INDEX

Explanations

instances of manual or hands-on activities

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

from scratch or by hand

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 avoient

-0.52

 chrétien

-0.48

 hendes

-0.45

了嗎

-0.44

 poslední

-0.43

 secundario

-0.43

 navideño

-0.43

 tarvit

-0.43

 pengh

-0.43

 campamento

-0.42

POSITIVE LOGITS

AtIndexPath

0.66

databind

0.63

asList

0.61

toArray

0.60

存于互联网档案馆

0.60

ToScroll

0.60

ISupport

0.58

 JAXBElement

0.58

 BytesLike

0.57

randomUUID

0.56

Activations Density 0.143%

instances of manual or hands-on activities

from scratch or by hand

No Comments

No Known Activations