INDEX

Explanations

Unstructured text

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

 soft

-0.37

å¥³åŃĲ

-0.36

 Soft

-0.35

 lady

-0.35

å¥³æĸ¹

-0.35

è½¯

-0.35

soft

-0.34

 softer

-0.33

å¨ĺ

-0.32

å¥³æĢ§

-0.32

POSITIVE LOGITS

çļĦçĶ·äºº

0.43

ä¸Īå¤«

0.35

çĶ·äºº

0.34

çĶ·ä¸»è§Ĵ

0.34

èĢģåħ¬

0.33

çĶ·

0.32

çĶ·æĢ§

0.32

çĶ·æĸ¹

0.32

çĶ·åıĭ

0.32

 Ð¼ÑĥÐ¶

0.32

Activations Density 0.001%

Unstructured text

No Comments

No Known Activations