INDEX

Explanations

Code snippets

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_35.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.35.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

å¿«è®¯

-0.25

 Route

-0.25

ä½ķç§į

-0.24

 localize

-0.24

åĹ£

-0.24

UNET

-0.23

æľŁå¾ħ

-0.23

åī¯ç§ĺä¹¦éķ¿

-0.23

 randomized

-0.23

 Armed

-0.23

POSITIVE LOGITS

ee

0.27

egrator

0.27

ä¾¯

0.27

å°±å¾Ī

0.26

å¾Ī

0.26

izi

0.26

å°±æľīäºĨ

0.25

rogen

0.25

enburg

0.25

Activations Density 37.506%

Code snippets

No Known Activations