INDEX

Explanations

completeness

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

è¹²

-0.25

äº²åĪĩ

-0.24

éĵ¾

-0.24

 Webcam

-0.24

 reprint

-0.24

 Chains

-0.23

edis

-0.23

 spaghetti

-0.23

 Weed

-0.23

 STRICT

-0.23

POSITIVE LOGITS

EPS

0.29

æ¸¥

0.28

 parks

0.28

pite

0.27

 park

0.26

elah

0.26

erver

0.26

fst

0.25

çłĢ

0.25

èĪ±

0.25

Activations Density 0.001%

completeness

No Comments

No Known Activations