INDEX

Explanations

proper nouns, specifically names of people and locations

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

proper names

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

 palp

-0.52

 незавершена

-0.52

 transfieras

-0.51

󠁮

-0.51

 Belajar

-0.49

Außerdem

-0.48

Gruß

-0.48

 оригіналу

-0.47

 preju

-0.46

🎳

-0.46

POSITIVE LOGITS

kuuta

0.54

KommentareTeilen

0.50

expandindo

0.47

uxxxx

0.47

 صوتيه

0.45

 './../

0.45

aarrggbb

0.43

ţiile

0.41

LLocation

0.41

 Wiktionnaire

0.41

Activations Density 0.056%

proper nouns, specifically names of people and locations

proper names

No Comments

No Known Activations