INDEX

Explanations

Code and forum posts

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

 diversity

-0.27

ruptions

-0.26

åŃĹåı·

-0.26

 bitten

-0.26

ä¸¤ç±»

-0.24

ç»¼åĲĪä½ĵ

-0.24

PTION

-0.24

ä¹Ĵ

-0.24

chest

-0.23

RECT

-0.23

POSITIVE LOGITS

å¤¹

0.28

 ///<

0.26

.executeQuery

0.26

ÅĽcie

0.25

gement

0.25

å¼¦

0.24

 Pratt

0.24

opa

0.24

 Void

0.24

loy

0.23

Activations Density 0.001%

Code and forum posts

No Comments

No Known Activations