|
|
--- |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- en |
|
|
base_model: |
|
|
- prithivMLmods/Qwen3-1.7B-ft-bf16 |
|
|
pipeline_tag: text-generation |
|
|
library_name: transformers |
|
|
tags: |
|
|
- Non-Reasoning |
|
|
- text-generation-inference |
|
|
datasets: |
|
|
- prithivMLmods/Nemotron-Safety-30K |
|
|
--- |
|
|
|
|
|
 |
|
|
|
|
|
# **Computron-Bots-1.7B-R1** |
|
|
|
|
|
> **Computron-Bots-1.7B-R1** is a **general-purpose safe question-answering model** fine-tuned from **Qwen3-1.7B**, specifically designed for **direct and efficient factual responses** without complex reasoning chains. It provides straightforward, accurate answers across diverse topics, making it ideal for knowledge retrieval, information systems, and applications requiring quick, reliable responses. |
|
|
|
|
|
> \[!note] |
|
|
> GGUF: [https://huggingface.co/prithivMLmods/Computron-Bots-1.7B-R1-GGUF](https://huggingface.co/prithivMLmods/Computron-Bots-1.7B-R1-GGUF) |
|
|
|
|
|
|
|
|
## **Key Features** |
|
|
1. **Direct Question Answering Excellence** |
|
|
Trained to provide clear, concise, and accurate answers to factual questions across a wide range of topics without unnecessary elaboration or complex reasoning steps. |
|
|
|
|
|
2. **General-Purpose Knowledge Base** |
|
|
Capable of handling diverse question types including factual queries, definitions, explanations, and general knowledge questions with consistent reliability. |
|
|
|
|
|
3. **Efficient Non-Reasoning Architecture** |
|
|
Optimized for fast, direct responses without step-by-step reasoning processes, making it perfect for applications requiring immediate answers and high throughput. |
|
|
|
|
|
4. **Compact yet Knowledgeable** |
|
|
Despite its 1.7B parameter size, delivers strong performance for factual accuracy and knowledge retrieval with minimal computational overhead. |
|
|
|
|
|
## **Quickstart with Transformers** |
|
|
```python |
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer |
|
|
|
|
|
model_name = "prithivMLmods/Computron-Bots-1.7B-R1" |
|
|
|
|
|
model = AutoModelForCausalLM.from_pretrained( |
|
|
model_name, |
|
|
torch_dtype="auto", |
|
|
device_map="auto" |
|
|
) |
|
|
tokenizer = AutoTokenizer.from_pretrained(model_name) |
|
|
|
|
|
prompt = "What is the capital of France?" |
|
|
|
|
|
messages = [ |
|
|
{"role": "system", "content": "You are a knowledgeable assistant that provides direct, accurate answers to questions."}, |
|
|
{"role": "user", "content": prompt} |
|
|
] |
|
|
|
|
|
text = tokenizer.apply_chat_template( |
|
|
messages, |
|
|
tokenize=False, |
|
|
add_generation_prompt=True |
|
|
) |
|
|
|
|
|
model_inputs = tokenizer([text], return_tensors="pt").to(model.device) |
|
|
|
|
|
generated_ids = model.generate( |
|
|
**model_inputs, |
|
|
max_new_tokens=256, |
|
|
temperature=0.7, |
|
|
do_sample=True |
|
|
) |
|
|
|
|
|
generated_ids = [ |
|
|
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids) |
|
|
] |
|
|
|
|
|
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0] |
|
|
print(response) |
|
|
``` |
|
|
|
|
|
## **Intended Use** |
|
|
- **Knowledge Base Systems**: Quick factual retrieval for databases and information systems. |
|
|
- **Educational Tools**: Direct answers for students and learners seeking factual information. |
|
|
- **Customer Support Bots**: Efficient responses to common questions and inquiries. |
|
|
- **Search Enhancement**: Improving search results with direct, relevant answers. |
|
|
- **API Integration**: Lightweight question-answering service for applications and websites. |
|
|
- **Research Assistance**: Quick fact-checking and information gathering for researchers. |
|
|
|
|
|
## **Limitations** |
|
|
1. **Non-Reasoning Architecture**: |
|
|
Designed for direct answers rather than complex reasoning, problem-solving, or multi-step analysis tasks. |
|
|
|
|
|
2. **Limited Creative Tasks**: |
|
|
Not optimized for creative writing, storytelling, or tasks requiring imagination and artistic expression. |
|
|
|
|
|
3. **Context Dependency**: |
|
|
May struggle with questions requiring extensive context or nuanced understanding of complex scenarios. |
|
|
|
|
|
4. **Parameter Scale Constraints**: |
|
|
The 1.7B parameter size may limit performance on highly specialized or technical domains compared to larger models. |
|
|
|
|
|
5. **Base Model Limitations**: |
|
|
Inherits any limitations from Qwen3-1.7B's training data and may reflect biases present in the base model. |
|
|
|
|
|
6. **Conversational Depth**: |
|
|
While excellent for Q&A, may not provide the depth of engagement expected in extended conversational scenarios. |
