INDEX

Explanations

send

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_33.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.33.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Ð¾Ð½Ð¸

-0.37

 THEY

-0.37

They

-0.36

they

-0.35

 ÐŀÐ½Ð¸

-0.33

è¿Ļæīįæĺ¯

-0.32

ä»ĸä»¬

-0.31

 They

-0.31

 they

-0.29

sending

-0.29

POSITIVE LOGITS

out

0.38

him

0.36

 ther

0.31

 videos

0.29

 them

0.28

 questions

0.28

 queries

0.28

off

0.28

Him

0.28

 cups

0.28

Activations Density 0.031%

send

No Comments

No Known Activations