INDEX

Explanations

Acronyms/initials/names/coding

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_19.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.19.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

equalsIgnoreCase

-0.27

wa

-0.27

iras

-0.26

 distr

-0.25

äºİ

-0.25

ama

-0.25

reb

-0.24

ya

-0.24

conut

-0.24

que

-0.24

POSITIVE LOGITS

ucer

0.26

PY

0.26

azzi

0.26

race

0.25

PY

0.24

æ§Į

0.24

 metre

0.24

_ALLOW

0.24

 touch

0.24

py

0.23

Activations Density 0.022%

Acronyms/initials/names/coding

No Comments

No Known Activations

Acronyms/initials/names/coding

No Comments

No Known Activations