INDEX

Explanations

model modelalphanumeric identifiers and model numbers

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

references to Intel processor models and architectures (like i3, i5, i7, i386, i686, x86).

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

The neuron spikes on occurrences of Intel CPU/architecture identifiers of the form “i” plus a digit (e.g. “i3”, “i7”, “i386”).

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

<em>

-3.16

↵↵

-3.09

In

-3.06

…”

-2.78

↵

-2.78

-2.77

(),

-2.66

…..

-2.66

….

-2.63

<i>

-2.61

POSITIVE LOGITS

嚒

2.73

潆

2.72

 kimdir

2.64

 continúas

2.55

ቷ

2.50

 continú

2.45

 discul

2.41

)，

2.39

 brilla

2.38

 desbloquear

2.36

Activations Density 0.033%

model modelalphanumeric identifiers and model numbers

references to Intel processor models and architectures (like i3, i5, i7, i386, i686, x86).

The neuron spikes on occurrences of Intel CPU/architecture identifiers of the form “i” plus a digit (e.g. “i3”, “i7”, “i386”).

No Comments

No Known Activations

model modelalphanumeric identifiers and model numbers

references to Intel processor models and architectures (like i3, i5, i7, i386, i686, x86).

The neuron spikes on occurrences of Intel CPU/architecture identifiers of the form “i” plus a digit (e.g. “i3”, “i7”, “i386”).

No Comments

No Known Activations