INDEX

Explanations

programming constructs related to message and enum definitions in a specific syntax

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

source._repeated

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

EndContext

-0.64

TestBed

-0.60

存于互联网档案馆

-0.60

Havolalar

-0.57

rungsseite

-0.55

 <>",

-0.55

LookAnd

-0.54

rawDesc

-0.53

новништво

-0.52

="@+

-0.52

POSITIVE LOGITS

culable

0.39

dė

0.38

 kautta

0.38

leville

0.34

 remplir

0.34

 umě

0.34

 jäsen

0.33

ypeł

0.33

 enemmän

0.32

 aikaa

0.31

Activations Density 0.157%

programming constructs related to message and enum definitions in a specific syntax

source._repeated

No Comments

No Known Activations