Upload README.md with huggingface_hub

README.md

@@ -113,16 +113,17 @@ After the audit, configurations were updated to apply corrections such as **bf16
 
 The current trainings were interrupted prematurely due to cloud credit exhaustion (AMD MI300X, 192GB VRAM). None of them generated the final JSON. Extracted from the Hugging Face Hub (`Cukinator/cpu1-ablation-checkpoints-v3`), here are the completion percentages of each run before termination:
 
-- **Target steps:** 62,500 steps (50M) / 12,500 steps (10M)
-- `run_01` (50M): 12,173 steps → **19.5%**
-- `run_02` (50M): 10,963 steps → **17.5%**
-- `run_04` (50M): 6,074 steps → **9.7%**
-- `run_05` (50M): 4,390 steps → **7.0%**
-- `run_06` (50M): 4,237 steps → **6.8%**
-- `run_08` (50M): 3,858 steps → **6.2%**
-- `run_13` (10M): 9,290 steps → **74.3%**
-- `run_14` (10M): 9,775 steps → **78.2%**
-- `run_15` (10M): 2,350 steps → **18.8%**
+| Run | Architecture | Params | Target Steps | Last Step | Progress |
+|-----|-------------|-------:|-------------:|----------:|---------:|
+| run_01 | Transformer + BPE (16K vocab) + FP16 | 54.7M | 62,500 | 12,173 | **19.5%** |
+| run_02 | Transformer + Byte + LocalByteDecoder | 38.8M | 62,500 | 10,963 | **17.5%** |
+| run_04 | MLGRU + Byte + Ternary | 38.9M | 62,500 | 6,074 | **9.7%** |
+| run_05 | + FPResidual | 39.0M | 62,500 | 4,390 | **7.0%** |
+| run_06 | + Bolmo patch embedding | 39.0M | 62,500 | 4,237 | **6.8%** |
+| run_08 | Folded Transformer + Byte + Ternary | 38.9M | 62,500 | 3,858 | **6.2%** |
+| run_13 | Small CPU-1 + BPE (4K vocab) | 12.5M | 12,500 | 9,290 | **74.3%** |
+| run_14 | Small CPU-1 byte-level (Qwen distillation) | 10.7M | 12,500 | 9,775 | **78.2%** |
+| run_15 | Small CPU-1 byte + hidden distillation | 10.7M | 12,500 | 2,350 | **18.8%** |
 
 The extremely slow progress (<20% in 1 day for 50M models) was caused by a severe CPU context-switching bottleneck when orchestrating 16 concurrent processes on a single MI300X GPU.