INDEX

Explanations

last

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

æ»ļ

-0.31

AtPath

-0.29

ä¸ĬæĿ¥

-0.28

roller

-0.28

.pg

-0.27

 roller

-0.26

plier

-0.26

erval

-0.26

_che

-0.25

 rolled

-0.24

POSITIVE LOGITS

 last

0.41

-last

0.38

last

0.37

LAST

0.35

 Last

0.34

(last

0.33

Last

0.32

onor

0.30

,last

0.30

 lasts

0.30

Activations Density 1.715%

last

No Known Activations