Crimson_Dawn-v0.2-exl2

exl2 quant (measurement.json in main branch)

---

check revisions for quants

---

Taking what seemed to work out well with Crimson_Dawn-v0.1, the new Crimson_Dawn-v0.2 is the same training methodology, training on Mistral-Nemo-Base-2407 this time I've added significantly more data, as well as trained using RSLoRA as opposed to regular LoRA. Another key change is training on ChatML as opposed to Mistral Formatting.

Quants!

full / <strong>exl2</strong> / gguf

Prompting

The v0.2 models are trained on ChatML, the prompting structure goes a little something like this:

<|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

Context and Instruct

The v0.2 models are trained on ChatML, please use that Context and Instruct template.

Current Top Sampler Settings

Spicy_Temp <br/> Violet_Twilight-Nitral-Special <br/>

Training

Training was done twice over 2 epochs each on two 2x NVIDIA A6000 GPUs using LoRA. A two-phased approach was used in which the base model was trained 2 epochs on RP data, the LoRA was then applied to base. Finally, the new modified base was trained 2 epochs on instruct, and the new instruct LoRA was applied to the modified base, resulting in what you see here.

<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>