INDEX

Explanations

with

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_7.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.7.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

iali

-0.28

çĸıå¯¼

-0.28

TAR

-0.27

 mate

-0.25

Rem

-0.25

 reap

-0.24

iki

-0.24

iate

-0.24

etch

-0.24

REM

-0.24

POSITIVE LOGITS

çĳŁ

0.30

ä»ĸçļĦ

0.30

äºĨä»ĸçļĦ

0.30

æĺ¯ä»ĸ

0.28

ä»ĸæĺ¯

0.27

 giÃ³

0.27

 winds

0.27

ä½Ĩæĺ¯ä»ĸ

0.26

é£İ

0.26

venir

0.25

Activations Density 0.002%

with

No Comments

No Known Activations