Update README.md
Browse files
README.md
CHANGED
|
@@ -50,4 +50,99 @@ repetition_penalty: 1.15
|
|
| 50 |
do_sample: true
|
| 51 |
temperature: 0.5
|
| 52 |
top_p: 0.5
|
| 53 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
do_sample: true
|
| 51 |
temperature: 0.5
|
| 52 |
top_p: 0.5
|
| 53 |
+
```
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
# InstructWise 470M - A virtual assistant.
|
| 60 |
+
|
| 61 |
+
Introduction- InstructWise is a model created to act as helpful virtual assistant while maintaing the memory efficiency, this model was fine-tuned on
|
| 62 |
+
## Features
|
| 63 |
+
|
| 64 |
+
- **Creative Content Generation:** NeXGen excels at generating creative writing, including stories, poetry, and fictional narratives.
|
| 65 |
+
- **Contextual Awareness:** The model understands context, ensuring coherent and contextually appropriate responses.
|
| 66 |
+
- **User-Friendly Interface:** NeXGen offers an intuitive and user-friendly interface for seamless integration into various applications.
|
| 67 |
+
- **Versatility:** From content creation to educational support, NeXGen adapts to different writing styles and applications.
|
| 68 |
+
- **Advanced Architecture:** Built on the latest advancements in natural language processing, NeXGen offers high-quality text generation.
|
| 69 |
+
|
| 70 |
+
## Uses
|
| 71 |
+
|
| 72 |
+
NeXGen finds application in various domains, including:
|
| 73 |
+
|
| 74 |
+
- **Content Creation:** Generate marketing copy, stories, and product descriptions.
|
| 75 |
+
- **Assistance in Writing:** Aid authors, bloggers, and students in drafting articles and essays.
|
| 76 |
+
- **Chatbot Development:** Power conversational agents with human-like responses.
|
| 77 |
+
- **Prototyping and Idea Generation:** Facilitate brainstorming sessions for product development.
|
| 78 |
+
- **Social Media Content:** Generate engaging captions for social media posts.
|
| 79 |
+
- **Personal Assistant Applications:** Assist users in drafting emails and messages.
|
| 80 |
+
|
| 81 |
+
## Direct Use Cases
|
| 82 |
+
|
| 83 |
+
NeXGen can be directly employed for:
|
| 84 |
+
|
| 85 |
+
- **Automated Email Drafting:** Quickly compose emails with NeXGen's assistance.
|
| 86 |
+
- **Blog Post Generation:** Generate sections or entire articles based on a given topic.
|
| 87 |
+
- **Code Commenting:** Improve code documentation with clear and concise comments.
|
| 88 |
+
- **Storyline Creation for Games:** Create dynamic and engaging storylines for video games.
|
| 89 |
+
- **Learning Material Generation:** Develop study guides and educational content.
|
| 90 |
+
- **Personal Journaling Assistance:** Receive prompts and suggestions for journaling.
|
| 91 |
+
|
| 92 |
+
## Getting Started
|
| 93 |
+
|
| 94 |
+
To download NeXGen use this code:
|
| 95 |
+
```python
|
| 96 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
| 97 |
+
|
| 98 |
+
# Specify the model name from Hugging Face Model Hub
|
| 99 |
+
model_name = "CrabfishAI/NeXGen-small"
|
| 100 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
| 101 |
+
model = AutoModelForCausalLM.from_pretrained(model_name)
|
| 102 |
+
|
| 103 |
+
def generate_text(prompt, max_length=100, num_beams=5, no_repeat_ngram_size=2, top_k=50, top_p=0.95, temperature=0.7):
|
| 104 |
+
input_ids = tokenizer.encode(prompt, return_tensors="pt")
|
| 105 |
+
|
| 106 |
+
# Ensure attention_mask is provided
|
| 107 |
+
attention_mask = input_ids.ne(tokenizer.pad_token_id).float()
|
| 108 |
+
|
| 109 |
+
# Generate output text
|
| 110 |
+
output = model.generate(
|
| 111 |
+
input_ids,
|
| 112 |
+
max_length=max_length,
|
| 113 |
+
num_beams=num_beams,
|
| 114 |
+
no_repeat_ngram_size=no_repeat_ngram_size,
|
| 115 |
+
top_k=top_k,
|
| 116 |
+
top_p=top_p,
|
| 117 |
+
temperature=temperature,
|
| 118 |
+
attention_mask=attention_mask # Pass attention_mask to the generation method
|
| 119 |
+
)
|
| 120 |
+
|
| 121 |
+
decoded_output = tokenizer.decode(output[0], skip_special_tokens=True)
|
| 122 |
+
return decoded_output
|
| 123 |
+
|
| 124 |
+
# Example usage:
|
| 125 |
+
prompt = "Your prompt here"
|
| 126 |
+
generated_text = generate_text(prompt, max_length=200)
|
| 127 |
+
|
| 128 |
+
print("Generated Text:")
|
| 129 |
+
print(generated_text)
|
| 130 |
+
```
|
| 131 |
+
|
| 132 |
+
## Limitation
|
| 133 |
+
|
| 134 |
+
1. **Content Quality**: The model's output may vary in quality, and there's a possibility it might generate content that is nonsensical, irrelevant, or grammatically incorrect.
|
| 135 |
+
|
| 136 |
+
2. **Bias and Sensitivity**: The model is trained on diverse data, but it may inadvertently exhibit biases or generate content that is sensitive or inappropriate. Exercise caution and review generated text before use.
|
| 137 |
+
|
| 138 |
+
3. **Inappropriate Language**: The model might generate text that includes offensive language or inappropriate content. Be mindful of this, especially in applications where maintaining a respectful and inclusive tone is essential.
|
| 139 |
+
|
| 140 |
+
4. **Ambiguous Prompts**: The quality of generated text is highly dependent on the prompt provided. Ambiguous or unclear prompts may result in less coherent or relevant outputs.
|
| 141 |
+
|
| 142 |
+
## Disclaimer
|
| 143 |
+
|
| 144 |
+
- **Use with Caution**: This model is a tool that should be used with caution. Always review and validate the generated text before incorporating it into any application or publication.
|
| 145 |
+
|
| 146 |
+
- **Not for Critical Applications**: Avoid using the model for critical applications where accuracy and reliability are paramount. The model is intended for creative and exploratory purposes.
|
| 147 |
+
|
| 148 |
+
- **Ongoing Improvement**: The model may be updated or fine-tuned for better performance. Stay informed about updates and consider using the latest version for improved results.
|