INDEX

Explanations

0

np_max-act · gemini-2.0-flash

The neuron is primarily detecting special structural tokens that mark system/role delimiters (e.g. “<|start_header_id|>”, “<|begin_of_text|>”, etc.).

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

htaking

-0.07

허

-0.07

noloj

-0.07

_NS

-0.06

>NN

-0.06

 cropping

-0.06

ısıyla

-0.06

 использ

-0.06

 PROPERTY

-0.06

 weeks

-0.06

POSITIVE LOGITS

担当

0.07

 China

0.07

 statewide

0.06

سبة

0.06

동

0.06

Som

0.06

 inclination

0.06

China

0.06

ทำ

0.06

 posture

0.06

Activations Density 0.004%

0

The neuron is primarily detecting special structural tokens that mark system/role delimiters (e.g. “<|start_header_id|>”, “<|begin_of_text|>”, etc.).

No Comments

No Known Activations