INDEX

Explanations

important keywords and phrases associated with action or status changes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

recent times, significantly higher

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Biôgrafia

-0.58

السكان

-0.53

 ویکی‌پدی

-0.50

:✨

-0.49

IUrlHelper

-0.49

 Савезне

-0.49

లాలు

-0.48

 podjela

-0.46

 Italijani

-0.46

 Waray

-0.46

POSITIVE LOGITS

 disambiguazione

0.47

 trouw

0.46

 ujednoznacz

0.46

 lạc

0.43

AddTagHelper

0.43

∭

0.43

MockBean

0.40

According

0.40

Autowired

0.40

참고

0.39

Activations Density 0.005%

important keywords and phrases associated with action or status changes

recent times, significantly higher

No Comments

No Known Activations