Update README.md

README.md

@@ -3,20 +3,21 @@ language: en
 license: apache-2.0
 tags:
 - text-generation
-- mpt
+- mistralai
 - moe
 - question-answering
 - natural-questions
-base_model: mosaicml/mpt-7b
+base_model:
+- mistralai/Mixtral-8x7B-v0.1
 ---
 
-# MPT-7B-MoE Fine-tuned on Natural Questions
+# Mixtral-8x7B MoE Fine-tuned on Natural Questions
 
-This model is a fine-tuned version of [mosaicml/mpt-7b](https://huggingface.co/mosaicml/mpt-7b) on the Natural Questions dataset.
+This model is a fine-tuned version of [MistralAI/Mixtral-8x7B-0.1](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1) on the Natural Questions dataset.
 
 ## Model Description
 
-- **Base Model**: mosaicml/mpt-7b
+- **Base Model**: mistralai/Mixtral-8x7B-0.1
 - **Architecture**: Mixture-of-Experts (MoE) variant
 - **Training Dataset**: Natural Questions (NQ) - annotated for expert routing
 - **Task**: Question Answering / Text Generation
@@ -24,7 +25,7 @@ This model is a fine-tuned version of [mosaicml/mpt-7b](https://huggingface.co/m
 ## Training Details
 
 The model was fine-tuned using:
-- DeepSpeed with ZeRO optimization
+- DeepSpeed with ZeRO Stage 2 optimization
 - HuggingFace Accelerate
 - Custom MoE expert annotations
 
@@ -65,7 +66,7 @@ Training was performed using the configuration in the repository. See `train_mix
 
 ## Limitations and Biases
 
-This model inherits limitations and biases from the base MPT-7B model and the Natural Questions dataset.
+This model inherits limitations and biases from the base Mixtral-8x7B model and the Natural Questions dataset.
 Users should be aware of potential biases in question-answering outputs.
 
 ## Citation
@@ -73,9 +74,9 @@ Users should be aware of potential biases in question-answering outputs.
 If you use this model, please cite:
 
 ```bibtex
-@misc{mpt-7b-moe-nq,
-  author = {Your Name},
-  title = {MPT-7B-MoE Fine-tuned on Natural Questions},
+@misc{mixtral-8x7b-moe-nq-finetuned,
+  author = {Vijay Venkatraman},
+  title = {Mixtral-8x7B-0.1 Fine-tuned on Natural Questions},
   year = {2025},
   publisher = {HuggingFace},
   howpublished = {\url{https://huggingface.co/vvijayk/mixtral-8x7b-moe-nq-finetuned}}
@@ -84,6 +85,7 @@ If you use this model, please cite:
 
 ## Acknowledgements
 
-- Base model by MosaicML
+- Base model by Mistral AI
 - Natural Questions dataset by Google Research
 - Training infrastructure using DeepSpeed and HuggingFace Accelerate
+