INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.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

éĺ²æ±Ľ

-0.28

Ð¾ÑĢÐ°

-0.26

.Mode

-0.25

arching

-0.25

-responsive

-0.24

airo

-0.24

èĢĥéªĮ

-0.24

 FieldType

-0.24

]↵↵

-0.23

ais

-0.23

POSITIVE LOGITS

atisfied

0.26

æŁ

0.25

 fits

0.25

ogi

0.25

 Raptors

0.24

lander

0.24

 mats

0.24

æįķ

0.23

åī§

0.23

 dram

0.23

Activations Density 1.381%

No Known Activations

This feature has no known activations.