prithivMLmods/Bellatrix-1.5B-xElite

<pre align="center"> ___ __ __ __ __ ( \( __)( ) ( ) /\ ( )( \( )( \/ ) ) < )) )( )( /()\ )( ) / )( ) ( (__/(__)(__)(__)()()() ()\)(__)(/\_) </pre>

Bellatrix-1.5B-xElite

Bellatrix-1.5B-xElite is based on a reasoning-based model designed for the QWQ synthetic dataset entries. The pipeline's instruction-tuned, text-only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks. These models outperform many of the available open-source options. Bellatrix is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions utilize supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF).

Quickstart with Transformers

Here provides a code snippet with apply_chat_template to show you how to load the tokenizer and model and how to generate contents.

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "prithivMLmods/Bellatrix-1.5B-xElite"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

prompt = "Give me a short introduction to large language model."
messages = [
    {"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."},
    {"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=512
)
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]

Intended Use:

1. 1.Multilingual Dialogue Systems:
2. 2.Designed for conversational AI applications, capable of handling dialogue across multiple languages.
3. 3.Useful in customer service, chatbots, and other dialogue-centric use cases.

1. 1.Reasoning and QWQ Dataset Applications:
2. 2.Optimized for tasks requiring logical reasoning and contextual understanding, particularly in synthetic datasets like QWQ.

1. 1.Agentic Retrieval:
2. 2.Supports retrieval-augmented generation tasks, helping systems fetch and synthesize information effectively.

1. 1.Summarization Tasks:
2. 2.Excels in summarizing long or complex text while maintaining coherence and relevance.

1. 1.Instruction-Following Tasks:
2. 2.Can execute tasks based on specific user instructions due to instruction-tuning during training.

1. 1.Language Generation:
2. 2.Suitable for generating coherent and contextually relevant text in various domains and styles.

Limitations:

1. 1.Synthetic Dataset Bias:
2. 2.Optimization for QWQ and similar datasets may make the model less effective on real-world or less structured data.

1. 1.Data Dependency:
2. 2.Performance may degrade on tasks or languages not well-represented in the training dataset.

1. 1.Computational Requirements:
2. 2.The optimized transformer architecture may demand significant computational resources, especially for fine-tuning or large-scale deployments.

1. 1.Potential Hallucinations:
2. 2.Like most auto-regressive models, it may generate plausible-sounding but factually incorrect or nonsensical outputs.

1. 1.RLHF-Specific Biases:
2. 2.Reinforcement Learning with Human Feedback (RLHF) can introduce biases based on the preferences of the annotators involved in the feedback process.

1. 1.Limited Domain Adaptability:
2. 2.While effective in reasoning and dialogue tasks, it may struggle with highly specialized domains or out-of-distribution tasks.

1. 1.Multilingual Limitations:
2. 2.Although optimized for multilingual use, certain low-resource languages may exhibit poorer performance compared to high-resource ones.

1. 1.Ethical Concerns:
2. 2.May inadvertently generate inappropriate or harmful content if safeguards are not applied, particularly in sensitive applications.

1. 1.Real-Time Usability:
2. 2.Latency in inference time could limit its effectiveness in real-time applications or when scaling to large user bases.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here! Summarized results can be found here!