INDEX

Explanations

code-related terms and constructs in programming, particularly in Java

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

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

onOptions

-0.49

transQ

-0.44

StructEnd

-0.42

 scissor

-0.42

BarStyle

-0.41

 runt

-0.41

StreetMap

-0.41

TagMode

-0.40

scryfall

-0.40

 PeEnEo

-0.39

POSITIVE LOGITS

 المعيارى

0.60

RTLU

0.49

:✨

0.44

 mêmes

0.42

PYX

0.42

 Italijani

0.38

Efq

0.38

ⓧ

0.37

webElementXpaths

0.37

تقاوى

0.36

Activations Density 0.341%

code-related terms and constructs in programming, particularly in Java

No Comments

No Known Activations