INDEX

Explanations

descriptions of chemical structures and bonding in materials, particularly graphene and graphite

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

The neuron identifies molecular structure descriptors with phrases containing "-like" (especially "graphene-like" and "graphitic-like") in scientific papers about carbon materials.

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

 تضيفلها

-0.60

parsedMessage

-0.56

Personendaten

-0.55

 snippetHide

-0.54

 ویکی‌پدیا

-0.53

 hidrá

-0.48

pegno

-0.48

 Wiktionnaire

-0.47

 calendriers

-0.47

 Administrativna

-0.47

POSITIVE LOGITS

like

0.42

similar

0.42

 similar

0.41

esque

0.39

 Similar

0.38

tfrac

0.38

 ähnliche

0.35

richTextPanel

0.34

 аналоги

0.34

 similarly

0.34

Activations Density 0.384%

descriptions of chemical structures and bonding in materials, particularly graphene and graphite

The neuron identifies molecular structure descriptors with phrases containing "-like" (especially "graphene-like" and "graphitic-like") in scientific papers about carbon materials.

No Comments

No Known Activations