Update README.md

README.md

@@ -40,7 +40,7 @@ LLM, please do feel free to play with it!
 ### Model Sources
 
 - **Repository:** [gpjt/ddp-base-model-from-scratch](https://github.com/gpjt/ddp-base-model-from-scratch)
-- **Blog post:** [Writing an LLM from scratch, part 29 -- using DistributedDataParallel to train a base model from scratch in the cloud](https://www.gilesthomas.com/2026/01/llm-from-scratch-29-ddp-training-a-base-model-in-the-cloud)
+- **Blog post:** [Writing an LLM from scratch, part 32b -- Interventions: gradient clipping](https://www.gilesthomas.com/2026/02/llm-from-scratch-32b-interventions-gradient-clipping)
 
 ## How to Get Started with the Model
 
@@ -78,9 +78,10 @@ number of tokens.  It's [both dumb and ignorant](https://www.gilesthomas.com/202
 
 ## Training Details
 
-- **Machine type:** TODO
+- **Machine type:** 8a A100 with 40GiB per GPU
 - **Tokens:**  3,260,190,720 (Chinchilla-optimal of 20x parameters) rounded up to the nearest batch.
 - **Dataset:** [gpjt/fineweb-gpt2-tokens](https://huggingface.co/datasets/gpjt/fineweb-gpt2-tokens)
 - **Micro-batch size:** 12
-- **Global batch size:** TODO
+- **Global batch size:** 96
 - **Dropout:** 0.1
+- **Gradient clipping**: at 3.5