INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_35.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.35.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

èĿī

-0.26

 bubble

-0.25

çĽħ

-0.23

cor

-0.23

/pm

-0.23

ACLE

-0.23

è´¯

-0.22

hosts

-0.22

ervlet

-0.22

 fold

-0.22

POSITIVE LOGITS

 water

0.77

water

0.76

æ°´

0.74

 Water

0.73

Water

0.70

_water

0.68

.water

0.66

 WATER

0.63

-water

0.62

 agua

0.59

Activations Density 2.582%

No Known Activations

This feature has no known activations.