Update README.md

README.md

@@ -5,11 +5,17 @@ base_model:
 tags:
 - mergekit
 - merge
-
+- solar
+license: apache-2.0
 ---
-# solar-hermes
+# Nous-Hermes-2-SOLAR-18B
+
+This is a SOLAR-like model upscaled to 18B. 
+It is a frankenmerge model created using mergekit, alternating layers of Nous-Hermes-2-SOLAR-10.7B and SOLAR-10.7B-Instruct.
+
+Evaluations coming soon! 
 
-This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
+This model has very good writing capabilities (compared to SOLAR-10.7B), specially for role-playing.
 
 ## Merge Details
 ### Merge Method
@@ -54,3 +60,20 @@ merge_method: passthrough
 dtype: float16
 
 ```
+
+
+### Usage
+
+You can use the provided template:
+
+```
+tokenizer = AutoTokenizer.from_pretrained("vicgalle/Nous-Hermes-2-SOLAR-18B")
+model = AutoModelForCausalLM.from_pretrained("vicgalle/Nous-Hermes-2-SOLAR-18B", torch_dtype=torch.float16, load_in_4bit=True)
+
+conversation = [ {'role': 'system', 'content': SYSTEM_PROMPT}, {'role': 'user', 'content': USER_PROMPT} ] 
+prompt = tokenizer.apply_chat_template(conversation, tokenize=False, add_generation_prompt=True)
+inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+
+outputs = model.generate(**inputs, use_cache=True, max_new_tokens=1024, do_sample=True, temperature=0.8)
+output_text = tokenizer.decode(outputs[0]) 
+```