INDEX

Explanations

terms associated with various categories of science, health, and specific technical fields

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

pain management

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

oredCriteria

-0.59

 AssemblyTitle

-0.55

舺

-0.54

 OFDb

-0.49

تقاوى

-0.49

 المعيارى

-0.48

astéro

-0.47

rungsseite

-0.47

 disambiguazione

-0.47



-0.45

POSITIVE LOGITS

AndEndTag

0.44

fortawesome

0.43

 referrerpolicy

0.41

ポート

0.38

 snippetHide

0.38

umnos

0.37

mum

0.36

fizierung

0.35

ärsk

0.35

migrationBuilder

0.35

Activations Density 0.159%

terms associated with various categories of science, health, and specific technical fields

pain management

No Comments

No Known Activations