Evals Template

Overview

This repository contains the run.py script and associated files for conducting evaluations using LLMs from the Anthroppic and OpenAI APIs. It is designed to handle tasks such as generating responses to prompts, caching results, and managing API interactions efficiently.

Setup

Prerequisites

• Python 3.11
• Virtual environment tool (e.g., virtualenv)

Installation

1. Create and Activate a Virtual Environment:

   virtualenv --python python3.11 .venv
   source .venv/bin/activate

2. Install Required Packages:

   pip install -r requirements.txt

3. Install Pre-Commit Hooks:

   make hooks

4. Create a SECRETS file

   touch SECRETS
   echo OPENAI_API_KEY=<INSERT_HERE> >> SECRETS
   echo ANTHROPIC_API_KEY=<INSERT_HERE> >> SECRETS
   echo ACEDEMICNYUPEREZ_ORG=org-<INSERT_HERE> >> SECRETS
   echo FARAI_ORG=org-<INSERT_HERE> >> SECRETS

Usage

Running Inference

• Basic Run on the MMLU Dataset: You must specify an experiment directory to store results. All the logs and hydra config for that experiment will be automatically saved there. Check out teh default config in evals/conf/config.yaml for more options.

  python3 -m evals.run ++exp_dir=exp/test_run1

• Advanced Usage with Overrides: To test a different model, limit to 5 samples, and print output:

  python3 -m evals.run ++exp_dir=exp/test_run2 ++limit=5 ++print_prompt_and_response=true ++language_model.model=gpt-3.5-turbo-instruct ++reset=true

• Creating and using a new prompt: If you want to create a new prompt you can add it to the prompt folder. E.g. creating a prompt woth chain of thought evals/conf/prompt/cot.yaml and then use it instead of zero-shot like this:

  python3 -m evals.run ++exp_dir=exp/test_run3 prompt=cot

  A prompt files contains a messages field that has the prompt in the standard openAI messages format. You can use string templating e.g. $question within the prompt which can be filled in via the code.
• Creating and using a LLM config: You can create new specific LLM param configs so you don't have to override lots of parameters. E.g. creating a config for gpt-4 with temperature 0.8 in evals/conf/language_model/gpt-4-temp-0.8.yaml:

  python3 -m evals.run ++exp_dir=exp/test_run5 ++language_model=gpt-4-temp-0.8

• Control your API usage: Control number of threads with anthropic_num_threads and openai_fraction_rate_limit which you can set via the command line or in the config file.

Running Finetuning

• Basic Run: Prepare a jsonl file according to the openai format and run the following command:

  for n_epochs in 4 8; do python3 -m evals.apis.finetuning.run $jsonl_path --n_epochs $n_epochs --notes test_run --no-ask_to_validate_training --organization FARAI_ORG; done

• Use the CLI: There are a few helper functions to do things like list all the files on the server and delete files if it gets full.

  python3 -m evals.apis.finetuning.cli list_all_files --organization FARAI_ORG
  python3 -m evals.apis.finetuning.cli delete_all_files --organization FARAI_ORG

• Set-up Weights and Biases: You can set up weights and biases to log your finetuning runs. You will need to set up a weights and biases account and then run the following command:

  wandb login

  You can then run the finetuning script with the --use_wandb flag to log your runs. You will need to provide the project name via --project_name too.

Features

• Hydra for Configuration Management: Hydra enables easy overriding of configuration variables. Use ++ for overrides. You can reference other variables within variables using ${var} syntax.

• Caching Mechanism: Caches prompt calls to avoid redundant API calls. Cache location defaults to $exp_dir/cache. This means you can kill your run anytime and restart it without worrying about wasting API calls.

• Prompt History Logging: For debugging, human-readable .txt files are stored in $exp_dir/prompt_history, timestamped for easy reference.

• LLM Inference API Enhancements:

  • Ability to double the rate limit if you pass a list of models e.g. ["gpt-3.5-turbo", "gpt-3.5-turbo-0613"]
  • Manages rate limits efficiently, bypassing the need for exponential backoff.
  • Allows custom filtering of responses via is_valid function.
  • Provides a running total of cost and model timings for performance analysis.
  • Utilise maximum rate limit by setting max_tokens=None for OpenAI models.

• Logging finetuning runs with Weights and Biases:

  • Logs finetuning runs with Weights and Biases for easy tracking of experiments.

• Usage Tracking:

  • Tracks usage of OpenAI and Anthropic APIs so you know how much they are being utilised within your organisation.

Repository Structure

• evals/run.py: Main script for evaluations.
• evals/apis/inference: Directory containing modules for LLM inference
• evals/apis/finetuning: Directory containing scripts to finetune OpenAI models and log with weights and biases
• evals/apis/usage: Directory containing two scripts to get usage information from OpenAI and Anthropic
• evals/conf: Directory containing configuration files for Hydra. Check out prompt and language_model for examples of how to create useful configs.
• evals/data_models: Directory containing Pydantic data models
• evals/load: Directory containing code to download and process MMLU
• tests: Directory containing unit tests
• scripts: Example scripts on how to run sweep experiments

Contributing

Contributions to this repository are welcome. Please follow the standard procedures for submitting issues and pull requests.

---