INDEX

Explanations

Code/debugging

np_max-act · gemini-2.0-flash

The neuron responds to programming‐language declaration keywords (e.g. data types and access modifiers) and other code‐structure tokens.

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

}%

-0.07

{})

-0.07

 precios

-0.06

 personals

-0.06

 Worlds

-0.06

 Noble

-0.06

 Bars

-0.06

κος

-0.06

 людини

-0.06

從

-0.06

POSITIVE LOGITS

Tensor

0.08

 дека

0.07

 minut

0.07

itches

0.07

 canoe

0.06

 fundraising

0.06

usz

0.06

Mrs

0.06

 Crowley

0.06

��

0.06

Activations Density 0.038%

Code/debugging

The neuron responds to programming‐language declaration keywords (e.g. data types and access modifiers) and other code‐structure tokens.

No Comments

No Known Activations