INDEX

Explanations

abstract

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

çĻ¾ç§ĳ

-0.30

[list

-0.26

åľº

-0.26

èŀį

-0.25

acho

-0.25

ÑĨÐµÐ½

-0.25

geh

-0.24

fra

-0.23

xa

-0.23

ertiary

-0.23

POSITIVE LOGITS

0.45

 provided

0.32

squ

0.30

 Provided

0.29

provided

0.27

letters

0.26

 letters

0.26

sand

0.25

received

0.25

ernals

0.24

Activations Density 0.032%

abstract

No Known Activations