INDEX

Explanations

two or three

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_30.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.30.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 scrub

-0.26

 Confirmation

-0.26

mite

-0.26

ighted

-0.25

ç»Ł

-0.25

å¤©èĬ±

-0.25

Uniform

-0.24

é©¼

-0.24

Coc

-0.24

ankind

-0.24

POSITIVE LOGITS

 parallel

0.42

 sides

0.35

 basic

0.35

 side

0.34

 elements

0.34

parallel

0.32

 basics

0.32

 fundamentals

0.32

 paralle

0.31

 fundamental

0.31

Activations Density 0.887%

two or three

No Known Activations