INDEX

Explanations

categories and references related to various topics or entities

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

references to categories and listings

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

Categories and links

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

saraba

-0.81

XmlAccessType

-0.79

adpleegd

-0.73

RSpec

-0.69

Erreferentziak

-0.66

Personendaten

-0.66

AlterField

-0.65

الدراسه

-0.64

nodoc

-0.64

ագրություններ

-0.63

POSITIVE LOGITS

<table>

0.59

 Categories

0.55

 Template

0.54

}\|

0.51

Template

0.51

EndInit

0.51

विक

0.49

↵↵↵↵↵

0.49

 Wikimedia

0.49

فيه

0.48

Activations Density 0.076%

categories and references related to various topics or entities

references to categories and listings

Categories and links

No Comments

No Known Activations