INDEX

Explanations

numbers and codes

np_max-act · gemini-2.0-flash

This neuron fires on product names and model identifiers—especially brand names and alphanumeric model codes.

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Properties

-0.07

 humanities

-0.07

 crud

-0.06

 coron

-0.06

 Jian

-0.06

 lief

-0.06

Club

-0.06

 donc

-0.06

@brief

-0.06

POSITIVE LOGITS

.IS

0.07

 نک

0.07

ISS

0.07

ัดส

0.07

 обс

0.06

 projects

0.06

任

0.06

 DataContext

0.06

円

0.06

三

0.06

Activations Density 0.036%

numbers and codes

This neuron fires on product names and model identifiers—especially brand names and alphanumeric model codes.

No Comments

No Known Activations