INDEX

Explanations

references to possessive pronouns and articles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 unknownFields

-0.42

échelle

-0.41

essentiel

-0.38

 disambiguazione

-0.38

 partout

-0.37

atmosphère

-0.37

setVerticalGroup

-0.36

 everybody

-0.36

Something

-0.35

 chamber

-0.34

POSITIVE LOGITS

 ability

0.71

 involvement

0.65

 participation

0.59

 appearance

0.56

 efforts

0.54

AFX

0.52

 keen

0.52

 importance

0.52

 ongoing

0.51

 عدم

0.51

Activations Density 0.370%

references to possessive pronouns and articles

No Comments

No Known Activations