INDEX

Explanations

Code/foreign language snippets

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_32.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.32.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

emp

-0.33

 perfect

-0.28

æī¯

-0.28

 perfectly

-0.27

æ¡§

-0.27

Perfect

-0.26

perfect

-0.26

 medios

-0.25

éĺŁåĳĺä»¬

-0.25

 Perfect

-0.25

POSITIVE LOGITS

çŃ·

0.28

ä¸°å¯Į

0.28

å¤§çļĦ

0.28

 part

0.28

 affluent

0.27

ä¸°å¯ĮçļĦ

0.27

çļĦéĩįå¤§

0.27

 rich

0.26

 hind

0.26

hof

0.25

Activations Density 0.001%

Code/foreign language snippets

No Known Activations