add optional kwargs to readme

README.md

@@ -40,17 +40,35 @@ m.predict('hello world')
 #### Llama::Model.new
 
 ```ruby
-def self.new(
- model, # path to model file, e.g. "models/7B/ggml-model-q4_0.bin"
- n_predict: 128 # number of tokens to predict
- seed: Time.now.to_i, # RNG seed
-)
+require 'llama'
+
+Llama::Model.new('models/7B/ggml-model-q4_0.bin')
+```
+
+Optional arguments:
+
+```ruby
+seed # RNG seed (default Time.now.to_i)
+n_predict # number of tokens to predict (default: 128, -1 = infinity)
+threads # number of threads to use during computation (default: 4)
+top_k # top-k sampling (default: 40)
+top_p # top-p sampling (default: 0.9)
+repeat_last_n # last n tokens to consider for penalize (default: 64)
+repeat_penalty # penalize repeat sequence of tokens (default: 1.1)
+ctx_size # size of the prompt context (default: 512)
+ignore_eos # ignore end of stream token and continue generating
+memory_f32 # use f32 instead of f16 for memory key+value
+temp # temperature (default: 0.8)
+n_parts # number of model parts (default: -1 = determine from dimensions)
+batch_size # batch size for prompt processing (default: 8)
+keep # number of tokens to keep from the initial prompt (default: 0, -1 = all)
+mlock # force system to keep model in RAM rather than swapping or compressing
 ```
 
 #### Llama::Model#predict
 
 ```ruby
-def predict(prompt)
+model.predict('hello world')
 ```
 
 ## Development