INDEX

Explanations

th

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_24.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.24.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nett

-0.27

æĸ°é¢ĸ

-0.26

errat

-0.25

ths

-0.24

nda

-0.23

åħ³æĢĢ

-0.23

 scratched

-0.23

elts

-0.23

icy

-0.23

æķĪçĽĬ

-0.23

POSITIVE LOGITS

 century

0.36

 anniversary

0.36

 percentile

0.35

 Anniversary

0.35

 edition

0.34

-floor

0.33

irty

0.31

-century

0.31

ousand

0.29

 stage

0.29

Activations Density 1.381%

th

No Comments

No Known Activations