INDEX

Explanations

terms related to medical or biological concepts, particularly those concerning genetic or cellular behavior

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

word compounds and following words

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

Vidite

-0.66

 defaultstate

-0.64

 disambiguazione

-0.61

 OMITBAD

-0.59

 الاطلاع

-0.58

 <<<<<<<<<<<<<<

-0.56

#+#

-0.56

rungsseite

-0.55

PreferredItem

-0.54

Биография

-0.54

POSITIVE LOGITS

SupportActionBar

0.37

 AssemblyTitle

0.35

DrawerToggle

0.31

serviceWorker

0.31

 face

0.29

 system

0.29

CodedInputStream

0.28

 class

0.28

to

0.28

messageInfo

0.27

Activations Density 0.140%

terms related to medical or biological concepts, particularly those concerning genetic or cellular behavior

word compounds and following words

No Comments

No Known Activations