INDEX

Explanations

devices and platforms

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

punctuation marks that separate items in lists or technical specifications, particularly commas and periods in contexts involving software, hardware, or web browser compatibility information.

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

The neuron flags list‐style or enumeration tokens—things like colons, commas, version or model numbers, and other markers of technical or browser‐support lists.

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

in

-1.38

selbe

-1.32

 there

-1.23

extré

-1.16

 each

-1.14

-1.12

قای

-1.09

LEMMA

-1.09

 dejamos

-1.09

 also

-1.09

POSITIVE LOGITS

sierten

1.34

⁛

1.29

 Præ

1.26

 réalisé

1.25

豋

1.24

 ہے۔

1.23

鞮

1.23

undos

1.22

廰

1.21

and

1.20

Activations Density 0.221%

devices and platforms

punctuation marks that separate items in lists or technical specifications, particularly commas and periods in contexts involving software, hardware, or web browser compatibility information.

The neuron flags list‐style or enumeration tokens—things like colons, commas, version or model numbers, and other markers of technical or browser‐support lists.

No Comments

No Known Activations

devices and platforms

punctuation marks that separate items in lists or technical specifications, particularly commas and periods in contexts involving software, hardware, or web browser compatibility information.

The neuron flags list‐style or enumeration tokens—things like colons, commas, version or model numbers, and other markers of technical or browser‐support lists.

No Comments

No Known Activations