INDEX

Explanations

m

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

-m

-0.29

mÄ±

-0.28

 mÄ±

-0.28

_mr

-0.27

mPid

-0.25

éĢĦ

-0.25

ä¸įåĲĪæł¼

-0.25

AUTO

-0.25

 weather

-0.24

POSITIVE LOGITS

kep

0.27

 Braz

0.26

_keeper

0.26

èĹ»

0.25

 Mostly

0.25

unct

0.25

 Imag

0.24

 Entire

0.24

EA

0.24

å¾Īéĩįè¦ģçļĦ

0.24

Activations Density 0.227%

m

No Known Activations