INDEX

Explanations

quoted strings and shell commands

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

variable references in shell scripts and configuration files, particularly those prefixed with a dollar sign.

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

This neuron detects shell-style variable or command substitutions (tokens beginning with “$”, “$(”, or “${”).

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

 ederim

-1.65

ܜ

-1.55

蘗

-1.55

</sup>

-1.52

hoea

-1.51

 mendes

-1.49

</h2>

-1.48

 seems

-1.48

 emphasized

-1.46

 potente

-1.46

POSITIVE LOGITS

1.87

1.75

 They

1.67

逦

1.66

er

1.64

1.52

 faisant

1.46

ie

1.46

 pię

1.45

yah

1.45

Activations Density 0.069%

quoted strings and shell commands

variable references in shell scripts and configuration files, particularly those prefixed with a dollar sign.

This neuron detects shell-style variable or command substitutions (tokens beginning with “$”, “$(”, or “${”).

No Comments

No Known Activations

quoted strings and shell commands

variable references in shell scripts and configuration files, particularly those prefixed with a dollar sign.

This neuron detects shell-style variable or command substitutions (tokens beginning with “$”, “$(”, or “${”).

No Comments

No Known Activations