INDEX

Explanations

statements related to medical classification and drug properties

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

drugs and medications

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

setVerticalGroup

-0.52

 Wachs

-0.33

 suavemente

-0.30

 bañ

-0.29

 pertemuan

-0.28

velocityY

-0.28

 înc

-0.28

 exteriores

-0.27

 adelante

-0.27

CREATED

-0.27

POSITIVE LOGITS

 drugs

0.74

 medication

0.73

drugs

0.72

 médicament

0.71

 medications

0.69

 médicaments

0.68

 pharmaceutical

0.67

 drug

0.67

 pharmaceuticals

0.67

FDA

0.67

Activations Density 0.202%

statements related to medical classification and drug properties

drugs and medications

No Comments

No Known Activations