INDEX

Explanations

questions and code snippets

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_32.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ä¸¤æĿ¡

-0.26

èµ°

-0.26

åĬŀ

-0.26

illes

-0.26

endi

-0.25

passed

-0.24

 Chandler

-0.24

èĤ¤èī²

-0.24

 Conte

-0.23

HC

-0.23

POSITIVE LOGITS

tie

0.34

Tie

0.29

ieg

0.28

tie

0.26

åľĪ

0.26

 reach

0.26

iti

0.25

xD

0.24

 Bubble

0.24

igy

0.24

Activations Density 0.001%

questions and code snippets

No Known Activations