INDEX

Explanations

.

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_3/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.3.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

{}

-0.07

 proposes

-0.07

 demonstrates

-0.07

aleigh

-0.07

 vertical

-0.07

ymbol

-0.07

 stacking

-0.07

准备

-0.07

евых

-0.07

 division

-0.07

POSITIVE LOGITS

 Balanced

0.07

 HinderedRotor

0.07

_executor

0.07

 Shapes

0.06

 İng

0.06

Motor

0.06

 діяль

0.06

Brain

0.06

 moveTo

0.06

.lastIndexOf

0.06

Activations Density 0.371%

.

No Comments

No Known Activations