107 lines
4.2 KiB
Markdown
107 lines
4.2 KiB
Markdown
---
|
|
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. |