INDEX

Explanations

is

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_30.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.30.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ium

-0.26

iya

-0.25

vana

-0.25

 bidi

-0.25

ersen

-0.25

å®ĥåı¯ä»¥

-0.24

çĬ¶

-0.24

 Wrapped

-0.24

Alle

-0.24

 conv

-0.23

POSITIVE LOGITS

 that

0.41

 ÑĩÑĤÐ¾

0.29

that

0.29

ä¼°è®¡

0.27

 bahwa

0.26

 ráº±ng

0.26

Åķ

0.26

Ð½ÑıÑĤ

0.25

çļĦæĺ¯

0.25

olith

0.25

Activations Density 0.010%

is

No Known Activations