INDEX

Explanations

number 2

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

jur

-0.27

 Vand

-0.26

 instruct

-0.26

ativity

-0.25

 panc

-0.25

reeze

-0.24

 consistency

-0.24

 conc

-0.23

Coach

-0.23

ä¸»åĬ¨æĢ§

-0.23

POSITIVE LOGITS

ovsky

0.28

 Giov

0.28

 reim

0.27

UGHT

0.26

(limit

0.25

quisition

0.24

cly

0.23

èĻ¹æ¡¥

0.23

UserProfile

0.23

é¢Ħçº¦

0.23

Activations Density 0.222%

number 2

No Comments

No Known Activations