled, open, solid, dashed markers and lines

No Comments

No Known Activations