INDEX

Explanations

mirrors and symmetry

np_max-act · gemini-2.0-flash

references to mirroring or symmetry—concepts of reflection, parity, or equal counterparts.

oai_token-act-pair · gpt-5 Triggered by @liuliujiujiu05

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_0.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.0.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

TokenType

-0.29

Tup

-0.26

osit

-0.26

ä¿ĿèŃī

-0.25

[]>↵

-0.25

éģĵå£«

-0.24

åºĹä¸»

-0.24

 messageType

-0.24

lar

-0.23

é¥ķ

-0.23

POSITIVE LOGITS

“

0.31

åĪĨ

0.30

assa

0.26

ilor

0.26

noc

0.25

nett

0.25

isson

0.24

äºĶæľĪ

0.24

-side

0.24

 robin

0.23

Activations Density 5.066%

mirrors and symmetry

references to mirroring or symmetry—concepts of reflection, parity, or equal counterparts.

No Comments

No Known Activations