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

egment

-0.26

Ea

-0.25

 puss

-0.25

Ð¸Ð¶

-0.24

(Parcel

-0.24

parated

-0.24

 meds

-0.24

/simple

-0.24

 kÃ¶n

-0.24

è¯´ä»Ģä¹Ī

-0.23

POSITIVE LOGITS

 reference

0.32

 references

0.28

source

0.28

 source

0.27

_published

0.26

.reference

0.25

 rÃ©fÃ©rence

0.25

è§īå¾Ĺå¾Ī

0.25

 authoritative

0.25

obo

0.25

Activations Density 0.000%

No Comments

No Known Activations