INDEX

Explanations

elements related to data visualization and plotting in programming

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

user input identifier

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

 ब्रेकडाउन

-0.74

OGND

-0.73

цездатний

-0.65

 PeEnEo

-0.63

 autorytatywna

-0.59

quelize

-0.59

 betweenstory

-0.58

 Roskov

-0.57

 архивлан

-0.56

Diweddarwch

-0.56

POSITIVE LOGITS

 tight

0.41

HEIM

0.34

 ajust

0.31

WebElementEntity

0.31

 acord

0.30

ViewSet

0.30

 tasteful

0.29

 diken

0.29

HasKey

0.29

 clearly

0.29

Activations Density 0.009%

elements related to data visualization and plotting in programming

user input identifier

No Comments

No Known Activations