INDEX

Explanations

Death and resurrection

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

ç½ĳç«Ļåľ°åĽ¾

-0.27

 fins

-0.26

âĬĶ

-0.26

 ksztaÅĤt

-0.26

imoto

-0.25

itzer

-0.25

_MSK

-0.25

 Anders

-0.24

 retali

-0.23

à¸Ļà¸²à¸¡

-0.23

POSITIVE LOGITS

 sunrise

0.28

resenter

0.27

eos

0.27

è®°å¿ĨåĬĽ

0.26

visibility

0.26

æ»¨å·ŀ

0.24

 visibility

0.24

 sufficiently

0.24

_bn

0.24

.wp

0.24

Activations Density 0.001%

Death and resurrection

No Comments

No Known Activations