INDEX

Explanations

initialisms and abbreviations

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

tab characters or whitespace used for formatting and indentation in structured documents.

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

The neuron is specialized to pick up page‐break or other non‐printing control characters (e.g. form‐feed marks).

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

 firmy

-1.94

 okno

-1.89

terrasse

-1.82

<bos>

-1.77

kurtka

-1.69

 我们

-1.69

our

-1.69

Nguy

-1.67

不

-1.63

 trhu

-1.60

POSITIVE LOGITS

1.85

 coreana

1.66

 такі

1.63

 coreano

1.60

1.59

 оригі

1.53

 decisivo

1.50

 使っ

1.50

 CONCERN

1.49

 casero

1.48

Activations Density 0.000%

initialisms and abbreviations

tab characters or whitespace used for formatting and indentation in structured documents.

The neuron is specialized to pick up page‐break or other non‐printing control characters (e.g. form‐feed marks).

No Comments

No Known Activations

initialisms and abbreviations

tab characters or whitespace used for formatting and indentation in structured documents.

The neuron is specialized to pick up page‐break or other non‐printing control characters (e.g. form‐feed marks).

No Comments

No Known Activations