INDEX

Explanations

spr

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_11.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.11.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

olumn

-0.30

æ½ĩ

-0.28

 canned

-0.26

 occult

-0.26

-kind

-0.26

æĶ¹å»º

-0.26

scribed

-0.26

æĺĨä»ĳ

-0.25

åıĺçİ°

-0.25

 ministries

-0.24

POSITIVE LOGITS

fur

0.34

èĥĮä¸Ĭ

0.27

fur

0.27

utions

0.27

uv

0.26

atta

0.25

Empty

0.25

getClass

0.24

 disag

0.24

 Empty

0.24

Activations Density 0.005%

spr

No Comments

No Known Activations