INDEX

Explanations

once

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

Not in Any Lists

No Comments

Negative Logits

now

-0.41

çİ°åľ¨

-0.39

[now

-0.35

Now

-0.34

now

-0.32

ä½Ĩçİ°åľ¨

-0.32

Now

-0.32

now

-0.31

èĢĮçİ°åľ¨

-0.31

NOW

-0.31

POSITIVE LOGITS

 again

0.96

again

0.87

Again

0.83

 Again

0.82

 AGAIN

0.71

åĨįæ¬¡

0.68

_again

0.65

 novamente

0.59

 ÑģÐ½Ð¾Ð²Ð°

0.59

 lagi

0.56

Activations Density 0.798%

once

No Comments

No Known Activations