INDEX

Explanations

terms associated with brightness and vividness

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

light or visual intensity

np_acts-logits-general · gemini-2.0-flash

shining bright

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yntaxException

-0.78

 protoimpl

-0.65

 ModelExpression

-0.64

complexContent

-0.60

 ब्रेकडाउन

-0.60

ритори

-0.58

WebMethod

-0.55

Closer

-0.55

 Normdatei

-0.54

Климат

-0.54

POSITIVE LOGITS

 shining

1.02

 brightly

1.01

 brightest

0.99

 brighter

0.98

 bright

0.97

Shining

0.95

 shine

0.95

 BRIGHT

0.94

 Bright

0.94

 shines

0.90

Activations Density 0.100%

terms associated with brightness and vividness

light or visual intensity

shining bright

No Comments

No Known Activations