INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_2.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.2.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

çľĭå¾ħ

-0.28

å¾Ģ

-0.25

-Feb

-0.25

 marginal

-0.24

-Jul

-0.24

atÃ³

-0.24

 standard

-0.24

_Pods

-0.24

 FEATURE

-0.23

åĨ·

-0.23

POSITIVE LOGITS

ulti

0.29

vide

0.24

its

0.24

die

0.23

el

0.23

_match

0.23

å¾Īæĥ³

0.23

Updater

0.23

.es

0.23

 stresses

0.23

Activations Density 35.971%

No Known Activations

This feature has no known activations.