INDEX

Explanations

references to clients, legal matters, and evidence-related terms

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

legal, financial, or experimental reports

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

 resourceCulture

-0.52

 betweenstory

-0.47

aarrggbb

-0.46

 nonUne

-0.45

 Chip

-0.41

tac

-0.39

complexType

-0.39

WireFormat

-0.39

twin

-0.38

adaptiveStyles

-0.38

POSITIVE LOGITS

 Biôgrafia

0.55

UnknownFieldSet

0.54

XmlAccessType

0.54

insee

0.47

seamnă

0.44

lewati

0.43

 дописавши

0.40

>{@

0.39

 scurt

0.38

 până

0.36

Activations Density 0.086%

references to clients, legal matters, and evidence-related terms

legal, financial, or experimental reports

No Comments

No Known Activations