INDEX

Explanations

numbers and mathematical classifications or relationships

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

classification codes and numbers

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

WebElementEntity

-0.71

 InputDecoration

-0.51

 ſta

-0.51

mybatisplus

-0.50

 للمعارف

-0.50

Παραπομπές

-0.47

 faſt

-0.47

şört

-0.47

 canst

-0.45

 queſta

-0.45

POSITIVE LOGITS

 ujednoznacz

0.44

arXiv

0.39

.,

0.37

 aggiorn

0.35

 Мексичка

0.35

LookAnd

0.34

 Bidang

0.34

 Italijanski

0.33

 topical

0.32

 topics

0.32

Activations Density 0.004%

numbers and mathematical classifications or relationships

classification codes and numbers

No Comments

No Known Activations