Hes

README.md

@@ -27,13 +27,13 @@ The primary intended users of the model are researchers and hobbyists in natural
 
 ## How to Get Started with the Model
 
-- Command line interface: https://github.com/lm-sys/FastChat#vicuna-weights.  
-- APIs (OpenAI API, Huggingface API): https://github.com/lm-sys/FastChat/tree/main#api.  
+Command line interface: https://github.com/lm-sys/FastChat#vicuna-weights.  
+APIs (OpenAI API, Huggingface API): https://github.com/lm-sys/FastChat/tree/main#api.  
 
 ## Training Details
 
 Vicuna v1.3 is fine-tuned from LLaMA with supervised instruction fine-tuning.
-The training data is around 125K conversations collected from ShareGPT.com.
+The training data is around 140K conversations collected from ShareGPT.com.
 See more details in the "Training Details of Vicuna Models" section in the appendix of this [paper](https://arxiv.org/pdf/2306.05685.pdf).
 
 ## Evaluation

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/home/ubuntu/model_weights/vicuna-33b-v1.3",
+  "_name_or_path": "/home/ubuntu/model_weights/vicuna-33b-gpt4-20230521/",
   "architectures": [
     "LlamaForCausalLM"
   ],
@@ -18,6 +18,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "float16",
   "transformers_version": "4.28.1",
-  "use_cache": true,
+  "use_cache": false,
   "vocab_size": 32000
 }