INDEX

Explanations

out

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 forth

-0.28

onaut

-0.26

ä¾

-0.25

oteric

-0.25

unate

-0.24

rin

-0.24

ek

-0.23

ovation

-0.23

oft

-0.23

ogne

-0.23

POSITIVE LOGITS

å¨ģèĥģ

0.27

é£İéĻ©ç®¡çĲĨ

0.27

çī¹åĪ«æĺ¯

0.26

é£İéĻ©

0.26

icularly

0.26

çģ¾

0.26

_IMPORTED

0.25

 especially

0.25

-bs

0.24

åĪĴ

0.24

Activations Density 0.036%

out

No Comments

No Known Activations