INDEX

Explanations

programming code and syntax elements

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

.js, we reasoned, specific codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

 LoggerFactory

-0.56

epresidente

-0.46

CHEZ

-0.44

 généraux

-0.43

 Stuff

-0.43

 Vann

-0.43

ade

-0.42

GeneratedMessage

-0.42

 Sunday

-0.42

 raise

-0.42

POSITIVE LOGITS

 Roskov

0.90

SourceChecksum

0.84

 utafitiHapana

0.80

AddTagHelper

0.80

Personensuche

0.75

 ویکی‌پدی

0.72

Brainz

0.71

abestanden

0.70

出版年

0.68

 Mahomet

0.68

Activations Density 0.003%

programming code and syntax elements

.js, we reasoned, specific codes

No Comments

No Known Activations