Large language models have led to remarkable progress on many NLP tasks, and researchers are turning to ever-larger text corpora to train them. Some of the largest corpora available are made by scraping significant portions of the internet, and are f
requently introduced with only minimal documentation. In this work we provide some of the first documentation for the Colossal Clean Crawled Corpus (C4; Raffel et al., 2020), a dataset created by applying a set of filters to a single snapshot of Common Crawl. We begin by investigating where the data came from, and find a significant amount of text from unexpected sources like patents and US military websites. Then we explore the content of the text itself, and find machine-generated text (e.g., from machine translation systems) and evaluation examples from other benchmark NLP datasets. To understand the impact of the filters applied to create this dataset, we evaluate the text that was removed, and show that blocklist filtering disproportionately removes text from and about minority individuals. Finally, we conclude with some recommendations for how to created and document web-scale datasets from a scrape of the internet.
In this article we present a new and effective method to Estimate the static fatigue age for
springs, when it works by static load or static strain for a long time at deferent
temperature, this case is very important because the spring is a battery
for force and it must
save allow value of the force.
At the first we present a theory study for “creep and relaxation “and how we can accelerate
the test to get the results, then we suggest method to accelerate the test to know when the
spring retch to the minimum allowable value of the force at work temperature degree, this
method depend on tests at samples at deferent temperature degrees, higher than work and
storage temperature, and then use the results to estimate the static fatigue age for springs at
real storage and work temperature.
This study was related to the urban development of Homs City and
its surrounding between 1972 – 2010 by using remote sensing
technique and geographic information system. The results of the
study indicated that the size of the city was increased 60
0% times ,
randomly and eliminating most of crop lands.
Fiber reinforced polymer (FRP) plates have gained popularity in the strengthening of reinforced concrete (RC) members due to the high strength to weight ratio, the ease of installation and low maintenance costs compared to other systems such as steel
plates. Also, external bonding of fiber reinforced polymer (FRP) plate has been proven to be an effective method to strengthen and damage RC structures. However, not much attention has been given to the long-term behavioral aspects of FRP-strengthened RC members. It is difficult to accurately predict the long-term deflection, crack width and behavior of FRP-strengthened RC members.
This study presents a method for predicting the deflections at any time of the service life of normal and high strength concrete structures, loaded at any age, creep coefficient and shrinkage strain and the participation of the compressive steel for RC beams, and for RC beams strengthened with FRP plates.
The accurate prediction of deflections is a complex problem which requires the use of non-linear and time-dependent analytical methods. These methods are, in general, time consuming and require great effort[6-13]. However, at the design stage, simple but reliable methods which take into account the most important parameters influencing the long-term deflections may be very useful to adequately design the structure. For that purpose, many simplified methods have been developed [1-2-3-4-14]. Equations have been programmed to access the method combines the simplicity and accuracy and provides valuable information about the influence of each parameter on the increasing deflections with time.
The analytical values are compared to the experimental results from some existing papers [8-9], and to the results obtained from ACI code [1]. It is found that the analytical method is in a good agreement with the experimental results from some existing papers
[8-9].
