INDEX

Explanations

code

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

_typeof

-0.30

 virtual

-0.28

lify

-0.28

ãģ¦ãĤĭ

-0.26

daemon

-0.26

çķĻ

-0.26

 tail

-0.26

ymous

-0.26

 wolf

-0.25

 Bear

-0.25

POSITIVE LOGITS

æĭ¥æĮ¤

0.26

Ð°Ð·Ð²Ð°Ð½Ð¸Ðµ

0.25

æĥ¯

0.25

enus

0.24

vas

0.24

maybe

0.24

 Constraints

0.24

ÑĭÐ²Ð°ÑĤÑĮ

0.24

 Relations

0.24

ç»°

0.23

Activations Density 40.588%

code

No Comments

No Known Activations