INDEX

Explanations

references to healthcare and related professionals

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Copy

-0.46

makeConstraints

-0.45

 COPY

-0.44

 copy

-0.44

shadowOpacity

-0.42

 ся

-0.42

σ

-0.42

곱

-0.41

 STRIP

-0.41

ToScroll

-0.41

POSITIVE LOGITS

 healthcare

1.12

 Healthcare

1.11

Healthcare

1.09

healthcare

1.00

Health

0.69

 Health

0.68

 santé

0.67

HEALTH

0.65

Kjelder

0.65

health

0.64

Activations Density 0.010%

references to healthcare and related professionals

No Comments

No Known Activations