INDEX

Explanations

references to programming concepts and structures, particularly related to objects and JSON manipulation

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

Code, file paths, or markup

np_acts-logits-general · gemini-2.0-flash

code object

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

+#+

-0.49

chromedriver

-0.40

GraphicsUnit

-0.39

'{@

-0.39

pes

-0.36

smo

-0.36

/***/

-0.35

évaluateur

-0.35

QtGui

-0.33

EventId

-0.33

POSITIVE LOGITS

ariConfig

0.55

 Infórmanos

0.54

 nakalista

0.52

 Италијани

0.51

Diweddarwch

0.49

 Wikimedijinoj

0.48

MessageOf

0.47

parsedMessage

0.46

JsonInclude

0.46

rungsseite

0.45

Activations Density 0.358%

references to programming concepts and structures, particularly related to objects and JSON manipulation

Code, file paths, or markup

code object

No Comments

No Known Activations