INDEX

Explanations

references to specific individuals and their roles within a community or organization

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

requests to solicit or inform

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

̈́

-0.37

CMC

-0.36

vr

-0.31

opre

-0.31

bestimm

-0.31

ograft

-0.30

cmc

-0.30

 Gesetze

-0.29

 jamás

-0.29

 Jahrgang

-0.29

POSITIVE LOGITS

WarningLevel

0.58

 Италијани

0.54

msgTypes

0.53

 contingency

0.52

 يتيمه

0.52

initializeApp

0.51

 Deceased

0.51

solicited

0.50

Hentet

0.50

ModelSerializer

0.50

Activations Density 0.008%

references to specific individuals and their roles within a community or organization

requests to solicit or inform

No Comments

No Known Activations