Upload README.md with huggingface_hub

README.md

@@ -30,29 +30,42 @@ This is the OpenLLM small model trained for 10,000 steps on the SQUAD dataset.
 
 ### Using the Model
 
+This model uses a custom configuration format and requires the OpenLLM framework to load properly.
+
 ```python
-from transformers import AutoTokenizer, AutoModelForCausalLM
+# Load using the OpenLLM framework
+from core.src.model import GPTModel
+import json
 import torch
 
-# Load model and tokenizer
-model_name = "lemms/openllm-small-extended-10k"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name)
+# Load configuration
+with open("config.json", "r") as f:
+    config = json.load(f)
+
+# Create model instance
+model = GPTModel(config["model_config"])
+
+# Load trained weights
+model.load_state_dict(torch.load("pytorch_model.bin", map_location="cpu"))
+
+# Load tokenizer
+import sentencepiece as spm
+tokenizer = spm.SentencePieceProcessor()
+tokenizer.load("tokenizer.model")
 
 # Generate text
 prompt = "The future of artificial intelligence"
-inputs = tokenizer(prompt, return_tensors="pt")
+tokens = tokenizer.encode(prompt)
+inputs = torch.tensor([tokens], dtype=torch.long)
 
 with torch.no_grad():
     outputs = model.generate(
-        inputs["input_ids"],
+        inputs,
         max_length=100,
-        temperature=0.7,
-        do_sample=True,
-        pad_token_id=tokenizer.eos_token_id
+        temperature=0.7
     )
 
-generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
+generated_text = tokenizer.decode(outputs[0].tolist())
 print(generated_text)
 ```
 
@@ -66,16 +79,17 @@ model, tokenizer = load_model_and_tokenizer("lemms/openllm-small-extended-10k")
 
 # Generate text
 prompt = "The history of machine learning"
-inputs = tokenizer(prompt, return_tensors="pt")
+tokens = tokenizer.encode(prompt)
+inputs = torch.tensor([tokens], dtype=torch.long)
 
 with torch.no_grad():
     outputs = model.generate(
-        inputs["input_ids"],
+        inputs,
         max_length=100,
         temperature=0.7
     )
 
-print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+print(tokenizer.decode(outputs[0].tolist()))
 ```
 
 ## Model Architecture