INDEX

Explanations

references to entities or participants in a document

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

legal entities and roles

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

 propOrder

-0.63

цездатний

-0.57

 ویکی‌پدی

-0.56

OGND

-0.55

Autoritní

-0.52

oa̍t

-0.51

 ब्रेकडाउन

-0.51

Попис

-0.51

 kasarigan

-0.49

Diweddarwch

-0.49

POSITIVE LOGITS

bo

0.40

참고

0.38

stead

0.38

AccessorTable

0.38

isat

0.37

istern

0.36

disposing

0.36

correspond

0.36

vista

0.36

 also

0.35

Activations Density 0.022%

references to entities or participants in a document

legal entities and roles

No Comments

No Known Activations

references to entities or participants in a document

legal entities and roles

No Comments

No Known Activations