INDEX

Explanations

structured data and numerical representations

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

medical statistics counts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 avoient

-0.49

 étoient

-0.44

 voulons

-0.44

 fédé

-0.37

 يتيمه

-0.37

 piş

-0.36

leyebilirsiniz

-0.36

hanced

-0.35

 navegación

-0.35

 pouvons

-0.34

POSITIVE LOGITS

extAlignment

0.47

RectangleBorder

0.46

原始内容存档于

0.45

expandindo

0.45

MemoryWarning

0.45

XmlAccessorType

0.44

verifyException

0.44

 crossorigin

0.42

queryInterface

0.42

↘

0.42

Activations Density 0.316%

structured data and numerical representations

medical statistics counts

No Comments

No Known Activations