INDEX

Explanations

No Explanations Found

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

çī¹æ´¾

-0.29

.Batch

-0.27

ersed

-0.26

æ»ļåĬ¨

-0.25

æĺ¯æľĢ

-0.25

è¯¾é¢ĺ

-0.25

.mods

-0.24

 egreg

-0.24

least

-0.24

è²¬ä»»

-0.24

POSITIVE LOGITS

.sam

0.28

sam

0.27

çĽĬ

0.27

 approximate

0.26

arem

0.26

 subsequent

0.26

 short

0.25

are

0.25

 sequential

0.25

å®£

0.25

Activations Density 0.000%

No Known Activations

This feature has no known activations.