INDEX

Explanations

names and code snippets

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

reeze

-0.31

å®Ļ

-0.26

proved

-0.26

uib

-0.26

/commons

-0.26

 bÃ¨

-0.26

 views

-0.25

quee

-0.24

::$

-0.24

é¹³

-0.24

POSITIVE LOGITS

pec

0.27

elist

0.26

latex

0.25

 latex

0.25

boxes

0.25

à¹Ĥà¸ª

0.24

PEC

0.23

hei

0.23

-lat

0.23

copy

0.23

Activations Density 0.005%

names and code snippets

No Comments

No Known Activations