INDEX

Explanations

completeness

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

 PARAMETERS

-0.27

 thÆ°

-0.26

itored

-0.25

isans

-0.25

-Origin

-0.24

REC

-0.24

.Weight

-0.24

aris

-0.24

.tw

-0.24

ikh

-0.23

POSITIVE LOGITS

è°Ī

0.29

 Hart

0.28

çĤ¼

0.28

Py

0.27

á»©ng

0.26

 hart

0.25

hart

0.25

 part

0.25

hack

0.25

Activations Density 0.010%

completeness

No Comments

No Known Activations