INDEX

Explanations

references to legal processes and documentation

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

legal actions and complaints

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

 estekak

-0.86

 للاسماء

-0.73

CloseOperation

-0.72

NameInMap

-0.65

 wireType

-0.64

 MainAxisSize

-0.61

 Taktlose

-0.61

aarrggbb

-0.60

MemoryWarning

-0.60

 oprot

-0.59

POSITIVE LOGITS

 warned

0.38

 koke

0.33

 dokonce

0.32

 protested

0.32

sł

0.31

 complained

0.31

 Rücksicht

0.30

 warns

0.30

 kości

0.29

 الرياضيه

0.29

Activations Density 0.058%

references to legal processes and documentation

legal actions and complaints

No Comments

No Known Activations