INDEX

Explanations

computer code/technical

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mange

-0.27

 tá»ĳ

-0.25

èĪĽ

-0.25

_async

-0.24

ovel

-0.24

åħĪéĶĭ

-0.24

_rat

-0.24

Scheduler

-0.24

orte

-0.23

nap

-0.23

POSITIVE LOGITS

0.34

æ°´å¹³

0.29

no

0.28

hus

0.28

 honors

0.27

es

0.27

 Middle

0.27

ers

0.26

 grade

0.25

Activations Density 0.007%

computer code/technical

No Comments

No Known Activations