INDEX

Explanations

structure-activity analyses

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

periods or full stops at the end of paragraphs.

oai_token-act-pair · claude-3-7-sonnet-20250219

Triggered by @neilrathi

The neuron strongly activates on end-of-sentence periods, i.e. sentence-terminating punctuation.

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

Negative Logits

 alho

-2.08

 plafon

-1.98

 faciliter

-1.83

 trouver

-1.81

 différents

-1.81

………………

-1.73

……….

-1.71

 разных

-1.70

を見つけ

-1.70

 ſont

-1.69

POSITIVE LOGITS

has

2.69

and

2.39

 allow

2.33

 will

2.05

 create

2.02

 truly

1.97

 give

1.97

 allows

1.92

 make

1.91

 actually

1.88

Activations Density 0.104%

structure-activity analyses

periods or full stops at the end of paragraphs.

The neuron strongly activates on end-of-sentence periods, i.e. sentence-terminating punctuation.

No Known Activations

structure-activity analyses

periods or full stops at the end of paragraphs.

The neuron strongly activates on end-of-sentence periods, i.e. sentence-terminating punctuation.

No Known Activations