---
license: other
license_name: tongyi-qianwen
license_link: https://huggingface.co/anthracite-org/magnum-v2-72b/blob/main/LICENSE
language:
  - en
  - fr
  - de
  - es
  - it
  - pt
  - ru
  - zh
  - ja
pipeline_tag: text-generation
tags:
- chat
---

## This repo contains EXL2 quants of the model. If you need the original weights, please find them [here](https://huggingface.co/anthracite-org/magnum-v2-72b).
## Base repo only contains the measurement file, see revisions for your quant of choice.

- [measurement.json](https://huggingface.co/anthracite-org/magnum-v2-72b-exl2/tree/main)
- [3.0bpw](https://huggingface.co/anthracite-org/magnum-v2-72b-exl2/tree/3.0bpw)
- [4.0bpw](https://huggingface.co/anthracite-org/magnum-v2-72b-exl2/tree/4.0bpw)
- [6.0bpw](https://huggingface.co/anthracite-org/magnum-v2-72b-exl2/tree/6.0bpw)


![image/png](https://cdn-uploads.huggingface.co/production/uploads/6491e00e057b0928b3e07b75/u8B-5bEeroN549uxUIisV.png)

This is the seventh (Lucky!) in a series of models designed to replicate the prose quality of the Claude 3 models, specifically Sonnet and Opus. This model is fine-tuned on top of [Qwen-2 72B Instruct](https://huggingface.co/Qwen/Qwen2-72B-Instruct).

## Prompting
Model has been Instruct tuned with the ChatML formatting. A typical input would look like this:

```py
"""<|im_start|>user
Hi there!<|im_end|>
<|im_start|>assistant
Nice to meet you!<|im_end|>
<|im_start|>user
Can I ask a question?<|im_end|>
<|im_start|>assistant
"""
```

## Credits
- [anthracite-org/Stheno-Data-Filtered](https://huggingface.co/datasets/anthracite-org/Stheno-Data-Filtered)
- [anthracite-org/kalo-opus-instruct-22k-no-refusal](https://huggingface.co/datasets/anthracite-org/kalo-opus-instruct-22k-no-refusal)
- [anthracite-org/nopm_claude_writing_fixed](https://huggingface.co/datasets/anthracite-org/nopm_claude_writing_fixed)

This model has been a team effort, and the credits goes to all members of Anthracite.

## Training
The training was done for 2 epochs. We used 8x [AMD Instinct™ MI300X Accelerators](https://www.amd.com/en/products/accelerators/instinct/mi300/mi300x.html) for the full-parameter fine-tuning of the model.

We also trained with a weight decay of 0.01 to help further stabilize the loss trajectory and mitigate catastrophic forgetting, and utilize a peak learning rate of 4e-6 to prevent the 2nd epoch loss from dropping too significantly (as it is a strong indicator of overfitting).
![image/png](https://cdn-uploads.huggingface.co/production/uploads/6491e00e057b0928b3e07b75/hVd5gNqSLOlWTkUb0A7iE.png)

Sample Packing was done for 16k tokens rather than the 8k tokens used in our previous runs.

[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)

## Safety
...