INDEX

Explanations

patients

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_33.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.33.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

éĹ¸

-0.32

 inter

-0.27

 worm

-0.26

 ridge

-0.26

é³ŀ

-0.26

gs

-0.26

âĤĵ

-0.25

de

-0.25

ing

-0.25

 bird

-0.25

POSITIVE LOGITS

.coll

0.35

_coll

0.30

èģĶç³»æĪĳä»¬

0.28

 Coll

0.27

æ³¼

0.26

esan

0.26

(coll

0.26

 Locker

0.26

ibil

0.25

-Mart

0.25

Activations Density 0.014%

patients

No Known Activations