INDEX

Explanations

reflections

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_19/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.19.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

浦

-0.07

 обра

-0.07

 Empresa

-0.07

Mgr

-0.06

_preds

-0.06

 Пра

-0.06

ipl

-0.06

訓

-0.06

'.↵

-0.06

 phục

-0.06

POSITIVE LOGITS

 dahi

0.07

 didReceiveMemoryWarning

0.07

 preserve

0.07

 refining

0.06

 brutal

0.06

.forEach

0.06

 součas

0.06

 awaits

0.06

 adjust

0.06

_OFFSET

0.06

Activations Density 0.020%

reflections

No Comments

No Known Activations