INDEX

Explanations

Code and text fragments

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

Not in Any Lists

No Comments

Negative Logits

çĥĽ

-0.27

dq

-0.27

èĩĤ

-0.26

åłĩ

-0.26

#endregion

-0.25

 parcel

-0.24

×¤×¡

-0.24

è¹°

-0.24

.Spring

-0.23

ank

-0.23

POSITIVE LOGITS

odium

0.26

ç§°

0.26

aghan

0.25

-pressure

0.25

chio

0.24

è±º

0.24

iei

0.24

%C

0.24

AGMA

0.24

 mish

0.23

Activations Density 0.001%

Code and text fragments

No Comments

No Known Activations