Update README.md

README.md

@@ -23,20 +23,22 @@ app before March 7, 2023.
 ### Sources
 
 - **Repository:** [Open-Assistant/model/model_training](https://github.com/LAION-AI/Open-Assistant/tree/main/model/model_training)
-- **Demo:** View Assistant replies for 250 random prompts [here](https://open-assistant.github.io/oasst-model-eval/?f=https%3A%2F%2Fraw.githubusercontent.com%2FOpen-Assistant%2Foasst-model-eval%2Fmain%2Fsampling_reports%2Foasst-sft%2F2023-03-09_andreaskoepf_oasst-1_12b_7000_sampling_noprefix_lottery.json) ([sampling code](https://github.com/Open-Assistant/oasst-model-eval/blob/3d71f3be100c05cd8ddb568365e036a29fbff8c7/model_eval/manual/sampling_report.py)).
+- **Demo:** [Continuations for 250 random prompts](https://open-assistant.github.io/oasst-model-eval/?f=https%3A%2F%2Fraw.githubusercontent.com%2FOpen-Assistant%2Foasst-model-eval%2Fmain%2Fsampling_reports%2Foasst-sft%2F2023-03-09_andreaskoepf_oasst-1_12b_7000_sampling_noprefix_lottery.json) ([sampling code](https://github.com/Open-Assistant/oasst-model-eval/blob/3d71f3be100c05cd8ddb568365e036a29fbff8c7/model_eval/manual/sampling_report.py)).
 
-### Reply Generation
+### Prompting
 
 Two special tokens are used to mark the beginning of user and assistant turns:
 `<|prompter|>` and `<|assistant|>`. Each turn ends with a `<|endoftext|>` token.
 
-Input example:
+Input prompt example:
 ```
-<|prompter|>What is a meme, and what's the history behind this word?
-<|endoftext|><|assistant|>
+<|prompter|>What is a meme, and what's the history behind this word?<|endoftext|><|assistant|>
 ```
+The input ends with the `<|assistant|>` token to signal that the model should 
+start generating the assistant reply.
 
-Generated continuation:
+
+Example of generated continuation (typical sampling 0.2):
 ```
 A visual joke often accompanied by text that spreads rapidly from individual
 to individual. The exact origin of the term "meme" is unclear, but it was
@@ -56,7 +58,9 @@ fields ranging from anthropology to psychology to marketing.<|endoftext|>
 
 The model is known to fail horribly at answering math and coding questions.
 
-Beware of hallucinations: Outputs are often factually wrong or misleading. Replies might look convincing (at first glance) while containing completely made up false statements.
+Beware of hallucinations: Outputs are often factually wrong or misleading. 
+Replies might look convincing (at first glance) while containing completely 
+made up false statements.
 
 This model is usable only for English conversations.