INDEX

Explanations

fictional characters

np_max-act · gemini-2.0-flash

The neuron activates on placeholder tokens representing named entities and their numeric identifiers (e.g., “NAME,” digits, and related fragments).

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

 жод

-0.07

.om

-0.07

 Пом

-0.07

 quoi

-0.07

 Rent

-0.07

وان

-0.06

 phần

-0.06

Minimal

-0.06

 určit

-0.06

 unprecedented

-0.06

POSITIVE LOGITS

 Miles

0.06

oooooooo

0.06

..........

0.06

cause

0.06

're

0.06

ανα

0.06

атель

0.06

 chịu

0.06

 sidewalks

0.06

Vk

0.05

Activations Density 0.052%

fictional characters

The neuron activates on placeholder tokens representing named entities and their numeric identifiers (e.g., “NAME,” digits, and related fragments).

No Comments

No Known Activations