license: llama3.1
language:
- en
base_model:
- microsoft/phi-4
pipeline_tag: text-generation
library_name: transformers
tags:
- text-generation-inference
- llama3.1
- phi4
- LlamaForCausalLM
- Corpus
- trl
model-index:
- name: Megatron-Opus-14B-2.0
results:
- task:
type: text-generation
name: Text Generation
dataset:
name: IFEval (0-Shot)
type: wis-k/instruction-following-eval
split: train
args:
num_few_shot: 0
metrics:
- type: inst_level_strict_acc and prompt_level_strict_acc
value: 66.94
name: averaged accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FMegatron-Opus-14B-2.0
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: BBH (3-Shot)
type: SaylorTwift/bbh
split: test
args:
num_few_shot: 3
metrics:
- type: acc_norm
value: 54.7
name: normalized accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FMegatron-Opus-14B-2.0
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MATH Lvl 5 (4-Shot)
type: lighteval/MATH-Hard
split: test
args:
num_few_shot: 4
metrics:
- type: exact_match
value: 27.79
name: exact match
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FMegatron-Opus-14B-2.0
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: GPQA (0-shot)
type: Idavidrein/gpqa
split: train
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 14.54
name: acc_norm
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FMegatron-Opus-14B-2.0
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MuSR (0-shot)
type: TAUR-Lab/MuSR
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 10.52
name: acc_norm
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FMegatron-Opus-14B-2.0
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MMLU-PRO (5-shot)
type: TIGER-Lab/MMLU-Pro
config: main
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 46.34
name: accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=prithivMLmods%2FMegatron-Opus-14B-2.0
name: Open LLM Leaderboard
Megatron-Opus-14B-2.0 [ Exp ]
[Megatron-Opus-14B-2.0 ] Exp finetuned from Microsoft's Phi-4 is a state-of-the-art open model developed with a focus on responsible problem solving and advanced reasoning capabilities. Built upon a diverse blend of synthetic datasets, carefully filtered public domain websites, and high-quality academic books and Q&A datasets, Megatron-Opus-14B-2.0 ensures that small, capable models are trained with datasets of exceptional depth and precision.
Megatron-Opus-14B-2.0 adopts a robust safety post-training approach using open-source and in-house synthetic datasets. This involves a combination of SFT (Supervised Fine-Tuning) and iterative DPO (Direct Preference Optimization) techniques, ensuring helpful and harmless outputs across various safety categories.
Dataset Info
Megatron-Opus-14B-2.0 is fine-tuned on a carefully curated synthetic dataset generated using an advanced pipeline optimized for Chain of Thought (CoT) reasoning and Responsible Problem Breakdown (RPB) methodologies. This ensures that the model excels at:
- Logical reasoning
- Step-by-step problem-solving
- Breaking down complex tasks into manageable parts
The dataset also emphasizes responsible decision-making and fairness in generating solutions.
Run with Transformers
# pip install accelerate
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
tokenizer = AutoTokenizer.from_pretrained("prithivMLmods/Megatron-Opus-14B-2.0")
model = AutoModelForCausalLM.from_pretrained(
"prithivMLmods/Megatron-Opus-14B-2.0",
device_map="auto",
torch_dtype=torch.bfloat16,
)
input_text = "Explain the concept of black holes."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")
outputs = model.generate(**input_ids, max_new_tokens=64)
print(tokenizer.decode(outputs[0]))
For chat-style interactions, use tokenizer.apply_chat_template:
messages = [
{"role": "user", "content": "Explain the concept of black holes."},
]
input_ids = tokenizer.apply_chat_template(messages, return_tensors="pt", return_dict=True).to("cuda")
outputs = model.generate(**input_ids, max_new_tokens=256)
print(tokenizer.decode(outputs[0]))
Intended Use
Megatron-Opus-14B-2.0 is tailored for a wide range of applications, especially those involving advanced reasoning, multilingual capabilities, and responsible problem-solving. Its primary use cases include:
Responsible Problem Solving
- Breaking down complex problems into logical, actionable steps.
- Offering ethical, well-rounded solutions in academic and professional contexts.
Advanced Reasoning Tasks
- Excelling in mathematics, logic, and scientific reasoning.
- Providing detailed explanations and systematic answers.
Content Generation
- Assisting in generating high-quality content for various domains, including creative writing and technical documentation.
- Supporting marketers, writers, and educators with detailed and well-structured outputs.
Educational Support
- Acting as a virtual tutor for students by generating practice questions, answers, and detailed explanations.
- Helping educators design learning material that promotes critical thinking and step-by-step problem-solving.
Customer Support & Dialogue Systems
- Enabling chatbots and virtual assistants to provide accurate, helpful, and responsible responses.
- Enhancing customer service with reasoning-driven automation.
Limitations
Despite its strengths, Megatron-Opus-14B-2.0 has some limitations that users should be aware of:
Bias and Fairness
- While great effort has been made to minimize biases, users should critically assess the model’s output in sensitive scenarios to avoid unintended bias.
Contextual Interpretation
- The model may occasionally misinterpret highly nuanced prompts or ambiguous contexts, leading to suboptimal responses.
Knowledge Cutoff
- Megatron-Opus-14B-2.0’s knowledge is static and based on the data available at the time of training. It does not include real-time updates or information on recent developments.
Safety and Harmlessness
- Despite post-training safety alignment, inappropriate or harmful outputs may still occur. Continuous monitoring and human oversight are advised when using the model in critical contexts.
Computational Requirements
- Deploying Megatron-Opus-14B-2.0 efficiently may require substantial computational resources, particularly for large-scale deployments or real-time applications.
Open LLM Leaderboard Evaluation Results
Detailed results can be found here! Summarized results can be found here!
| Metric | Value (%) |
|---|---|
| Average | 36.80 |
| IFEval (0-Shot) | 66.94 |
| BBH (3-Shot) | 54.70 |
| MATH Lvl 5 (4-Shot) | 27.79 |
| GPQA (0-shot) | 14.54 |
| MuSR (0-shot) | 10.52 |
| MMLU-PRO (5-shot) | 46.34 |