INDEX

Explanations

chemical formulas and scientific text

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

The neuron detects numerical subscripts in chemical formulas, particularly focusing on the number 2.

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

The neuron fires strongly on numeric and element‐symbol tokens that occur as part of chemical formulas (e.g. the “C6H10O5,” “Nb2,” “Cr2Ge2Te6,” etc.).

oai_token-act-pair · o4-mini Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_22/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

 laranja

-1.52

 retrô

-1.34

攥

-1.25

 знал

-1.23

 Melhor

-1.19

 bemerk

-1.17

᯾

-1.17

 wła

-1.17

видео

-1.16

фикации

-1.13

POSITIVE LOGITS

</em>

1.40

<bos>

1.31

CLINICAL

1.14

1.10

1.09

__*/

1.08

濑

1.06

とな

1.01

된

1.01

所谓

1.00

Activations Density 0.017%

chemical formulas and scientific text

The neuron detects numerical subscripts in chemical formulas, particularly focusing on the number 2.

The neuron fires strongly on numeric and element‐symbol tokens that occur as part of chemical formulas (e.g. the “C6H10O5,” “Nb2,” “Cr2Ge2Te6,” etc.).

No Comments

No Known Activations

chemical formulas and scientific text

The neuron detects numerical subscripts in chemical formulas, particularly focusing on the number 2.

The neuron fires strongly on numeric and element‐symbol tokens that occur as part of chemical formulas (e.g. the “C6H10O5,” “Nb2,” “Cr2Ge2Te6,” etc.).

No Comments

No Known Activations