INDEX

Explanations

code syntax elements like brackets and semicolons

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

array or list indexing operations with bracket notation in programming code.

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

The neuron is triggered by numeric literals used as indices (e.g. array or list bracket-index expressions).

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

ircut

-1.36

わかります

-1.27

見えて

-1.23

的部分

-1.23

 отме

-1.21

関わ

-1.20

ליה

-1.13

ぶん

-1.13

 llavero

-1.12

Isn

-1.11

POSITIVE LOGITS

 then

1.49

}(),

1.36

主演

1.29

ي

1.27

 ")";

1.27

قر

1.24

「

1.23

sikal

1.22

WHEREAS

1.21

بعد

1.20

Activations Density 0.027%

code syntax elements like brackets and semicolons

array or list indexing operations with bracket notation in programming code.

The neuron is triggered by numeric literals used as indices (e.g. array or list bracket-index expressions).

No Comments

No Known Activations

code syntax elements like brackets and semicolons

array or list indexing operations with bracket notation in programming code.

The neuron is triggered by numeric literals used as indices (e.g. array or list bracket-index expressions).

No Comments

No Known Activations