INDEX

Explanations

Processes, transfers, or locations

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 toward

-0.56

 towards

-0.52

èµ°åĲĳ

-0.45

 Towards

-0.43

åĪ°è¾¾

-0.41

 hacia

-0.40

æĬµè¾¾

-0.38

 reaches

-0.37

 reaching

-0.37

 reach

-0.37

POSITIVE LOGITS

åħĭæľį

0.48

 overcoming

0.46

ç©¿è¿ĩ

0.40

æĳĨèĦ±

0.40

 overcome

0.40

ç»ıè¿ĩ

0.34

æĶ¾å¼ĥäºĨ

0.34

ç©¿éĢı

0.34

æĶ¾å¼ĥ

0.33

 Skip

0.33

Activations Density 0.001%

Processes, transfers, or locations

No Known Activations