INDEX

Explanations

language

np_max-act · gemini-2.0-flash

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

َأ

-0.07

landır

-0.07

 Floyd

-0.06

(et

-0.06

_detalle

-0.06

aptors

-0.06

OffsetTable

-0.06

 Validates

-0.06

roduction

-0.06

 lacked

-0.06

POSITIVE LOGITS

-site

0.07

.getBytes

0.07

 onMouse

0.07

Partner

0.06

 acquire

0.06

<body

0.06

 kişisel

0.06

 Weak

0.06

 seated

0.06

 Initiative

0.06

Activations Density 0.016%

language

No Comments

No Known Activations