INDEX

Explanations

references to figures and tables in academic or research documents

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

figures or citations

np_acts-logits-general · gemini-2.0-flash

citations and bracketed references

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

tagHelperRunner

-0.54

 चीज़ों

-0.45

IsContent

-0.45

Hozzáférés

-0.45

abestanden

-0.43

Viited

-0.39

EncBinary

-0.39

AndEndTag

-0.39

writeField

-0.37

 calefacción

-0.37

POSITIVE LOGITS

rungsseite

0.65

 PeEnEo

0.54

 EClass

0.54

 censiti

0.48

webElementXpaths

0.48

 Италијани

0.46

فایل‌لار

0.44

 referrerpolicy

0.44

principalTable

0.44

Бахар

0.44

Activations Density 0.613%

references to figures and tables in academic or research documents

figures or citations

citations and bracketed references

No Comments

No Known Activations