INDEX

Explanations

terms related to health measurements and correlations, specifically in a clinical context

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

blood pressure and HF

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

 '\\;'

-0.57

 lèvres

-0.51

̈́

-0.49

tagHelperRunner

-0.49

 cherchés

-0.48

 illustrationer

-0.48

 løpet

-0.47

Rifer

-0.46

 lærer

-0.45

 betweenstory

-0.45

POSITIVE LOGITS

kháu

0.48

EndGlobalSection

0.42

andExpect

0.39

iloc

0.38

 nahilalakip

0.37

 significantly

0.37

findpost

0.37

Kapp

0.36

wnież

0.36

ともに

0.35

Activations Density 0.207%

terms related to health measurements and correlations, specifically in a clinical context

blood pressure and HF

No Comments

No Known Activations