INDEX

Explanations

individual letters

np_max-act · gemini-2.0-flash

The neuron activates on tokens that are parts of Unicode character names in mapping files (e.g. “CYRILLIC … SMALL LETTER …”).

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

Aus

-0.06

_USER

-0.06

 standalone

-0.06

Cluster

-0.06

 Applicants

-0.06

…but

-0.06

fmt

-0.06

Comparator

-0.06

RenderingContext

-0.06

POSITIVE LOGITS

 offsetY

0.07

工具

0.07

_UPDATE

0.07

	position

0.06

 seasons

0.06

ща

0.06

Restr

0.06

рова

0.06

也是

0.06

 retal

0.06

Activations Density 0.001%

individual letters

The neuron activates on tokens that are parts of Unicode character names in mapping files (e.g. “CYRILLIC … SMALL LETTER …”).

No Comments

No Known Activations

individual letters

The neuron activates on tokens that are parts of Unicode character names in mapping files (e.g. “CYRILLIC … SMALL LETTER …”).

No Comments

No Known Activations