My couch. MacBook Pro. Runpod for compute. Axolotl for training. OpenAI Playground and Anthropic Console for prompt engineering. Sometimes PromptKnit.
@mattshumer_
-
Complete AI Development Stack: Runpod, Axolotl, and Prompt Engineering
By
–
My couch. MacBook Pro. Runpod for compute. Axolotl for training. OpenAI Playground and Anthropic Console for prompt engineering. Sometimes PromptKnit.
-
AI Models Learn Faster With Less Training Data
By
–
Doubt it. As models get smarter, less data will be needed to teach them new things. So if today, it takes millions of examples to teach it a new language, a couple years from now, it may only need a few thousand.
-
Adding Example Inputs Outputs Improves Prompt Performance
By
–
Try adding example inputs/outputs to the prompt!
-
LLM Training Data Dependencies and Implications
By
–
Depends on the LLM and its training data. Will have to think more about the second question.
-
Fine-tune Mistral-7B with synthetic prompt-response datasets
By
–
Switch all prompts to mistral-large. Run the repo many times, collect the prompts / responses for each input / output. Format as JSONL, train a Mistral-7B or similar on all of them, and slot it in.
-
API Optimization and qLoRA Fine-tuning Techniques
By
–
On APIs, it’s hard but if you’re going after a specific use-case it just takes a bit of tinkering with the prompt. If not, just do a qLoRA with a couple hundred negative examples and you’re all set.
-
Prompt First: When to Fine-tune AI Models
By
–
Always, always try prompting first. The only time you should fine-tune is when you NEED to.
-
High-Quality Data Training for AI Models
By
–
Really good data with that style. You don’t need a ton of it (few hundred examples is totally fine), just make sure it’s really high-quality.