Update README.md

README.md

@@ -7,6 +7,7 @@ datasets:
 language:
 - en
 library_name: transformers
+pipeline_tag: text-generation
 ---
 
 # Snowflake-G1-Tiny
@@ -178,6 +179,77 @@ None. Yet
 - Trained on FlameF0X/Mixture-of-Thoughts-2048T dataset
 - Inspired by GPT architecture with custom optimizations
 
+## Loading and Using the Model
+
+You can load the SnowflakeCore-G1-Tiny model and generate text using the provided script or directly in your Python code.
+
+### Using the Provided Script
+
+Run the following command from the project root:
+
+```bash
+python GPT/generate_text.py
+```
+
+You will be prompted to enter a prompt, and the model will generate text in response.
+
+### Loading the Model in Python
+
+You can also load the model and tokenizer in your own Python scripts as follows:
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
+
+# Load model and tokenizer
+model = AutoModelForCausalLM.from_pretrained(
+    "FlameF0X/SnowflakeCore-G1-Tiny",
+    trust_remote_code=True,
+    force_download=True,
+    use_safetensors=True,
+)
+tokenizer = AutoTokenizer.from_pretrained(
+    "FlameF0X/SnowflakeCore-G1-Tiny",
+    trust_remote_code=True,
+    force_download=True,
+    use_safetensors=True,
+)
+
+def custom_greedy_generate(prompt, max_length=50):
+    model.eval()
+    input_ids = tokenizer(prompt, return_tensors="pt").input_ids
+    generated = input_ids
+    with torch.no_grad():
+        for _ in range(max_length):
+            outputs = model(input_ids=generated)
+            next_token_logits = outputs["logits"][:, -1, :]
+            next_token_id = torch.argmax(next_token_logits, dim=-1).unsqueeze(-1)
+            generated = torch.cat((generated, next_token_id), dim=1)
+            if next_token_id.item() == tokenizer.eos_token_id:
+                break
+    return tokenizer.decode(generated[0], skip_special_tokens=True)
+
+# Example usage
+prompt = "Once upon a time"
+print(custom_greedy_generate(prompt))
+```
+
+## Demo
+
+Example usage with the provided script:
+
+```
+Enter a prompt: Hello, I am Alex and
+
+Generated text: Hello, I am Alex andbourg Chip Chip Chip Chip Chip Chip Chip ChipCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCosCos
+```
+
+**Explanation:**
+- When you run the script, you will be prompted to enter a text prompt (e.g., `Hello, I am Alex and`).
+- The model will then generate a continuation of your prompt, printing the result under "Generated text:".
+- The output shown above is a sample from the model. The actual output may vary depending on the model's training, prompt, and generation settings.
+- This demo demonstrates how the model can extend a given prompt with its learned language patterns. The repetitive or unusual output is typical for small or early-stage models and can be improved with further training or tuning.
+
 ---
 
 *Model trained with PyTorch and mixed precision for optimal performance. For technical support or questions, please open an issue in the repository.*