INDEX

Explanations

terms related to emergency risk communication and public health guidance

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

health and disease

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

httphttps

-0.55

AnchorStyles

-0.54

 pinulongan

-0.45

sidemargin

-0.43

trale

-0.42

 BorderRadius

-0.41

 الحره

-0.41

ViewChild

-0.40

بوابة

-0.40

 IBOutlet

-0.39

POSITIVE LOGITS

 COVID

0.60

 coronavirus

0.57

 pandemic

0.56

 Coronavirus

0.55

 Covid

0.53

 virus

0.52

 covid

0.51

 epidemiological

0.51

 Ebola

0.50

 disease

0.49

Activations Density 0.053%

terms related to emergency risk communication and public health guidance

health and disease

No Comments

No Known Activations