INDEX

Explanations

mention of specific characters or entities in various contexts

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

names with nicknames

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

 nakalista

-0.40

⅘

-0.40

Rüyada

-0.40

OrBuilder

-0.39

Pyx

-0.39

BeforeEach

-0.38

tinyos

-0.37

utica

-0.36

COG

-0.36

 bambú

-0.36

POSITIVE LOGITS

TemporalType

0.47

简称

0.46

XmlRootElement

0.46

 ویکی‌پدیای

0.42

WEBPACK

0.41

 nicknamed

0.40

 thread

0.39

etCode

0.39

 nicknames

0.39

 called

0.38

Activations Density 0.390%

mention of specific characters or entities in various contexts

names with nicknames

No Comments

No Known Activations