INDEX

Explanations

import

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_35.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.35.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ä¸Ģæīĭ

-0.26

 outnumber

-0.25

ÃŃtica

-0.24

æı¡

-0.23

 Benedict

-0.22

 larger

-0.22

 improv

-0.22

MLM

-0.22

å¼§

-0.22

 bigger

-0.22

POSITIVE LOGITS

in

0.43

ein

0.33

å¸¸çĶ¨

0.28

et

0.28

ite

0.28

ÐµÑħ

0.28

ean

0.27

thin

0.27

off

0.26

Activations Density 0.247%

import

No Comments

No Known Activations