INDEX

Explanations

love or some form of fall

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

public

-0.32

iere

-0.27

 public

-0.26

æµĳ

-0.26

 introdu

-0.26

.consume

-0.25

åĲĳä¸ľ

-0.25

 sich

-0.24

ogue

-0.24

é¾Ļ

-0.24

POSITIVE LOGITS

ä¸įè¯¥

0.29

.FromResult

0.25

ä¸įåĥı

0.25

åı¯ä¸įæĺ¯

0.25

flush

0.24

éĹ®é¢ĺ

0.24

hide

0.24

ä¸įåºĶè¯¥

0.23

 Transmit

0.23

 backed

0.23

Activations Density 0.002%

love or some form of fall

No Comments

No Known Activations