INDEX

Explanations

references to medical terminology and studies

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

Academic/research studies and designs

np_acts-logits-general · gemini-2.0-flash

describes states or roles

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

 synes

-0.33

 forhold

-0.32

 default

-0.31

 oxygen

-0.30

 pienso

-0.30

 alvor

-0.29

 évidence

-0.28

 egyes

-0.28

 notizia

-0.27

 obcí

-0.27

POSITIVE LOGITS

⟬

0.69

IntoConstraints

0.69

 Paglinawan

0.69

ConstraintMaker

0.65

 للاسماء

0.62

RTEE

0.62

utilisons

0.61

+#+

0.61

########.

0.60

évaluateur

0.60

Activations Density 1.488%

references to medical terminology and studies

Academic/research studies and designs

describes states or roles

No Comments

No Known Activations