tcsenpai/multi1
1
0
Fork 0
mirror of https://github.com/tcsenpai/multi1.git synced 2025-06-06 02:55:21 +00:00
Code Issues 4 Packages Projects Releases Wiki Activity

fix wording ambiguities

Browse Source
This commit is contained in:
Benjamin Klieger 2024-09-15 19:33:33 -07:00 committed by GitHub
parent 5c2c466a15
commit e80dca6cee
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@ -11,11 +11,11 @@ g1 demonstrates the potential of prompting alone to overcome straightforward LLM
### How it works
g1 on top of Llama3.1-70b creates reasoning chains, in principle a dynamic Chain of Thought, that allows the LLM to "think" and solve some logical problems that usually otherwise stump leading models.
g1 powered by Llama3.1-70b creates reasoning chains, in principle a dynamic Chain of Thought, that allows the LLM to "think" and solve some logical problems that usually otherwise stump leading models.
At each step, the LLM can choose to continue to another reasoning step, or provide a final answer. Each step is titled and visible to the user. The system prompt also includes tips for the LLM. There is a full explanation under Prompt Breakdown, but a few examples are asking the model to “include exploration of alternative answers” and “use at least 3 methods to derive the answer”.
The reasoning ability of the LLM is improved through combining Chain-of-Thought with the requirement to try multiple methods, explore alternative answers, question previous draft solutions, and consider the LLM’s limitations. This alone, without additonal training, is sufficient to achieve ~70% accuracy on the Strawberry problem (n=10, "How many Rs are in strawberry?"). Without prompting, Llama-3.1-70b had 0% accuracy and ChatGPT-4o had 30% accuracy.
The reasoning ability of the LLM is therefore improved through combining Chain-of-Thought with the requirement to try multiple methods, explore alternative answers, question previous draft solutions, and consider the LLM’s limitations. This alone, without any training, is sufficient to achieve ~70% accuracy on the Strawberry problem (n=10, "How many Rs are in strawberry?"). Without prompting, Llama-3.1-70b had 0% accuracy and ChatGPT-4o had 30% accuracy.
### Examples