INDEX

Explanations

of

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_18.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.18.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ä»°

-0.32

LEE

-0.28

renc

-0.28

riba

-0.27

 trophy

-0.26

rence

-0.25

Named

-0.25

Url

-0.24

 Nullable

-0.24

è¥¿æ¹ĸ

-0.24

POSITIVE LOGITS

.green

0.28

cola

0.27

oping

0.26

-green

0.25

Mol

0.25

çīŁ

0.25

ç»¿èī²

0.25

mol

0.25

Mari

0.25

_green

0.24

Activations Density 11.245%

of

No Known Activations