INDEX

Explanations

No Explanations Found

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

çļĦçĲĨçĶ±

-0.29

 household

-0.28

oltage

-0.25

age

-0.24

 slow

-0.24

å¸Ļ

-0.24

age

-0.24

ients

-0.24

 sparks

-0.24

ç¨ĭåº¦

-0.24

POSITIVE LOGITS

 Neck

0.28

ProcessEvent

0.28

 Grades

0.27

ÑĢÐ°Ð¹

0.27

FLT

0.26

CONF

0.25

!--

0.25

AIM

0.25

,↵↵↵↵

0.25

AIM

0.24

Activations Density 11.287%

No Known Activations

This feature has no known activations.