|
--- |
|
library_name: transformers |
|
tags: |
|
- code |
|
- chemistry |
|
- medical |
|
license: apache-2.0 |
|
datasets: |
|
- Locutusque/hyperion-v2.0 |
|
language: |
|
- en |
|
--- |
|
# Hyperion-2.0-Mistral-7B |
|
|
|
|
|
 |
|
|
|
## Model Details |
|
- **Model Name**: Locutusque/Hyperion-2.0-Mistral-7B |
|
- **Base Model**: mistralai/Mistral-7B-v0.1 |
|
- **Publisher**: Locutusque |
|
- **Model Type**: Question answering, conversational AI, code generation, medical text comprehension, mathematical reasoning, logical reasoning. |
|
- **Language**: Multi-domain, English language. |
|
- **License**: Apache-2.0 |
|
|
|
## Model Description |
|
`Locutusque/Hyperion-2.0-Mistral-7B` is a state-of-the-art language model fine-tuned on the Hyperion-v2.0 dataset for advanced reasoning across scientific domains. This model is designed to handle complex inquiries and instructions, leveraging the diverse and rich information contained in the Hyperion dataset. Its primary use cases include but are not limited to complex question answering, conversational understanding, code generation, medical text comprehension, mathematical reasoning, and logical reasoning. |
|
|
|
## Intended Use |
|
This model is intended for researchers and practitioners looking for a powerful tool to tackle challenging problems in scientific domains. It can be used in the following scenarios: |
|
- AI-driven tutoring systems for science, medicine, mathematics, and computer science. |
|
- Assistive tools for professionals requiring fast and accurate domain-specific information retrieval. |
|
- Platforms that require conversational AI capabilities with a focus on technical and scientific reasoning. |
|
- Automation in code generation and understanding complex programming context. |
|
|
|
## Training Data |
|
The `Locutusque/Hyperion-2.0-Mistral-7B` model was fine-tuned on the Hyperion-v2.0 dataset, which amalgamates various datasets rich in diversity and complexity, including programming, medical texts, mathematical problems, and reasoning tasks. |
|
|
|
## Quants |
|
|
|
ExLlamaV2: https://huggingface.co/bartowski/Hyperion-2.0-Mistral-7B-exl2 |
|
|
|
GGUF: https://huggingface.co/bartowski/Hyperion-2.0-Mistral-7B-GGUF |
|
|
|
## Evaluation Results |
|
0-shot AGIEval |
|
| Tasks |Version|Filter|n-shot| Metric |Value | |Stderr| |
|
|---------------------------------|-------|------|-----:|--------|-----:|---|-----:| |
|
|agieval_nous |N/A |none | 0|acc |0.3602|± |0.0929| |
|
| | |none | 0|acc_norm|0.3342|± |0.0764| |
|
| - agieval_aqua_rat | 1|none | 0|acc |0.2402|± |0.0269| |
|
| | |none | 0|acc_norm|0.2441|± |0.0270| |
|
| - agieval_logiqa_en | 1|none | 0|acc |0.2965|± |0.0179| |
|
| | |none | 0|acc_norm|0.3226|± |0.0183| |
|
| - agieval_lsat_ar | 1|none | 0|acc |0.2348|± |0.0280| |
|
| | |none | 0|acc_norm|0.2000|± |0.0264| |
|
| - agieval_lsat_lr | 1|none | 0|acc |0.3667|± |0.0214| |
|
| | |none | 0|acc_norm|0.3373|± |0.0210| |
|
| - agieval_lsat_rc | 1|none | 0|acc |0.4981|± |0.0305| |
|
| | |none | 0|acc_norm|0.4089|± |0.0300| |
|
| - agieval_sat_en | 1|none | 0|acc |0.6359|± |0.0336| |
|
| | |none | 0|acc_norm|0.5777|± |0.0345| |
|
| - agieval_sat_en_without_passage| 1|none | 0|acc |0.3883|± |0.0340| |
|
| | |none | 0|acc_norm|0.3544|± |0.0334| |
|
| - agieval_sat_math | 1|none | 0|acc |0.3500|± |0.0322| |
|
| | |none | 0|acc_norm|0.2682|± |0.0299| |
|
|
|
| Groups |Version|Filter|n-shot| Metric |Value | |Stderr| |
|
|------------|-------|------|-----:|--------|-----:|---|-----:| |
|
|agieval_nous|N/A |none | 0|acc |0.3602|± |0.0929| |
|
| | |none | 0|acc_norm|0.3342|± |0.0764| |
|
|
|
5-shot AGIEval coming soon. |
|
|
|
## How to Use |
|
```python |
|
from transformers import AutoModelForCausalLM, AutoTokenizer |
|
|
|
model_name = "Locutusque/Hyperion-1.5-Mistral-7B" |
|
tokenizer = AutoTokenizer.from_pretrained(model_name) |
|
model = AutoModelForCausalLM.from_pretrained(model_name) |
|
|
|
# For a text generation task |
|
input_text = "<|im_start|>user\nWhat are the implications of Einstein's theory of relativity in modern physics?<|im_end|>\n<|im_start|>assistant\n" |
|
input_ids = tokenizer.encode(input_text, return_tensors="pt") |
|
|
|
# Generate a response |
|
outputs = model.generate(input_ids, max_length=200, num_return_sequences=1, temperature=0.8, top_p=0.95, top_k=40, repetition_penalty=1.1) |
|
print(tokenizer.decode(outputs[0], skip_special_tokens=True)) |
|
``` |
|
## Known Limitations |
|
|
|
The diversity of the dataset could lead to inconsistencies in the model's responses due to variations in data formatting and annotation quality. |
|
|
|
This model is also very compliant, it will respond to any request. Please make sure to build upon this model with DPO if you plan on using it for enterprise-level deployment. |
|
|
|
## Licensing Information |
|
|
|
This model is released under the Apache-2.0 license. |
