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

Negative Logits

-0.31

 infiltr

-0.26

åį¤

-0.26

.emplace

-0.26

_states

-0.26

 until

-0.26

 bene

-0.25

 nitrogen

-0.25

æ®´

-0.25

 hateful

-0.25

POSITIVE LOGITS

(KP

0.31

(SP

0.27

 GENERIC

0.26

 UTIL

0.26

YYY

0.26

BAT

0.26

 ÐļÐ°

0.26

nerRadius

0.26

(SC

0.26

âĢ¦)↵↵

0.26

Activations Density 0.000%

No Known Activations

This feature has no known activations.