INDEX

Explanations

user reviews

np_max-act · gemini-2.0-flash

The neuron is primarily activated by numeric tokens (especially decimal numbers and ratings).

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

 construction

-0.07

setParameter

-0.06

 insistence

-0.06

انيا

-0.06

 erection

-0.06

 polic

-0.06

completion

-0.06

	left

-0.06

 territory

-0.06

 Reminder

-0.06

POSITIVE LOGITS

_pago

0.07

检

0.06

 retorna

0.06

anno

0.06

 Intelli

0.06

 CreateTable

0.06

็ม

0.06

 kutje

0.06

 geral

0.06

 BOOL

0.06

Activations Density 0.015%

user reviews

The neuron is primarily activated by numeric tokens (especially decimal numbers and ratings).

No Comments

No Known Activations