INDEX

Explanations

technical and foreign language text

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Predictor

-0.27

 Isle

-0.27

picker

-0.26

 Picker

-0.26

Spinner

-0.26

_sensitive

-0.25

ierz

-0.25

oline

-0.25

æĢĢæĬ±

-0.25

oglob

-0.25

POSITIVE LOGITS

å®¡è®®

0.31

uya

0.30

 turret

0.27

æĿĲæĸĻ

0.27

 Trustees

0.27

 material

0.26

èī¯

0.26

 trÃº

0.25

EDA

0.25

æĿĲ

0.25

Activations Density 0.000%

technical and foreign language text

No Comments

No Known Activations