INDEX

Explanations

different entities or titles related to media and entertainment

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

about [topic description]

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

 تضيفلها

-1.02

AddTagHelper

-0.75

новниш

-0.71

 Normdatei

-0.69

 Wicidata

-0.69

'{@

-0.68

featureID

-0.68

 snippetHide

-0.68

]")]

-0.66

 kasarigan

-0.65

POSITIVE LOGITS

About

0.55

 About

0.52

 ABOUT

0.49

ABOUT

0.46

 about

0.46

about

0.45

 acerca

0.35

 बारे

0.35

 yourself

0.34

 Acerca

0.32

Activations Density 0.029%

different entities or titles related to media and entertainment

about [topic description]

No Comments

No Known Activations