INDEX

Explanations

references to conflict or instances of hostility involving specific groups or individuals

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

Follows personal pronouns

np_acts-logits-general · gemini-2.0-flash

tried to

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 nonUne

-0.56

GIVEREF

-0.51

 argint

-0.45

 onPostExecute

-0.45

OOTDTY

-0.45

المناصب

-0.43

исленность

-0.42

ResponseWriter

-0.40

 Außer

-0.40

cocc

-0.40

POSITIVE LOGITS

 wireType

0.53

SharedCtor

0.44

RTGC

0.39

AntiForgeryToken

0.36

 insults

0.36

 estekak

0.35

CDCl

0.35

]++;

0.34

GraphicsUnit

0.34

 sní

0.34

Activations Density 0.360%

references to conflict or instances of hostility involving specific groups or individuals

Follows personal pronouns

tried to

No Comments

No Known Activations