INDEX

Explanations

family

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

Not in Any Lists

No Comments

Negative Logits

ecessarily

-0.27

AccessException

-0.25

IGH

-0.25

å°ĳ

-0.24

peed

-0.24

istine

-0.24

.Material

-0.23

vette

-0.23

upon

-0.23

cest

-0.23

POSITIVE LOGITS

 Hyde

0.27

dale

0.27

 calloc

0.26

ping

0.26

mor

0.25

 cosine

0.25

é¡ĺæĦı

0.24

horn

0.24

 wanted

0.23

 palindrome

0.23

Activations Density 0.016%

family

No Comments

No Known Activations