update README + typos

Author: VictorSanh

CONTRIBUTING.md

@@ -7,7 +7,7 @@ One of the best ways to contribute is by writing templates!
 A template is a piece of code written in a templating language called
 [Jinja](https://jinja.palletsprojects.com/en/3.0.x/). A template defines
 a function that maps an example from a dataset in the
-[HuggingFace library](https://huggingface.co/datasets) to two strings of
+[Hugging Face datasets library](https://huggingface.co/datasets) to two strings of
 text. The first is called the _prompt_ which provides all information that
 will be available to solve a task, such as the instruction and the context.
 The second piece is called the _output_, which is the desired response to the
@@ -23,7 +23,7 @@ and find an unclaimed one. Put your name under "Who's Prompting it?" and
 mark it yellow to show it's in progress.
 1. **Examine the dataset.** Select or type the dataset into the dropdown in the app.
 If the dataset has subsets (subsets are not the same as splits), you can select
-which one to work on. Note that templates are subset specific. You can find
+which one to work on. Note that templates are subset-specific. You can find
 out background information on the dataset by reading the information in the
 app. The dataset is a collection of examples, and each example is a Python
 dictionary. The sidebar will tell you the schema that each example has.
@@ -42,10 +42,10 @@ applied to the current example will appear in the right sidebar.
 through a handful of examples of the prompted dataset using the
 "Prompted dataset viewer" mode.
 1. **Write between 5 and 10 templates**. Repeat the steps 4 to 8 to create between 5
-and 10 (more if you want!) templates per dataset. Feel free to introduce some diversity
+and 10 (more if you want!) templates per dataset/subset. Feel free to introduce some diversity
 both in the format and the formulation.
 1. **Duplicate the template(s).** If the dataset you have chosen bear the same
-format as other datasets (for instance `MNLI` and `SNLI` have identical format),
+format as other datasets (for instance, `MNLI` and `SNLI` have identical formats),
 you can simply claim these datasets and duplicate the templates you have written
 to these additional datasets.
 1. **Upload the template(s).** Submit a PR using the instructions
@@ -56,7 +56,7 @@ to these additional datasets.
 Here is a quick crash course on using [Jinja](https://jinja.palletsprojects.com/en/3.0.x/)
 to write templates. More advanced usage is in the [cookbook](#jinja-cookbook).
 
-Generally in a template, you'll want to use a mix of hard-coded data that is
+Generally, in a template, you'll want to use a mix of hard-coded data that is
 task-specific and stays the same across examples, and commands that tailor the
 prompt and output to a specific example.
 
@@ -115,7 +115,7 @@ Is this a piece of news regarding {{"world politics"}}, {{"sports"}}, {{"busines
 
 A few miscellaneous things:
 
-* **Writing outputs.** When writing a template for an task that requires outputting
+* **Writing outputs.** When writing a template for a task that requires outputting
 a label, don't use articles or other stop words before the label name in the output.
 For example, in TREC, the output should be "Person", not "A person". The reason
 is that evaluations often look at the first word of the generated output to determine
@@ -145,7 +145,7 @@ Is this question asking for a {{"definition"}}, a {{"description"}}, a {{"manner
 ```
 * **Conditional generation format.** Always specify the output label `y` and separate it from the prompt
 by indicating the vertical bars `|||`. The `y` will be generated by a generative model
-conditionned on the prompted input you wrote. You can always transform an "infix" prompt format
+conditioned on the prompted input you wrote. You can always transform an "infix" prompt format
 ```jinja2
 Given that {{premise}}, it {{ ["must be true", "might be true", "must be false"][label] }} that {{hypothesis}}
 ```
@@ -158,7 +158,7 @@ Given that {{premise}}, it {{ "must be true, might be true, or must be false" }}
 ## Uploading Templates
 
 Once you save or modify a template, the corresponding file inside the `templates`
-directory in the repo will be modified. To upload it, following these steps:
+directory in the repo will be modified. To upload it, follow these steps:
 1. Run `make style` and `make quality`.
 2. Commit the modified template files (anything under `templates`) to git.
 3. Push to your fork on GitHub.

README.md

@@ -19,6 +19,13 @@ There are 3 modes in the app:
 - **Prompted dataset viewer**: check the templates you wrote or already written on entire dataset
 - **Sourcing**: write new prompts
 
+<img src="assets/promptsource_app.png" width="800">
+
+## Contributing
+Join the **Hackaprompt** and help writing templates!
+
+Contribution guidelines and step-by-step *HOW TO* are described [here](CONTRIBUTING.md).
+
 ## Writing Templates
 A prompt template is expressed in [Jinja](https://jinja.palletsprojects.com/en/3.0.x/).
 
@@ -28,28 +35,17 @@ and output. Generally, the prompt should provide information on the desired beha
 e.g., text passage and instructions, and the output should be a desired response.
 
 Here's an example for [AG News](https://huggingface.co/datasets/ag_news):
-```
+```jinja
 {{text}}
-Is this a piece of news regarding world politics, sports, business, or technology? |||
-{{ ["World politics", "Sport", "Business", "Technology"][label] }}
+Is this a piece of news regarding {{"world politics"}}, {{"sports"}}, {{"business"}}, or {{"science and technology"}}? |||
+{{ ["World politics", "Sports", "Business", "Science and technology"][label] }}
 ```
 
-## Contributing
-This is very much a work in progress, and help is needed and appreciated. Anyone wishing to
-contribute code can contact Steve Bach for commit access, or submit PRs from forks. Some particular
-places you could start:
-1. Try to express things! Explore a dataset and tell us what's hard to do to create templates you want
-2. Look in the literature. Are there prompt creation methods that do/do not fit well right now?
-3. Scalability testing. Streamlit is lightweight, and we're reading and writing all prompts on refresh.
-
-See also the [design doc](https://docs.google.com/document/d/1IQzrrAAMPS0XAn_ArOq2hyEDCVfeB7AfcvLUqgSnWxQ/).
-
-Before submitting a PR or pushing a new commit, please run style formattings and quality checks so that your newly added file look nice:
-```bash
-make style
-make quality
-```
+For more information, read the [Contribution guidelines](CONTRIBUTING.md).
 
+## Other documentation
+- Experimental context: [Experiment D](https://docs.google.com/document/d/1ar9cTRs9ZWxMkW-_9jF1ZtfY-u8IK_sO5QOtq1qnC2I/edit?usp=sharing)
+- Prompting interface: [design doc](https://docs.google.com/document/d/1IQzrrAAMPS0XAn_ArOq2hyEDCVfeB7AfcvLUqgSnWxQ/)
 ## Known Issues
 
 **Warning or Error about Darwin on OS X:** Try downgrading PyArrow to 3.0.0.