INDEX

Explanations

foreignsuffixes and chemical terms

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

numerical terms and measurement units, particularly in foreign languages or technical contexts.

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

Triggered by @neilrathi

The neuron fires on spelled-out numbers and numeric prefixes/affixes (e.g. “twenty,” “four,” “uno,” “duodeca,” “nonade,” “cinquante,” 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

Negative Logits

尨

-2.11

全く

-2.11

優しく

-2.03

 WHEN

-2.02

というわけで

-2.00

大好きです

-1.95

はっきり

-1.93

まったく

-1.90

 nové

-1.87

౧

-1.87

POSITIVE LOGITS

."

2.25

。”

2.20

"));

2.09

}*/

1.98

！”

1.97

},

1.97

1.85

[]

1.84

 उसकी

1.77

 flamme

1.76

Activations Density 0.020%

foreignsuffixes and chemical terms

numerical terms and measurement units, particularly in foreign languages or technical contexts.

The neuron fires on spelled-out numbers and numeric prefixes/affixes (e.g. “twenty,” “four,” “uno,” “duodeca,” “nonade,” “cinquante,” etc.).

No Known Activations

foreignsuffixes and chemical terms

numerical terms and measurement units, particularly in foreign languages or technical contexts.

The neuron fires on spelled-out numbers and numeric prefixes/affixes (e.g. “twenty,” “four,” “uno,” “duodeca,” “nonade,” “cinquante,” etc.).

No Known Activations