INDEX

Explanations

phrases indicating relationships, conditions, or components related to various subjects

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

questions and lists

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

Negative Logits

rungsseite

-0.61

Autoritní

-0.52

 gyhoeddwyd

-0.48

 nahilalakip

-0.47

 مرئيه

-0.47

 متعلقه

-0.47

发表于

-0.47

matchCondition

-0.46

GEBURTSDATUM

-0.45

ViewImports

-0.44

POSITIVE LOGITS

setVerticalGroup

0.54

stateProvider

0.41

setCellStyle

0.36

coration

0.35

IComponent

0.35

Assista

0.35

cookieParser

0.34

 faveur

0.34

InitStruct

0.33

AddTagHelper

0.32

Activations Density 0.282%

phrases indicating relationships, conditions, or components related to various subjects

questions and lists

No Known Activations