Last active
November 26, 2023 16:40
-
-
Save konijn/da37cf189c776997657c8da94c8a6733 to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| 25 Nov 2023 | |
| ----------- | |
| Realized I lost track of my last model?? | |
| Realized I need to take notes ;) | |
| Downloaded and installed https://github.com/RahulSChand/gpu_poor | |
| Name RTX-2060 | |
| bandwidth 336 GB | |
| compute 13 TFlops/s | |
| memory 6 GB | |
| Open LLM leaderboard | |
| https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard | |
| Ollama, easy quick installer for models | |
| https://ollama.ai | |
| 26 Nov 2023 | |
| ----------- | |
| https://github.com/fastai/fastbook | |
| Todo | |
| ---- | |
| Check out https://ollama.ai/fakezeta/neural-chat-7b-v3-1 | |
| Colab | |
| ----- | |
| data_table.enable_dataframe_formatter() | |
| Notes to self | |
| ------------- | |
| DROP -> Deep reading over paraphrases | |
| GSM8K -> Grade School Math 8K | |
| Winogrand -> a large-scale dataset of pronoun resolution problems | |
| TruthfulQA -> a benchmark dataset that measures the truthfulness of language models | |
| MMLU -? 10,000 multiple-choice questions that test the ability to understand the meaning of words & sentences in context. | |
| HellaSwag -> This tests common-sense reasoning capabilities. 2500 multiple-choice questions that require inferences about the world based on common-sense knowledge. | |
| ARC -> Evaluates question-answering abilities. 7,787 multiple-choice questions that test questions based on a given context |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment