Update README.md

README.md

@@ -134,14 +134,16 @@ Engram adds a hashed n-gram lookup at layers 1 and 8. Token sequences are compre
 
 ## Quick start
 
+Install: `pip install torch transformers safetensors`
+
 The model and tokenizer load through `transformers` with remote code. Batched generation with padding is not supported yet, so generate one prompt at a time.
 
 ```python
 from transformers import AutoTokenizer, AutoModelForCausalLM
 
-repo = "RuVI-AI/Pragya-EXP2"
+repo = "RuVI-AI/Pragya-EXP2"   # or a local folder
 tok = AutoTokenizer.from_pretrained(repo, trust_remote_code=True)
-model = AutoModelForCausalLM.from_pretrained(repo, trust_remote_code=True).eval()
+model = AutoModelForCausalLM.from_pretrained(repo, trust_remote_code=True).to("cpu").eval()
 
 inputs = tok("The Internet is", return_tensors="pt")
 out = model.generate(
@@ -151,6 +153,8 @@ out = model.generate(
     do_sample=True,
     temperature=0.7,
     top_k=20,
+    repetition_penalty=1.3,
+    no_repeat_ngram_size=3,
     use_cache=False,
     pad_token_id=tok.pad_token_id,
 )
@@ -159,19 +163,34 @@ print(tok.decode(out[0], skip_special_tokens=True))
 
 The tokenizer prepends `<|bos|>` by default. The string `<|bos|>` inside a prompt is treated as ordinary text.
 
----
+### Expected load output
+
+Loading prints a `LOAD REPORT` listing `model.lm_head.weight` as `MISSING`. This is expected. The checkpoint stores the embedding once, and the head is tied to it when the model loads. Check it with:
+
+```python
+print("tied:", model.model.transformer.wte.weight is model.model.lm_head.weight)
+```
+
+It should print `True`.
+
+### Hub download messages
+
+On first load, Transformers prints "A new version of the following files was downloaded". This is the remote code being fetched, not an error. Pin a `revision=` to avoid the message once you have a version you trust.
 
 ## Sampling
 
-At this size, use tight sampling:
+These settings produced fluent, on-topic output in testing:
 
 ```text
-temperature        0.6 – 0.9
-top_k              20 – 50
-repetition_penalty 1.2 – 1.4
+temperature        0.7
+top_k              20
+repetition_penalty 1.3
+no_repeat_ngram    3
 ```
 
----
+Greedy decoding (`do_sample=False`) repeats phrases after a few sentences. Use sampling.
+
+The model writes in a simple, children's-book register. This reflects the TinyStories-heavy data mix rather than a bug.
 
 ## Capabilities