INDEX

Explanations

mathematical and code symbols

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

The neuron is a strong indicator for code‐snippet tokens—especially “starred” LaTeX/math environments (e.g. align), DeclareMathOperator’s “{…}” syntax, Lisp’s “defun,” assembly’s “PTR,” etc.

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

基本的に

-1.74

 क्योंकि

-1.73

時々

-1.63

是因为

-1.59

鲼

-1.45

但她

-1.45

 будет

-1.44

 devido

-1.42

所以在

-1.41

 spesso

-1.40

POSITIVE LOGITS

 Produkten

2.19

荩

1.94

乆

1.94

亊

1.87

رنگ

1.83

مواد

1.80

衚

1.77

vantaged

1.73

泐

1.73

참고

1.73

Activations Density 0.002%

mathematical and code symbols

The neuron is a strong indicator for code‐snippet tokens—especially “starred” LaTeX/math environments (e.g. align), DeclareMathOperator’s “{…}” syntax, Lisp’s “defun,” assembly’s “PTR,” etc.

No Comments

No Known Activations

mathematical and code symbols

The neuron is a strong indicator for code‐snippet tokens—especially “starred” LaTeX/math environments (e.g. align), DeclareMathOperator’s “{…}” syntax, Lisp’s “defun,” assembly’s “PTR,” etc.

No Comments

No Known Activations

mathematical and code symbols

The neuron is a strong indicator for code‐snippet tokens—especially “starred” LaTeX/math environments (e.g. align*), DeclareMathOperator’s “*{…}” syntax, Lisp’s “defun,” assembly’s “PTR,” etc.

No Comments

No Known Activations

mathematical and code symbols

The neuron is a strong indicator for code‐snippet tokens—especially “starred” LaTeX/math environments (e.g. align*), DeclareMathOperator’s “*{…}” syntax, Lisp’s “defun,” assembly’s “PTR,” etc.

No Comments

No Known Activations

The neuron is a strong indicator for code‐snippet tokens—especially “starred” LaTeX/math environments (e.g. align), DeclareMathOperator’s “{…}” syntax, Lisp’s “defun,” assembly’s “PTR,” etc.

The neuron is a strong indicator for code‐snippet tokens—especially “starred” LaTeX/math environments (e.g. align), DeclareMathOperator’s “{…}” syntax, Lisp’s “defun,” assembly’s “PTR,” etc.