INDEX

Explanations

references to medical conditions or health-related topics

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

Numbers followed by parentheses

np_acts-logits-general · gemini-2.0-flash

numbers within parentheses

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pleaſure

-0.57

 purpoſe

-0.52

 ſche

-0.49

intenant

-0.48

 foundations

-0.45

 myſelf

-0.44

 preſent

-0.43

 lanjut

-0.43

 prochains

-0.43

 domains

-0.42

POSITIVE LOGITS

 Administrativna

0.50

 становника

0.48

Carriera

0.43

PostInfinity

0.43

intios

0.41

 فريبيس

0.41

GEBURTSDATUM

0.39

 disambiguazione

0.39

rawDesc

0.39

 gyhoeddwyd

0.38

Activations Density 1.407%

references to medical conditions or health-related topics

Numbers followed by parentheses

numbers within parentheses

No Comments

No Known Activations