INDEX

Explanations

{% templating %}{% block %} {% for %} {% set %}

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

Django/Jinja2 template syntax tags and delimiters.

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

This neuron detects template directive tags (the “{% … %}” blocks) in templating languages like Django/Jinja.

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

 PROFES

-2.02

︕

-2.00

躄

-2.00

 และ

-1.98

罍

-1.80

粜

-1.77

齶

-1.74

鶿

-1.73

糶

-1.70

 italie

-1.69

POSITIVE LOGITS

 only

1.72

()){

1.69

 about

1.65

not

1.62

 tomado

1.52

 undoubtedly

1.48

嫱

1.48

Why

1.44

 tecnico

1.41

Poznám

1.40

Activations Density 0.019%

{% templating %}{% block %} {% for %} {% set %}

Django/Jinja2 template syntax tags and delimiters.

This neuron detects template directive tags (the “{% … %}” blocks) in templating languages like Django/Jinja.

No Comments

No Known Activations

{% templating %}{% block %} {% for %} {% set %}

Django/Jinja2 template syntax tags and delimiters.

This neuron detects template directive tags (the “{% … %}” blocks) in templating languages like Django/Jinja.

No Comments

No Known Activations