INDEX

Explanations

people's last names or titles followed by "said"

oai_token-act-pair · gemini-2.0-flash

news/research articles

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_12/width_16k/average_l0_6

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

TestingModule

-0.56

uent

-0.45

ChromeDriver

-0.44

 sociaux

-0.44

 كومونز

-0.43

RegistryLite

-0.43

saco

-0.43

 sociais

-0.42

้าง

-0.41

stretch

-0.41

POSITIVE LOGITS

 typelib

0.75

arraycopy

0.65

 ComVisible

0.65

JpaRepository

0.63

GraphicsUnit

0.61

tvguidetime

0.61

 CascadeType

0.60

 مشين

0.60

 greateſt

0.59

 WaitForSeconds

0.58

Activations Density 0.512%

people's last names or titles followed by "said"

news/research articles

No Comments

No Known Activations