INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ä½Ĩè¿Ļ

-0.29

ea

-0.26

ead

-0.26

 Conv

-0.25

oto

-0.25

eft

-0.25

 ReturnValue

-0.24

 Lower

-0.24

eq

-0.23

çĽ¸åħ³

-0.23

POSITIVE LOGITS

 Christina

0.25

-encoded

0.24

imens

0.24

_actual

0.23

-covered

0.23

èįļ

0.23

 profes

0.23

enced

0.23

mland

0.22

pps

0.22

Activations Density 0.032%

No Known Activations

This feature has no known activations.