INDEX

Explanations

references to illustrative figures and tables in the text

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

figures in research papers

np_acts-logits-general · gemini-2.0-flash

referencing figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 disambiguazione

-0.63

 Meksiku

-0.57

 AssemblyCompany

-0.53

</em>

-0.52

+:+

-0.51

IntoConstraints

-0.48

</code>

-0.48

 pavimento

-0.47

 consultato

-0.47

GEBURTSDATUM

-0.47

POSITIVE LOGITS

اریخ

0.71

,:),

0.70

ритори

0.63

espèce

0.62

stående

0.61

moiselle

0.61

Manfaat

0.59

lapsingToolbar

0.59

paramref

0.59

 LABEL

0.59

Activations Density 0.844%

references to illustrative figures and tables in the text

figures in research papers

referencing figures

No Comments

No Known Activations