INDEX

Explanations

discussions related to healthcare practices and their effectiveness

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

clinical status and improvements

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

 Relaciones

-0.36

Rüyada

-0.36

árbol

-0.34

 swears

-0.33

 étranger

-0.33

principalColumn

-0.33

Ārējās

-0.33

 penge

-0.30

 constructeur

-0.30

ConstraintMaker

-0.29

POSITIVE LOGITS

 patients

0.73

Patients

0.70

 Patients

0.69

patients

0.65

 patient

0.63

LookAnd

0.58

 clinicians

0.57

 clinical

0.57

sizeCache

0.56

patient

0.55

Activations Density 0.154%

discussions related to healthcare practices and their effectiveness

clinical status and improvements

No Comments

No Known Activations