INDEX

Explanations

contrasting words

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

SCR

-0.37

å§¥

-0.31

SCR

-0.30

 present

-0.26

CTR

-0.25

CCR

-0.25

å¦Ĩ

-0.25

çŀ¿

-0.24

):

-0.24

/release

-0.24

POSITIVE LOGITS

é¬¯

0.26

kes

0.26

å°ĳäºĨ

0.25

irk

0.25

è·Ŀ

0.25

plevel

0.25

.compareTo

0.24

ped

0.24

\grid

0.24

singular

0.24

Activations Density 0.001%

contrasting words

No Comments

No Known Activations