INDEX

Explanations

Lists or e-commerce contexts

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_32.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æĶ¾è¿Ľ

-0.27

-build

-0.27

éĥ¡

-0.26

æµģéĢļ

-0.25

 unins

-0.25

 mail

-0.24

.gmail

-0.23

æ±©

-0.23

 build

-0.23

Build

-0.23

POSITIVE LOGITS

clo

0.31

esp

0.29

senal

0.27

spe

0.26

å»¶ç»Ń

0.26

uria

0.25

å¯¨

0.25

LEN

0.25

ilege

0.25

 Sant

0.25

Activations Density 0.001%

Lists or e-commerce contexts

No Comments

No Known Activations