INDEX

Explanations

code and image filenames

np_acts-logits-general · gemini-2.5-flash-lite

closing brackets, parentheses, or braces in source code.

oai_token-act-pair · claude-4-5-sonnet Triggered by @bfgcn7frf5

The neuron fires on numeric literals in the text (especially numbers, including integers and decimals).

oai_token-act-pair · o4-mini Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_10/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ってください

-1.46

 جالب

-1.41

 preparada

-1.34

كتشاف

-1.29

 gaining

-1.24

 multifaceted

-1.23

 življen

-1.22

ってみて

-1.21

 oseb

-1.20

 examining

-1.20

POSITIVE LOGITS

 แต่

1.34

 даже

1.24

 других

1.18

应该

1.18

оны

1.17

他们的

1.16

 тема

1.16

was

1.16

 debería

1.16

 сцена

1.16

Activations Density 0.002%

code and image filenames

closing brackets, parentheses, or braces in source code.

The neuron fires on numeric literals in the text (especially numbers, including integers and decimals).

No Comments

No Known Activations

code and image filenames

closing brackets, parentheses, or braces in source code.

The neuron fires on numeric literals in the text (especially numbers, including integers and decimals).

No Comments

No Known Activations