INDEX

Explanations

mathematical function notation and structure

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

f(variable) notation

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

 الحره

-0.65

 Photocase

-0.64

 betweenstory

-0.50

ISupport

-0.50

awtextra

-0.49

')))

-0.49

ContextCompat

-0.47

 CanadaChoose

-0.47

}}%

-0.47

httphttps

-0.47

POSITIVE LOGITS

󠁴

0.44

 húmedo

0.42

밌

0.41

raires

0.41

ercizio

0.40

blic

0.38

lī

0.38

ambién

0.37

醐

0.37

 sillas

0.36

Activations Density 0.303%

mathematical function notation and structure

f(variable) notation

No Comments

No Known Activations