INDEX

Explanations

records

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.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

çĽĳæĬ¤

-0.28

muj

-0.26

species

-0.26

 Corp

-0.25

effective

-0.25

istine

-0.25

Ens

-0.24

igm

-0.24

ç«ĭ

-0.24

æĬĹæĭĴ

-0.24

POSITIVE LOGITS

inality

0.28

ÑĢÑĥÑĩ

0.27

untime

0.26

ivia

0.25

æľīä»Ģä¹Ī

0.25

 toll

0.25

terminated

0.25

 trails

0.25

 terminated

0.24

 installed

0.24

Activations Density 0.029%

records

No Comments

No Known Activations