| 1. | | Show HN: New LLM Pre-Training and Post-Training Paradigms (sebastianraschka.com) |
| 2 points by rasbt on Aug 21, 2024 | past |
|
| 2. | | Developing an LLM: Building, Training, Finetuning (A 1h Video Explainer) (youtube.com) |
| 43 points by rasbt on June 14, 2024 | past | 12 comments |
|
| 3. | | Evaluating LLMs locally, on a laptop, with Llama 3 and Ollama (github.com/rasbt) |
| 2 points by rasbt on June 13, 2024 | past |
|
| 4. | | Understanding the LLM Development Cycle: Building, Training, Finetuning (sebastianraschka.com) |
| 3 points by rasbt on June 8, 2024 | past |
|
| 5. | | The latest major open LLM releases: Mixtral, Llama 3, Phi-3, and OpenELM (sebastianraschka.com) |
| 5 points by rasbt on May 12, 2024 | past |
|
| 6. | | Finetuning an LLM-Based Spam Classifier with LoRA from Scratch (github.com/rasbt) |
| 14 points by rasbt on May 11, 2024 | past |
|
| 7. | | Finetune a GPT Model for Spam Detection on Your Laptop in Just 5 Minutes (github.com/rasbt) |
| 3 points by rasbt on May 3, 2024 | past |
|
| 8. | | Insights from Finetuning LLMs for Classification Tasks (github.com/rasbt) |
| 2 points by rasbt on April 28, 2024 | past |
|
| 9. | | Tips for LLM Pretraining and Evaluating Reward Models (sebastianraschka.com) |
| 2 points by rasbt on March 31, 2024 | past |
|
| 10. | | Comparing 5 ways to implement Multihead Attention in PyTorch (github.com/rasbt) |
| 3 points by rasbt on March 8, 2024 | past |
|
| 11. | | AI Research in Feb 2024 – LoRA Successor, "Small" LLMs, Transparent LLM Research (sebastianraschka.com) |
| 3 points by rasbt on March 3, 2024 | past |
|
| 12. | | Understanding, using, and finetuning Gemma (lightning.ai) |
| 118 points by rasbt on Feb 24, 2024 | past | 48 comments |
|
| 13. | | Implementing Weight-Decomposed Low-Rank Adaptation (DoRA) from Scratch (sebastianraschka.com) |
| 96 points by rasbt on Feb 18, 2024 | past | 10 comments |
|
| 14. | | AI Research Papers in Jan 2024: Model Merging, Mixtures of Experts, Smaller LLMs (sebastianraschka.com) |
| 20 points by rasbt on Feb 3, 2024 | past |
|
| 15. | | Implementing a ChatGPT-like LLM from scratch, step by step (github.com/rasbt) |
| 739 points by rasbt on Jan 27, 2024 | past | 98 comments |
|
| 16. | | LoRA from scratch: implementation for LLM finetuning (lightning.ai) |
| 339 points by rasbt on Jan 22, 2024 | past | 78 comments |
|
| 17. | | Coding Self-Attention, Multi-Head Attention, Cross-Attention, Causal-Attention (sebastianraschka.com) |
| 142 points by rasbt on Jan 14, 2024 | past | 11 comments |
|
| 18. | | Noteworthy AI Research Papers of 2023 (sebastianraschka.com) |
| 3 points by rasbt on Jan 1, 2024 | past |
|
| 19. | | AI Research Papers in November 2023: hallucinations and reasoning capabilities (sebastianraschka.com) |
| 5 points by rasbt on Dec 9, 2023 | past |
|
| 20. | | Practical Tips for Finetuning LLMs Using LoRA (Low-Rank Adaptation) (sebastianraschka.com) |
| 342 points by rasbt on Nov 19, 2023 | past | 27 comments |
|
| 21. | | AI Research Papers (October 2023) (sebastianraschka.com) |
| 5 points by rasbt on Nov 4, 2023 | past |
|
| 22. | | AI and Open Source in 2023: A Review of the Year's Highs and Lows (sebastianraschka.com) |
| 2 points by rasbt on Oct 23, 2023 | past |
|
| 23. | | Takeaways from hundreds of LLM finetuning experiments with LoRA (lightning.ai) |
| 258 points by rasbt on Oct 13, 2023 | past | 39 comments |
|
| 24. | | Atlassian to Acquire Loom in $975M Deal (marketwatch.com) |
| 4 points by rasbt on Oct 12, 2023 | past | 2 comments |
|
| 25. | | OpenAI plans to unveil major developer updates on November 6 like memory storage (reuters.com) |
| 3 points by rasbt on Oct 12, 2023 | past |
|
| 26. | | AMD to acquire AI software startup as it seeks to catch up with Nvidia (cnbc.com) |
| 2 points by rasbt on Oct 11, 2023 | past |
|
| 27. | | AI chips, acquisitions, new "small" open-source LLMs, and new LoRA techniques (sebastianraschka.com) |
| 5 points by rasbt on Oct 9, 2023 | past |
|
| 28. | | AI news editorial from custom AI chips to new "small" LLMs like phi and Mistral (sebastianraschka.com) |
| 1 point by rasbt on Oct 8, 2023 | past |
|
| 29. | | OpenAI is exploring making its own AI chips (reuters.com) |
| 116 points by rasbt on Oct 6, 2023 | past | 94 comments |
|
| 30. | | Meta pitches EU to charge a $10 subscription for ad-free Facebook and Instagram (wsj.com) |
| 41 points by rasbt on Oct 3, 2023 | past | 70 comments |
|
|
| More |
