INDEX

Explanations

PIP mobility, Cordova apps

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

hexadecimal color codes or memory addresses (starting with "0x" followed by "ff").

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

The neuron activates on uppercase technical tokens—especially acronyms or hexadecimal constants.

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

-4.06

-3.91

-3.83

if

-3.58

-3.55

-3.39

-3.36

They

-3.13

we

-3.13

ll

-3.08

POSITIVE LOGITS

倂

2.63

燹

2.53

 agricultura

2.50

倶楽部

2.44

 vencedor

2.44

 Monfieur

2.42

饜

2.36

岛

2.30

เพราะ

2.27

日本的

2.25

Activations Density 0.002%

PIP mobility, Cordova apps

hexadecimal color codes or memory addresses (starting with "0x" followed by "ff").

The neuron activates on uppercase technical tokens—especially acronyms or hexadecimal constants.

No Comments

No Known Activations

PIP mobility, Cordova apps

hexadecimal color codes or memory addresses (starting with "0x" followed by "ff").

The neuron activates on uppercase technical tokens—especially acronyms or hexadecimal constants.

No Comments

No Known Activations