Upload README.md with huggingface_hub

README.md

@@ -34,11 +34,42 @@ This model is currently in **Autonomous Growth Mode**. It is training on an RTX
 - **Optimizer:** AdamW with a learning rate of 1e-4.
 - **Sync:** Auto-checkpointing every 2,500 steps to this repository.
 
-## 🛠️ Intended Use
+## 🛠️ Usage (Inference)
 
-DiffReaper-5 is intended for research into **Non-Autoregressive Generation**. Its primary strengths are:
-1. **Speed:** Parallel token generation eliminates the KV-cache bottleneck.
-2. **Coherence:** Focuses on global sequence structure rather than next-token probability.
+Unlike autoregressive models, DiffReaper-5 generates the entire response in parallel through iterative denoising. Use the following logic to run inference:
+
+```python
+import torch
+import torch.nn.functional as F
+
+def generate(model, tokenizer, prompt, steps=10):
+    model.eval()
+    with torch.no_grad():
+        p_tokens = tokenizer(prompt, return_tensors="pt").input_ids.to("cuda")
+        p_emb = model.token_embedding(p_tokens[:, :32]) # Hard conditioning
+        
+        # Start from pure noise
+        r_noise = torch.randn(1, 32, 1024).to("cuda")
+        
+        for i in range(steps):
+            t = torch.tensor([1000 - (i * (1000//steps)) - 1], device="cuda").long()
+            pred = model(torch.cat([p_emb, r_noise], dim=1), t)
+            r_0_pred = pred[:, 32:, :] # Extract response
+            r_noise = 0.4 * r_noise + 0.6 * r_0_pred # Iterative refinement
+            
+        # Map to vocab using Cosine Similarity
+        norm_weights = F.normalize(model.token_embedding.weight, dim=-1)
+        norm_r = F.normalize(r_noise, dim=-1)
+        logits = torch.matmul(norm_r, norm_weights.T)
+        return tokenizer.decode(torch.argmax(logits, dim=-1)[0])
+```
+
+## 🎯 Fine-tuning
+
+To fine-tune DiffReaper-5 on a custom dataset:
+1. **Objective:** Use `1 - F.cosine_similarity` between predicted and target embeddings.
+2. **Conditioning:** Ensure your data loader provides a fixed-length prompt prefix followed by the target response.
+3. **Architecture:** Maintain the 1024-dimensional latent space to stay compatible with the weights.
 
 ## 📈 Diagnostic: Cropmark
 
@@ -47,4 +78,4 @@ The model's progress is monitored via the **Cropmark Diagnostic**.
 - Results are logged in `checkpoint_log.txt` and uploaded periodically.
 
 ---
-*Created by Darwin (Oscar) & Clawd.*
+*Created by Darwin & Clawd.*