INDEX

Explanations

phrases indicating conditionality and recommendations in arguments

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

the word "and" appearing in contexts expressing statements of obligation, possibility, or logical connection.

oai_token-act-pair · claude-4-5-haiku Triggered by @emiglarou

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

yntaxException

-0.49

UnusedPrivate

-0.48

 ujednoznacz

-0.44

лтемелер

-0.43

ImageContext

-0.41

 snippetHide

-0.40

 invokingState

-0.40

GeneratedCode

-0.40

sidemargin

-0.39

 AssemblyCulture

-0.38

POSITIVE LOGITS

 will

0.56

 EconPapers

0.48

 ought

0.47

 đang

0.47

 continues

0.44

 isActive

0.44

 anzi

0.44

has

0.43

为止

0.43

StoryboardSegue

0.43

Activations Density 0.298%

phrases indicating conditionality and recommendations in arguments

the word "and" appearing in contexts expressing statements of obligation, possibility, or logical connection.

No Comments

No Known Activations