INDEX

Explanations

phrases related to collaboration and sharing in a team context

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

shared by friends or teammates

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

Gön

-0.40

 цвете

-0.38

 FormGroup

-0.37

komendasi

-0.36

VYMaps

-0.35

AntiForgeryToken

-0.33

 avoient

-0.33

Disliked

-0.32

колеп

-0.31

 लगा

-0.31

POSITIVE LOGITS

出版年

0.59

MLLoader

0.54

ReusableCell

0.53

PMailer

0.52

RegressionTest

0.52

 collaborators

0.52

 teammates

0.51

expandindo

0.51

IUrlHelper

0.48

İstinadlar

0.47

Activations Density 0.023%

phrases related to collaboration and sharing in a team context

shared by friends or teammates

No Comments

No Known Activations