INDEX

Explanations

references to adjustments and modifications in scientific or technical contexts

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

adjusting with water

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

脚注の使い方

-0.43

 ModelExpression

-0.43

 onPostExecute

-0.39

AdapterView

-0.38

Filmografie

-0.37

EndTag

-0.37

InstanceState

-0.37

glEnd

-0.37

ViewInit

-0.36

ISupport

-0.36

POSITIVE LOGITS

 defaultstate

0.53

 FILLER

0.51

rawDesc

0.50

filler

0.48

 filler

0.46

 fillers

0.45

Spacer

0.45

dummy

0.42

Hentet

0.42

 gynhyrchwyd

0.41

Activations Density 0.158%

references to adjustments and modifications in scientific or technical contexts

adjusting with water

No Comments

No Known Activations