INDEX

Explanations

percentages (v/v) or (w/v)

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

expressions indicating ratios or proportions, particularly "v/v" notation used in scientific contexts.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

The neuron fires on semantically content‐bearing tokens (i.e. non–stop words) rather than on common function words.

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

 doda

-1.39

楽しめます

-1.38

 otras

-1.38

粁

-1.37

前は

-1.37

-1.31

 erstes

-1.31

𝄆

-1.30

 chaque

-1.28

 zweiter

-1.27

POSITIVE LOGITS

who

1.88

 also

1.76

也不

1.60

のも

1.48

也

1.45

 wasn

1.45

所

1.44

也是

1.43

 hacienda

1.36

was

1.32

Activations Density 0.118%

percentages (v/v) or (w/v)

expressions indicating ratios or proportions, particularly "v/v" notation used in scientific contexts.

The neuron fires on semantically content‐bearing tokens (i.e. non–stop words) rather than on common function words.

No Comments

No Known Activations

percentages (v/v) or (w/v)

expressions indicating ratios or proportions, particularly "v/v" notation used in scientific contexts.

The neuron fires on semantically content‐bearing tokens (i.e. non–stop words) rather than on common function words.

No Comments

No Known Activations