INDEX

Explanations

phrases related to debate and speculation

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

expressed opinions or claims

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

 queſta

-0.73

 مشين

-0.68

EDEFAULT

-0.66

 EconPapers

-0.65

ChildScrollView

-0.65

CppMethod

-0.64

 surla

-0.62

parsedMessage

-0.61

 houſe

-0.61

typeorm

-0.60

POSITIVE LOGITS

 katkıda

0.31

이

0.29

 posibles

0.28

bahaya

0.28

minyak

0.28

 agrav

0.27

rások

0.27

 Mindest

0.26

 pomoci

0.26

jména

0.26

Activations Density 0.055%

phrases related to debate and speculation

expressed opinions or claims

No Comments

No Known Activations