Enhance results with prompt engineering strategies.
This guide shares strategies and tactics for getting better results from large language models (sometimes referred to as GPT models) like GPT-4o. The methods described here can sometimes be deployed in combination for greater effect. We encourage experimentation to find the methods that work best for you.
You can also explore example prompts which showcase what our models are capable of:
Prompt examples
Studying the usage of text-to-text transfer transformer to support code-related tasksA Mastropaolo, S Scalabrino, N Cooper, DN Palacio, D Poshyvanyk, ...2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE …, 2021| 249| 2021
A systematic literature review on the use of deep learning in software engineering researchC Watson, N Cooper, DN Palacio, K Moran, D PoshyvanykACM Transactions on Software Engineering and Methodology (TOSEM) 31 (2), 1-58, 2022| 117| 2022
An empirical study on the usage of bert models for code completionM Ciniselli, N Cooper, L Pascarella, D Poshyvanyk, M Di Penta, G Bavota2021 IEEE/ACM 18th International Conference on Mining Software Repositories …, 2021| 84| 2021
An empirical study on the usage of transformer models for code completionM Ciniselli, N Cooper, L Pascarella, A Mastropaolo, E Aghajani, ...IEEE Transactions on Software Engineering 48 (12), 4818-4837, 2021| 83| 2021
Translating video recordings of mobile app usages into replayable scenariosC Bernal
| print("Hello") |
| print("Hello") |
| base_model: NousResearch/Meta-Llama-3-8B | |
| model_type: LlamaForCausalLM | |
| tokenizer_type: AutoTokenizer | |
| load_in_8bit: true | |
| load_in_4bit: false | |
| strict: false | |
| datasets: | |
| - path: answerdotai/tiny_programs_haiku3_critiques |
| import boto3 | |
| s3 = boto3.resource("s3") | |
| my_bucket = s3.Bucket("s-eai-neox") | |
| file_paths = [] | |
| for my_bucket_object in my_bucket.objects.filter(Prefix="data/codepile/group1/"): | |
| # print(my_bucket_object.key) | |
| file_paths.append(f"s3a://s-eai-neox/{my_bucket_object.key}") | |
| print(len(file_paths)) | |
| from spark_session_builder import build_spark_session | |
| file_paths = file_paths[100:200] |
| import time | |
| import os | |
| from pyspark.ml import Pipeline | |
| from pyspark.ml.feature import RegexTokenizer, NGram, HashingTF, MinHashLSH | |
| from pyspark.sql.functions import col | |
| from spark_session_builder import build_spark_session | |
| spark = build_spark_session("spark://cpu64-dy-c6i-16xlarge-1:7077", 32, 128) | |
| db = spark.read.parquet("/fsx/shared/pilev2_parquet/StackExchange_ver4_non_local_dedupped/dataset.parquet").limit(1_000_000) # Stage 0 & 1 |
