INDEX

Explanations

use

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_7.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.7.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æªĶ

-0.28

 Boyd

-0.26

æ¬º

-0.25

 moot

-0.25

 Tribune

-0.25

åģľä¸ĭ

-0.24

vely

-0.24

 childbirth

-0.24

é©°

-0.24

.dep

-0.24

POSITIVE LOGITS

Sr

0.28

_REFRESH

0.28

å¹³è¡Į

0.27

çĻ½äºĳ

0.26

å²Ī

0.25

 parallel

0.25

parallel

0.25

Jr

0.25

refresh

0.24

 ì·¨

0.24

Activations Density 0.004%

use

No Comments

No Known Activations