INDEX

Explanations

references to biological processes and interactions

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

specific non-english words or technical terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 pouce

-0.30

<bos>

-0.29

 kaya

-0.26

 igjen

-0.26

gende

-0.25

又要

-0.24

 sanitarias

-0.24

 lieber

-0.24

 reward

-0.23

其他人

-0.23

POSITIVE LOGITS

 surla

0.93

таратура

0.78

 kasarigan

0.72

MLLoader

0.70

 فريبيس

0.68

TagMode

0.65

 Meksiku

0.64

AndroidJUnit

0.64

 EconPapers

0.63

 الرياضيه

0.62

Activations Density 0.622%

references to biological processes and interactions

specific non-english words or technical terms

No Comments

No Known Activations