INDEX

Explanations

mathematical symbols and formatting relevant to sequences and functions in mathematical expressions

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

mathematical formulas with indices and operations

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

parsedMessage

-0.73

BufferException

-0.66

 Италијани

-0.65

 мәкал

-0.58

 <=",

-0.57

 Савезне

-0.56

 kohdetta

-0.55



-0.55

 kasarigan

-0.54

 intptr

-0.54

POSITIVE LOGITS

paravant

0.36

 cahier

0.34

 literat

0.34

Colabor

0.34

 costado

0.34

 deuda

0.34

muerzo

0.33

 sidang

0.32

Weblinks

0.31

AsUp

0.31

Activations Density 0.050%

mathematical symbols and formatting relevant to sequences and functions in mathematical expressions

mathematical formulas with indices and operations

No Comments

No Known Activations