INDEX

Explanations

information about starting and managing a business

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

addressing you personally

np_acts-logits-general · gemini-2.5-flash-lite

This neuron detects second-person direct address — tokens in segments that speak to or describe "you" (the reader) in instructions or scenarios.

oai_token-act-pair · gpt-5-mini

Triggered by @vetterc0

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

 <<<<<<<<<<<<<<

-0.37

__(/*!

-0.36

AccessorTable

-0.33

Hozzáférés

-0.33

resave

-0.32

Kanpo

-0.30

 المعيارى

-0.30

cionario

-0.29

Enllaces

-0.29

.*")]

-0.29

POSITIVE LOGITS

 yourself

0.56

 informée

0.55

MLLoader

0.52

 CreateTagHelper

0.50

 SwitchCompat

0.49

OGND

0.48

InjectMocks

0.48

AndroidJUnit

0.47

ValueGeneration

0.47

GRANTED

0.47

Activations Density 0.078%

information about starting and managing a business

addressing you personally

This neuron detects second-person direct address — tokens in segments that speak to or describe "you" (the reader) in instructions or scenarios.

No Known Activations

information about starting and managing a business

addressing you personally

This neuron detects second-person direct address — tokens in segments that speak to or describe "you" (the reader) in instructions or scenarios.

No Known Activations