QuantFactory/AMD-Llama-350M-Upgraded-GGUF

This is quantized version of reflex-ai/AMD-Llama-350M-Upgraded created using llama.cpp

Original Model Card

AMD Llama 350M Upgraded

Model Description

The AMD Llama 350M Upgraded is a transformer-based causal language model built on the Llama architecture, designed to generate human-like text. This model has been upgraded from the original AMD Llama 135M model to provide enhanced performance with an increased parameter count of 332 million. It is suitable for various natural language processing tasks, including text generation, completion, and conversational applications.

Model Details

• Model Type: Causal Language Model
• Architecture: Llama
• Number of Parameters: 332 million
• Input Size: Variable-length input sequences
• Output Size: Variable-length output sequences

Usage

To use the AMD Llama 350M Upgraded model, you can utilize the transformers library. Here’s a sample code snippet to get started:

import torch
from transformers import LlamaForCausalLM, LlamaTokenizer

# Load the tokenizer and model
model_name = "reflex-ai/AMD-Llama-350M-Upgraded"
tokenizer = LlamaTokenizer.from_pretrained(model_name)
model = LlamaForCausalLM.from_pretrained(model_name)

# Set the model to evaluation mode
model.eval()

# Function to generate text
def generate_text(prompt, max_length=50):
    inputs = tokenizer.encode(prompt, return_tensors='pt', padding=True, truncation=True)
    attention_mask = (inputs != tokenizer.pad_token_id).long()

    if torch.cuda.is_available():
        inputs = inputs.to('cuda')
        attention_mask = attention_mask.to('cuda')

    with torch.no_grad():
        outputs = model.generate(inputs, attention_mask=attention_mask, max_length=max_length, num_return_sequences=1)
    
    generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return generated_text

# Example usage
prompt = "Once upon a time in a land far away,"
generated_output = generate_text(prompt, max_length=100)
print(generated_output)