INDEX

Explanations

sequence structures

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

programming code syntax and commands, especially in the context of hardware control.

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

The neuron activates on numeric literals (tokens consisting of digits) in code snippets.

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

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_22/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

how

-1.33

 from

-1.29

new

-1.16

 mengganggu

-1.13

 That

-1.07

 their

-1.07

If

-1.06

 that

-1.05

 Also

-1.04

 mengucap

-1.04

POSITIVE LOGITS

ӟ

1.21

ומים

1.20

 invari

1.13

かったです

1.10

 lefs

1.09

ניים

1.09

心裡

1.08

莇

1.08

 adequ

1.07

嘮

1.06

Activations Density 0.001%

sequence structures

programming code syntax and commands, especially in the context of hardware control.

The neuron activates on numeric literals (tokens consisting of digits) in code snippets.

No Comments

No Known Activations

sequence structures

programming code syntax and commands, especially in the context of hardware control.

The neuron activates on numeric literals (tokens consisting of digits) in code snippets.

No Comments

No Known Activations