INDEX

Explanations

technical terms and definitions related to processes and mechanisms

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

This neuron identifies punctuation used to introduce definitions or explanations, particularly parenthetical content, colons, and semicolons that precede clarifying text.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

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

elerik

-0.28

UpInside

-0.28

 astfel

-0.27

 quite

-0.24

fă

-0.24

mişti

-0.24

 mínima

-0.23

jegy

-0.23

 Plage

-0.23

hilangan

-0.23

POSITIVE LOGITS

 surla

0.89

NameInMap

0.80

 kasarigan

0.79

:✨

0.75

цездатний

0.73

########.

0.69

adaptiveStyles

0.68

EndGlobalSection

0.68

 Италијани

0.68

oa̍t

0.66

Activations Density 0.374%

technical terms and definitions related to processes and mechanisms

This neuron identifies punctuation used to introduce definitions or explanations, particularly parenthetical content, colons, and semicolons that precede clarifying text.

No Comments

No Known Activations

technical terms and definitions related to processes and mechanisms

This neuron identifies punctuation used to introduce definitions or explanations, particularly parenthetical content, colons, and semicolons that precede clarifying text.

No Comments

No Known Activations