INDEX

Explanations

-ing and past-tense verbs

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_18.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.18.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 asympt

-0.28

-human

-0.27

ips

-0.25

 deter

-0.24

 bounce

-0.24

ä¼İ

-0.23

amation

-0.23

 Doll

-0.23

ovich

-0.23

åı£ä¸Ń

-0.23

POSITIVE LOGITS

ä¼ģä¸ļæıĲä¾Ľ

0.28

åĨľä¸ļç§ĳæĬĢ

0.27

 deserved

0.26

lected

0.26

gia

0.25

 dearly

0.24

æĬķèµĦé¢Ŀ

0.24

ÑĤÐ¾Ðº

0.24

(BuildContext

0.24

çĽ´æİ¥å½±åĵį

0.24

Activations Density 15.598%

-ing and past-tense verbs

No Known Activations