3 (cont). One way to elicit reasoning is via "chain-of-thought (CoT) prompting", which gives examples of intermediate reasoning steps in-context. CoT prompting enables large LMs to do multi-step reasoning tasks, increasing the range of tasks that LMs can do.
PROMPT ENGINEERING
-
Context matters: Distinguishing GPT-4 from GPT-3.5 models
By
–
yes, pour ça que difficile de savoir si c'est gpt4 ou 3.5 sur du texte si court, par contre lorsqu'on oriente son contexte 😉
-
Temperature Parameter in AI Models: Deterministic vs Creative
By
–
Il faut savoir qu'il n'y a pas qu'une seule réponse unique. il y a un paramètre qui n'apparait pas dans chatgpt c'est la température. 0 c'est déterministe, 1 c'est créatif. en gros soit une réponse, soit un random.
-
Build Your Own Knowledge Hub by Curating AI Threads
By
–
Build your own perfect knowledge hub. Curate your threads by rewording your question until you’re happy with the answer. Manage your library of threads by deleting individual threads or all at once. pic.twitter.com/gkMMTo8HxD
— Perplexity (@perplexity_ai) 13 avril 2023Build your own perfect knowledge hub. Curate your threads by rewording your question until you’re happy with the answer. Manage your library of threads by deleting individual threads or all at once.
-
Perplexity Launches Multi-Line Text Input for Enhanced Writing
By
–
Use our new multi-line text input to easily write longer questions, edit your tweets, and get more detailed answers. ✍️ Try using Perplexity to edit text, debug code, and more! pic.twitter.com/i1sqeJWZ2b
— Perplexity (@perplexity_ai) 13 avril 2023Use our new multi-line text input to easily write longer questions, edit your tweets, and get more detailed answers. Try using Perplexity to edit text, debug code, and more!
-
LLM Failure Cases in Agent Loop Planning Tasks
By
–
Can you share some of the failure cases? Why specifically it didn’t work? LLM failed to stay in a loop, or follow a plan?
-
Jasper Template for Google-Compliant Product Reviews
By
–
Voilà un template #jasper pour créer une #reviewproduit conforme à #google J'ai fait sur jasper mais c'est compatible chatgpt https://
community.jasper.ai/recipes/3385ce
48-1686-4725-ae25-7f969a48a54b?fpr=copywriting-ai
… #ai #chatgpt -
Auto-GPT: Autonomous AI Agent Towards AGI
By
–
Auto-GPT: the next stage after #ChatGPT
— Pascal Bornet (@pascal_bornet) 13 avril 2023
🚀 Imagine ChatGPT with self-generated prompts, a memory & action abilities (e.g., file saving, emailing)
You just give it a goal and it achieves it
Auto-GPT is an autonomous agent leveraging GPT. A leap towards #AGI?
Crt: agentgpt pic.twitter.com/BNdx6j6pkDAuto-GPT: the next stage after #ChatGPT Imagine ChatGPT with self-generated prompts, a memory & action abilities (e.g., file saving, emailing) You just give it a goal and it achieves it Auto-GPT is an autonomous agent leveraging GPT. A leap towards #AGI? Crt: agentgpt
-
System Prompt Rules Cannot Fix Fundamental Problems
By
–
in case you think adding more rules in the system prompt will somehow fix this problem
-
Any Rule Can Be Bypassed With New Methods
By
–
sure you could make the prompt even more complex to try to combat this but the point here is that practically any rule you give it will always be able to be bypassed by some new method