INDEX

Explanations

terms related to medical research and procedures

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

followed by past-tense verbs

np_acts-logits-general · gemini-2.0-flash

past passive constructions

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 oprot

-0.68

 дописавши

-0.59

 TextAppearance

-0.57

 BoxDecoration

-0.57

AndEndTag

-0.56

queryInterface

-0.54

unhofer

-0.53

reactstrap

-0.52

MemoryWarning

-0.51

səhifə

-0.51

POSITIVE LOGITS

 centered

0.33

 radio

0.32

 consta

0.31

 leopardo

0.30

ชาย

0.30

vinfos

0.30

GEBURTSDATUM

0.30

Osob

0.30

netics

0.30

 involving

0.29

Activations Density 1.181%

terms related to medical research and procedures

followed by past-tense verbs

past passive constructions

No Comments

No Known Activations