INDEX

Explanations

key names, titles, and significant terms related to specific narratives or discussions

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

numbers and specific terms

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

 Italijani

-0.53

 للمعارف

-0.53

webElementXpaths

-0.52

 للاسماء

-0.51

乓

-0.50

LookAnd

-0.50

 HttpNotFound

-0.50

ViewImports

-0.49

 JpaRepository

-0.49

AddTagHelper

-0.48

POSITIVE LOGITS

DockStyle

0.43

 وتسجيلات

0.38

wyr

0.34

発

0.33

theria

0.33

consulté

0.32

AZA

0.32

 ویکی‌پدیا

0.32

śle

0.32

 chers

0.32

Activations Density 0.326%

key names, titles, and significant terms related to specific narratives or discussions

numbers and specific terms

No Comments

No Known Activations