INDEX

Explanations

forget

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_15/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 stimulus

-0.07

 Density

-0.07

 tặng

-0.06

 чолов

-0.06

=num

-0.06

 candidacy

-0.06

uild

-0.06

 relevance

-0.06

jsx

-0.06

POSITIVE LOGITS

Forget

0.09

 forgotten

0.09

 forget

0.09

 Forget

0.08

 Forgotten

0.08

forget

0.07

nette

0.07

 forgot

0.07

фров

0.07

gesi

0.07

Activations Density 0.010%

forget

No Comments

No Known Activations