INDEX

Explanations

scientific texts

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 choices

-0.29

 Choices

-0.28

jectives

-0.28

å·§

-0.26

å¼±çĤ¹

-0.26

choices

-0.26

++]

-0.26

inspection

-0.25

 choice

-0.25

++.

-0.25

POSITIVE LOGITS

Ð¸Ñı

0.26

ä¼¸åĩº

0.26

æ¤ħ

0.26

çļĦå¤§

0.25

ä¸¤åı¥

0.25

horn

0.25

vik

0.25

 terrestrial

0.25

à¸ªà¸²

0.25

åĲ¬åĪ°

0.24

Activations Density 1.666%

scientific texts

No Comments

No Known Activations