INDEX

Explanations

references to healthcare professionals and their qualifications

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

professionals and medical staff

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

 cris

-0.39

 Torino

-0.38

Rüyada

-0.38

 соц

-0.37

受

-0.37

WireFormatLite

-0.37

 POTENTIAL

-0.37

 disrupted

-0.37

 shook

-0.36

izarro

-0.36

POSITIVE LOGITS

 AssemblyCompany

0.63

physician

0.59

 physician

0.58

 GenerationType

0.56

 surgeon

0.55

 للمعارف

0.55

 Physician

0.54

 doctor

0.54

 practitioner

0.52

 professionisti

0.51

Activations Density 0.019%

references to healthcare professionals and their qualifications

professionals and medical staff

No Comments

No Known Activations