INDEX

Explanations

things that look like citations including brackets, numbers, and symbols like $, \, and ""

oai_token-act-pair · gemini-2.0-flash

Scientific/academic text

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 flashes

-0.60

Brightness

-0.59

LIOT

-0.57

 Spotlight

-0.57

 Lights

-0.56

 brightness

-0.56

 flash

-0.56

ANCHE

-0.56

 lights

-0.55

 Watt

-0.54

POSITIVE LOGITS

kháu

0.63

 gynhyrchwyd

0.59

SequentialGroup

0.58

 Italijanski

0.57

רום

0.56

otoro

0.55

NEA

0.52

 materiais

0.52

illère

0.52

 berea

0.50

Activations Density 1.403%

things that look like citations including brackets, numbers, and symbols like $, \, and ""

Scientific/academic text

No Comments

No Known Activations