INDEX

Explanations

Academic texts and code

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anela

-0.29

amus

-0.28

foot

-0.28

 contr

-0.27

(simp

-0.26

eren

-0.25

çĶ·

-0.24

inho

-0.23

èģĨåĲ¬

-0.23

ERC

-0.23

POSITIVE LOGITS

ä¼ĺ

0.30

_race

0.28

urities

0.26

è¾¹ä¸Ĭ

0.26

çī©ç§į

0.25

verty

0.25

 pseud

0.25

 {{↵

0.25

tlement

0.25

-translate

0.24

Activations Density 2.048%

Academic texts and code

No Comments

No Known Activations