INDEX

Explanations

terms related to medical accreditation and regulatory standards.

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

names of organizations and related regulatory information.

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

phrases related to official certifications, accreditations, and registrations for organizations or programs.

oai_token-act-pair · claude-3-5-sonnet-20240620 Triggered by @bot

Accreditation, certification, registration, or licensing

np_acts-logits-general · gemini-2.0-flash

certified and accredited

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Theſe

-0.81

 Majefty

-0.74

 Monfieur

-0.72

 purpoſe

-0.71

 Jefus

-0.69

 reaſon

-0.67

 greateſt

-0.66

 poffible

-0.66

 المعيارى

-0.64

 Anſ

-0.63

POSITIVE LOGITS

 certification

2.06

 certified

2.04

 certifications

1.72

 Certification

1.72

certified

1.70

 Certified

1.70

 certifying

1.67

 certific

1.67

 accreditation

1.66

 accredited

1.65

Activations Density 0.205%