INDEX

Explanations

numbers 1 and 2

np_max-act · gemini-2.0-flash

This neuron activates on isolated numeric tokens marking enumerated items or headings (e.g. list numbers or section labels).

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

 adjoining

-0.07

 coating

-0.07

.iv

-0.07

 مقدار

-0.06

 jehož

-0.06

talk

-0.06

 ushort

-0.06

 مذ

-0.06

쿠

-0.06

γή

-0.06

POSITIVE LOGITS

 numéro

0.07

papers

0.07

<AudioSource

0.06

_Function

0.06

(Employee

0.06

 Surv

0.06

.buy

0.06

.getItemId

0.06

uniacid

0.06

.currentTarget

0.06

Activations Density 0.030%

numbers 1 and 2

This neuron activates on isolated numeric tokens marking enumerated items or headings (e.g. list numbers or section labels).

No Comments

No Known Activations