INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_18.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.18.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÑĪÐµÐ¹

-0.27

loom

-0.26

 Jefferson

-0.25

ieu

-0.24

 laundering

-0.24

era

-0.24

 campaign

-0.24

 circ

-0.24

åºĦ

-0.24

 Campaign

-0.23

POSITIVE LOGITS

æľīçļĦ

0.26

iling

0.26

eg

0.25

omit

0.25

pas

0.25

theValue

0.25

uds

0.24

è®²ç©¶

0.24

ego

0.24

_pairs

0.24

Activations Density 9.979%

No Known Activations

This feature has no known activations.