INDEX

Explanations

references to health and well-being in various contexts

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

health and healthy

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

########.

-0.37

 utafitiHapana

-0.36

 mosquito

-0.36

 request

-0.35

 option

-0.33

 modi

-0.33

 later

-0.33

茅

-0.33

 annoying

-0.32

 Modi

-0.31

POSITIVE LOGITS

healthy

0.88

Healthy

0.88

 healthy

0.79

 Healthy

0.78

 sehat

0.71

HEALTH

0.70

Integrity

0.68

 HEALTH

0.67

 zdrow

0.67

 saludable

0.66

Activations Density 0.012%

references to health and well-being in various contexts

health and healthy

No Comments

No Known Activations