| 1. | | Flax debugging: making a hash of things (gilesthomas.com) |
| 2 points by gpjt 2 days ago | past | discuss |
|
| 2. | | 10Gb/s Ethernet: switching to a Broadcom SFP+ module (gilesthomas.com) |
| 189 points by gpjt 2 days ago | past | 165 comments |
|
| 3. | | Jax: Commitment Issues (gilesthomas.com) |
| 4 points by gpjt 3 days ago | past | discuss |
|
| 4. | | Jax Back Ends and Devices (gilesthomas.com) |
| 2 points by gpjt 13 days ago | past | discuss |
|
| 5. | | Using Safetensors with Flax (gilesthomas.com) |
| 2 points by gpjt 14 days ago | past |
|
| 6. | | First Looking into Jax (gilesthomas.com) |
| 3 points by gpjt 19 days ago | past |
|
| 7. | | 10Gb/s Ethernet: using mini-heatsinks with a 10GBASE-T SFP+ module (gilesthomas.com) |
| 3 points by gpjt 31 days ago | past |
|
| 8. | | 10Gb/s Ethernet: what I did to get it working in my home (gilesthomas.com) |
| 232 points by gpjt 50 days ago | past | 177 comments |
|
| 9. | | 10Gb Ethernet: what I had to (re)learn (gilesthomas.com) |
| 1 point by gpjt 51 days ago | past | 1 comment |
|
| 10. | | LLM from scratch, part 33 – what I learned from the appendices (gilesthomas.com) |
| 5 points by gpjt 57 days ago | past |
|
| 11. | | LLM from scratch (32l) – Interventions: updated instruction fine-tuning results (gilesthomas.com) |
| 1 point by gpjt 59 days ago | past |
|
| 12. | | How an LLM becomes more coherent as we train it (gilesthomas.com) |
| 3 points by gpjt 62 days ago | past |
|
| 13. | | LLM from scratch, part 32k – Interventions: gradient accumulation (gilesthomas.com) |
| 2 points by gpjt 64 days ago | past |
|
| 14. | | Provision: LLM-powered server setup from Markdown (provision.sh) |
| 2 points by gpjt 69 days ago | past |
|
| 15. | | LLM from scratch, part 32j – trying to train a better model in the cloud (gilesthomas.com) |
| 2 points by gpjt 70 days ago | past |
|
| 16. | | Writing an LLM from scratch, part 32i – Interventions: what is in the noise? (gilesthomas.com) |
| 1 point by gpjt 72 days ago | past |
|
| 17. | | Writing an LLM from scratch, part 32h – Interventions: full fat float32 (gilesthomas.com) |
| 7 points by gpjt 76 days ago | past |
|
| 18. | | Writing an LLM from scratch, part 32g – Interventions: weight tying (gilesthomas.com) |
| 2 points by gpjt 86 days ago | past |
|
| 19. | | Writing an LLM from scratch, part 32f – Interventions: weight decay (gilesthomas.com) |
| 6 points by gpjt 87 days ago | past |
|
| 20. | | Writing an LLM from scratch, part 32e – Interventions: the learning rate (gilesthomas.com) |
| 3 points by gpjt 3 months ago | past |
|
| 21. | | Writing an LLM from scratch, part 32d – Interventions: adding attention bias (gilesthomas.com) |
| 6 points by gpjt 4 months ago | past |
|
| 22. | | Writing an LLM from scratch, part 32c – Interventions: removing dropout (gilesthomas.com) |
| 1 point by gpjt 4 months ago | past |
|
| 23. | | Writing an LLM from scratch, part 32B – Interventions: gradient clipping (gilesthomas.com) |
| 2 points by gpjt 4 months ago | past |
|
| 24. | | Writing an LLM from scratch, part 32a – Interventions: training a baseline model (gilesthomas.com) |
| 1 point by gpjt 4 months ago | past |
|
| 25. | | Getting a Custom PyTorch LLM onto the Hugging Face Hub (gilesthomas.com) |
| 1 point by gpjt 4 months ago | past |
|
| 26. | | Writing an LLM from scratch, part 31 – the models are now on Hugging Face (gilesthomas.com) |
| 2 points by gpjt 5 months ago | past |
|
| 27. | | Writing an LLM from scratch, part 30 – digging into the LLM-as-a-judge results (gilesthomas.com) |
| 1 point by gpjt 5 months ago | past |
|
| 28. | | LLM from scratch, part 29 – using DDP to train a base model in the cloud (gilesthomas.com) |
| 2 points by gpjt 5 months ago | past |
|
| 29. | | LLM from scratch, part 28 – training a base model from scratch on an RTX 3090 (gilesthomas.com) |
| 540 points by gpjt 6 months ago | past | 121 comments |
|
| 30. | | Writing an LLM from scratch, part 27 – what's left, and what's next? (gilesthomas.com) |
| 1 point by gpjt 7 months ago | past |
|
|
| More |
