INDEX

Explanations

chemistry and code

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

format

-0.31

nab

-0.28

table

-0.27

çĽ´

-0.25

ourn

-0.25

 Cure

-0.25

AB

-0.24

IRECT

-0.24

ogram

-0.24

æľ½

-0.24

POSITIVE LOGITS

ovsky

0.25

å©µ

0.24

æľªæĿ¥

0.24

 ÄĲá»©c

0.23

 Rene

0.23

æ³ķä»¤

0.23

dere

0.23

å¯©

0.23

 duro

0.23

ophobic

0.23

Activations Density 0.027%

chemistry and code

No Comments

No Known Activations