Update README.md

README.md

@@ -23,44 +23,20 @@ datasets:
 library_name: transformers
 ---
 
+Quantized model => https://huggingface.co/anthracite-org/magnum-v2-123b
 
+**Quantization Details:**  
+Quantization is done using turboderp's ExLlamaV2 v0.2.2. 
 
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/6491e00e057b0928b3e07b75/hkPzhL-xYPeGGKCyAf3Qd.png)
-This is the sixth 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 [Mistral-Large-Instruct-2407](https://huggingface.co/mistralai/Mistral-Large-Instruct-2407).
+I use the default calibration datasets and arguments. The repo also includes a "measurement.json" file, which was used during the quantization process.
 
-## Prompting
-Model has been Instruct tuned with the Mistral formatting. A typical input would look like this:
+For models with bits per weight (BPW) over 6.0, I default to quantizing the `lm_head` layer at 8 bits instead of the standard 6 bits.
 
-```py
-<s>[INST] SYSTEM MESSAGE\nUSER MESSAGE[/INST] ASSISTANT MESSAGE</s>[INST] USER MESSAGE[/INST]
-```
 
-We also provide SillyTavern presets for [Context](https://huggingface.co/anthracite-org/Magnum-123b-v1/resolve/main/Magnum-Mistral-Context.json) and [Instruct](https://huggingface.co/anthracite-org/Magnum-123b-v1/raw/main/Magnum-Mistral-Instruct.json) respectively.
 
-The Mistral preset included in SillyTavern seems to be misconfigured by default, so we recommend using these as a replacement.
-
-## 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 1.5 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.
-
-In addition to this, we noticed that Mistral Large models seemed much more sensitive to learning rate adjustments than other models:
-
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/6491e00e057b0928b3e07b75/xCK3ISKF6pWcMyO7MEzTA.png)
-
-We hypothesize this is primarily due to the particularly narrow and low variance weight distributions typical of Mistral derived models regardless of their scale.
-
-In the end, due to the costs that would be involved in training another full 2 epochs run ($600) on an even lower rate, we settled on our third attempt: 2e-6 with an effective batch size of 64. We chose to publish the 1.5 epoch run after manually testing and comparing it.
-
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/6491e00e057b0928b3e07b75/d9_cBy-DuWrdnoVBbAvRV.png)
-Also, we notice a correlation between the significance of the 2nd epoch loss drop and the strength of the learning rate, implying 4e-6 leads to more catastrophic forgetting.
+---
 
-[<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)
+**Who are you? What's with these weird BPWs on [insert model here]?**  
+I specialize in optimized EXL2 quantization for models in the 70B to 100B+ range, specifically tailored for 48GB VRAM setups. My rig is built using 2 x 3090s with a Ryzen APU (APU used solely for desktop output—no VRAM wasted on the 3090s). I use TabbyAPI for inference, targeting context sizes between 32K and 64K.
 
-## Safety
-...
+Every model I upload includes a `config.yml` file with my ideal TabbyAPI settings. If you're using my config, don’t forget to set `PYTORCH_CUDA_ALLOC_CONF=backend:cudaMallocAsync` to save some VRAM.