INDEX

Explanations

chemistry texts

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_30.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.30.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

åģļå¼º

-0.28

oras

-0.27

Acts

-0.25

ig

-0.25

ä¹ĭåĪĿ

-0.24

ÑĥÑĪ

-0.24

igan

-0.24

edes

-0.24

esor

-0.23

imap

-0.23

POSITIVE LOGITS

ng

0.36

cke

0.32

rror

0.32

lsa

0.30

lement

0.30

SSION

0.29

è¡Į

0.29

nga

0.28

nge

0.28

rl

0.28

Activations Density 0.021%

chemistry texts

No Known Activations