INDEX

Explanations

references to medical imaging procedures and tests

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

medical diagnosis scans and biopsies

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

 שוליים

-0.49

 kaynağından

-0.46

Tikang

-0.44

 række

-0.43

 dagelijks

-0.42

ificent

-0.41

 Niveau

-0.40

 siguran

-0.40

 accett

-0.40

insics

-0.39

POSITIVE LOGITS

 exam

0.52

diagnos

0.51

MRI

0.51

 scan

0.51

0.50

 diagnosis

0.50

 inspection

0.49

 examination

0.49

oscopy

0.48

 autopsy

0.48

Activations Density 0.021%

references to medical imaging procedures and tests

medical diagnosis scans and biopsies

No Comments

No Known Activations