INDEX

Explanations

code/markup

np_max-act · gemini-2.0-flash

The neuron is triggered by numeric literals in code—i.e. tokens that are digits or number constants.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 protagonist

-0.07

 Markdown

-0.07

 conquest

-0.06

.units

-0.06

 pollen

-0.06

-chan

-0.06

Muon

-0.06

 можуть

-0.06

iotics

-0.06

Vehicle

-0.06

POSITIVE LOGITS

 creds

0.07

�

0.06

 ''),

0.06

 enlightenment

0.06

.par

0.06

 renewable

0.06

Рµ

0.06

 "(\<

0.06

rain

0.06

ATHER

0.06

Activations Density 0.012%

code/markup

The neuron is triggered by numeric literals in code—i.e. tokens that are digits or number constants.

No Comments

No Known Activations

code/markup

The neuron is triggered by numeric literals in code—i.e. tokens that are digits or number constants.

No Comments

No Known Activations