INDEX

Explanations

references to different religious affiliations and conditions related to mental health

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

specific concepts

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

AndEndTag

-0.57

HtmlAttribute

-0.53

FieldBuilder

-0.49

 arşivlendi

-0.49

queryInterface

-0.48

 resourceCulture

-0.44

dropIfExists

-0.43

tagHelperRunner

-0.43

uginosa

-0.42

entyfik

-0.42

POSITIVE LOGITS

IconModule

0.40

 challenge

0.39

 AspNetCore

0.37

Mog

0.37

 Challenging

0.36

lioz

0.36

mögens

0.36

 Chasse

0.36

Lan

0.36

:+:

0.36

Activations Density 0.130%

references to different religious affiliations and conditions related to mental health

specific concepts

No Comments

No Known Activations