Upload README.md

README.md

@@ -5,7 +5,17 @@ license: apache-2.0
 model_creator: Intel
 model_name: Neural Chat 7B v3-1
 model_type: mistral
-prompt_template: '{prompt}
+prompt_template: '### System:
+
+  {system_message}
+
+
+  ### User:
+
+  {prompt}
+
+
+  ### Assistant:
 
   '
 quantized_by: TheBloke
@@ -69,11 +79,17 @@ Here is an incomplete list of clients and libraries that are known to support GG
 <!-- repositories-available end -->
 
 <!-- prompt-template start -->
-## Prompt template: Unknown
+## Prompt template: Orca-Hashes
 
 ```
+### System:
+{system_message}
+
+### User:
 {prompt}
 
+### Assistant:
+
 ```
 
 <!-- prompt-template end -->
@@ -191,7 +207,7 @@ Windows Command Line users: You can set the environment variable by running `set
 Make sure you are using `llama.cpp` from commit [d0cee0d](https://github.com/ggerganov/llama.cpp/commit/d0cee0d36d5be95a0d9088b674dbb27354107221) or later.
 
 ```shell
-./main -ngl 32 -m neural-chat-7b-v3-1.Q4_K_M.gguf --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "{prompt}"
+./main -ngl 32 -m neural-chat-7b-v3-1.Q4_K_M.gguf --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "### System:\n{system_message}\n\n### User:\n{prompt}\n\n### Assistant:"
 ```
 
 Change `-ngl 32` to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
@@ -285,9 +301,9 @@ And thank you again to a16z for their generous grant.
 # Original model card: Intel's Neural Chat 7B v3-1
 
 
-## Finetuning on [habana](https://habana.ai/) HPU
+## Fine-tuning on [Habana](https://habana.ai/) Gaudi2
 
-This model is a fine-tuned model based on [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the open source dataset [Open-Orca/SlimOrca](https://huggingface.co/datasets/Open-Orca/SlimOrca). Then we align it with DPO algorithm. For more details, you can refer our blog: [NeuralChat: Simplifying Supervised Instruction Fine-Tuning and Reinforcement Aligning](https://medium.com/intel-analytics-software/neuralchat-simplifying-supervised-instruction-fine-tuning-and-reinforcement-aligning-for-chatbots-d034bca44f69) and [The Practice of Supervised Fine-tuning and Direct Preference Optimization on Habana Gaudi2](https://medium.com/@NeuralCompressor/the-practice-of-supervised-finetuning-and-direct-preference-optimization-on-habana-gaudi2-a1197d8a3cd3).
+This model is a fine-tuned model based on [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the open source dataset [Open-Orca/SlimOrca](https://huggingface.co/datasets/Open-Orca/SlimOrca). Then we align it with DPO algorithm. For more details, you can refer our blog: [The Practice of Supervised Fine-tuning and Direct Preference Optimization on Habana Gaudi2](https://medium.com/@NeuralCompressor/the-practice-of-supervised-finetuning-and-direct-preference-optimization-on-habana-gaudi2-a1197d8a3cd3).
 
 ## Model date
 Neural-chat-7b-v3-1 was trained between September and October, 2023.
@@ -317,10 +333,22 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 64
 - total_eval_batch_size: 8
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_ratio: 0.02
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.03
 - num_epochs: 2.0
 
+## Prompt Template
+
+```
+### System:
+{system}
+### User:
+{usr}
+### Assistant:
+
+```
+
+
 ## Inference with transformers
 
 ```shell
@@ -346,6 +374,5 @@ The NeuralChat team with members from Intel/SATG/AIA/AIPT. Core team members: Ka
 ## Useful links
 * Intel Neural Compressor [link](https://github.com/intel/neural-compressor)
 * Intel Extension for Transformers [link](https://github.com/intel/intel-extension-for-transformers)
-* Intel Extension for PyTorch [link](https://github.com/intel/intel-extension-for-pytorch)
 
 <!-- original-model-card end -->