INDEX

Explanations

Instructions

np_max-act · gemini-2.0-flash

The neuron activates on numeric tokens (digits or numbers), such as step indices and version numbers, highlighting numbered items in instruction text.

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

 tourist

-0.07

 feathers

-0.07

 pack

-0.07

اب

-0.07

_experiment

-0.07

 τα

-0.07

 rapide

-0.07

TZ

-0.07

_Insert

-0.06

 distraction

-0.06

POSITIVE LOGITS

 ENTER

0.06

$val

0.06

scr

0.06

、高

0.06

 เห

0.06

ůležit

0.06

 dém

0.06

 zvý

0.06

ــــــــ

0.06

 []);↵↵

0.05

Activations Density 0.076%

Instructions

The neuron activates on numeric tokens (digits or numbers), such as step indices and version numbers, highlighting numbered items in instruction text.

No Comments

No Known Activations