INDEX

Explanations

code punctuation

np_max-act · gemini-2.0-flash

This neuron fires on large code or technical blocks (e.g. multi-line code examples).

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

 grâce

-0.06

Vet

-0.06

输

-0.06

废

-0.06

θ

-0.06

oteric

-0.06

语

-0.06

ITAL

-0.06

 authenticated

-0.06

 duct

-0.06

POSITIVE LOGITS

(--

0.07

.ConnectionStrings

0.07

 gracefully

0.07

 температу

0.06

={

0.06

 Confidence

0.06

 Základní

0.06

烦

0.06

 alınan

0.06

 Adam

0.06

Activations Density 0.026%

code punctuation

This neuron fires on large code or technical blocks (e.g. multi-line code examples).

No Comments

No Known Activations