INDEX

Explanations

people

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pil

-0.27

æīĢ

-0.26

å¥¹åľ¨

-0.26

Pil

-0.26

å®¢è§Ĥ

-0.25

å·®

-0.25

åı¯çĸĳ

-0.25

èĩĨ

-0.25

å·®å¼Ĥ

-0.25

estate

-0.25

POSITIVE LOGITS

Principal

0.35

 Principal

0.30

è½

0.28

æŃ¤å¤ĸ

0.28

igar

0.27

è¿ĺåĮħæĭ¬

0.27

alesce

0.27

principal

0.27

è¿ĺæĥ³

0.27

è¿ĺä¼ļ

0.26

Activations Density 0.002%

people

No Comments

No Known Activations