INDEX

Explanations

references to website features and their impact on user experience and interaction

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

followed by benefits or advantages

np_acts-logits-general · gemini-2.0-flash

benefit or help

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

ҥ

-0.49

interopRequire

-0.48

 disambiguazione

-0.48

+#+#

-0.46

 devrez

-0.45

 gara

-0.44

rå

-0.44

 литров

-0.44

ళ

-0.44

ष्ठ

-0.44

POSITIVE LOGITS

 helps

1.41

 Helps

1.22

helps

1.19

 help

1.19

Helps

1.18

 improves

1.14

 assists

1.02

 enhances

1.01

 aiuta

1.01

 aids

1.00

Activations Density 0.392%

references to website features and their impact on user experience and interaction

followed by benefits or advantages

benefit or help

No Comments

No Known Activations