INDEX

Explanations

academic texts

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_14.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.14.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

æ³ª

-0.28

osti

-0.28

ÌĢ

-0.26

 currentState

-0.26

ä¸įæĥ³

-0.26

çĭİ

-0.25

 residences

-0.25

æ±Ĺ

-0.24

ä¸ŃåĽ½ä¼łç»Ł

-0.23

è¹Ħ

-0.23

POSITIVE LOGITS

eec

0.29

åīįä»»

0.25

æĮ¨

0.25

 plated

0.24

down

0.24

SECRET

0.23

 down

0.23

 dere

0.23

äºķ

0.23

 Trey

0.23

Activations Density 0.030%

academic texts

No Known Activations