INDEX

Explanations

specific technical terms and parameters related to programming and computer science

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

LaTeX code and file paths

np_acts-logits-general · gemini-2.0-flash

archaic german english

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 domestic

-0.34

 media

-0.33

انجليز

-0.31

bar

-0.31

 Comer

-0.31

 recent

-0.29

 dashed

-0.29

 Gart

-0.29

SE

-0.29

 minds

-0.29

POSITIVE LOGITS

$_"

0.56

ſelves

0.54

 uſed

0.54

 raiſ

0.53

 /\.(

0.52

 cauſe

0.52

 faſt

0.51

ſelf

0.51

 Majefty

0.51

Activations Density 2.399%

specific technical terms and parameters related to programming and computer science

LaTeX code and file paths

archaic german english

No Comments

No Known Activations