INDEX

Explanations

code

np_max-act · gemini-2.0-flash

The neuron detects Tailwind CSS utility class names (i.e. tokens inside class attributes).

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

 daddy

-0.07

ensus

-0.07

peed

-0.07

сер

-0.06

 foll

-0.06

 consensus

-0.06

 потреб

-0.06

สม

-0.06

 forks

-0.06

Yo

-0.06

POSITIVE LOGITS

 Bergen

0.06

 přen

0.06

_predicted

0.06

.datab

0.06

Fitness

0.06

 Scalars

0.06

بود

0.06

HOH

0.06

][]

0.06

 questa

0.06

Activations Density 0.111%

code

The neuron detects Tailwind CSS utility class names (i.e. tokens inside class attributes).

No Comments

No Known Activations