INDEX

Explanations

of

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.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

ROOM

-0.27

åģļçļĦ

-0.26

_-

-0.26

å°ıæĹ¶åīį

-0.26

allon

-0.26

.WriteAll

-0.26

isseur

-0.25

 Sell

-0.24

è·¨çķĮ

-0.24

 StringWriter

-0.23

POSITIVE LOGITS

uh

0.25

usher

0.25

æĭĽ

0.24

aid

0.24

aks

0.24

æĮĩåĲĳ

0.23

 tighten

0.23

 aiding

0.23

 interfere

0.23

-packages

0.23

Activations Density 32.932%

of

No Comments

No Known Activations