INDEX

Explanations

specific protein names and related terms

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

Sofia tells

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

 kasarigan

-0.54

outchou

-0.47

clusal

-0.45

 tropicales

-0.45

limard

-0.44

ffindor

-0.43

Demografía

-0.43

jspx

-0.42

goku

-0.42

GeneratedValue

-0.42

POSITIVE LOGITS

 tartalomajánló

0.65

 jsPsych

0.51

 виправивши

0.51

 NUKAT

0.49

 esternos

0.47

IndentedString

0.44

 censiti

0.43

xnn

0.40

 سكانية

0.40

 ddelweddau

0.40

Activations Density 0.215%

specific protein names and related terms

Sofia tells

No Comments

No Known Activations