INDEX

Explanations

mathematical and logical relationships in structured data

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

in a set or location

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

 ModelExpression

-0.65

featureID

-0.63

parsedMessage

-0.58

sidemargin

-0.54

 &___

-0.51

ValueGeneration

-0.50

fromnode

-0.50

DockStyle

-0.49

 Administrativna

-0.48

ReusableCell

-0.48

POSITIVE LOGITS

 primary

0.37

study

0.34

годня

0.34

Aiheesta

0.34

Ms

0.33

 primaries

0.33

 pino

0.33

here

0.33

новништво

0.32

primary

0.32

Activations Density 0.123%

mathematical and logical relationships in structured data

in a set or location

No Comments

No Known Activations

mathematical and logical relationships in structured data

in a set or location

No Comments

No Known Activations