Fix example parameters
Browse files
README.md
CHANGED
@@ -181,8 +181,10 @@ grammar = LlamaGrammar.from_json_schema(json.dumps({
|
|
181 |
}
|
182 |
}))
|
183 |
|
184 |
-
llm = Llama(model_path="./gorilla-openfunctions-v2.IQ3_M.gguf", n_gpu_layers=33, n_ctx=16384
|
185 |
response = llm.create_chat_completion(
|
|
|
|
|
186 |
messages = [
|
187 |
{
|
188 |
"role": "user",
|
@@ -215,6 +217,8 @@ response = llm.create_chat_completion(
|
|
215 |
print(json.loads(response["choices"][0]["text"]))
|
216 |
|
217 |
print(llm.create_chat_completion(
|
|
|
|
|
218 |
messages = [
|
219 |
{
|
220 |
"role": "user",
|
|
|
181 |
}
|
182 |
}))
|
183 |
|
184 |
+
llm = Llama(model_path="./gorilla-openfunctions-v2.IQ3_M.gguf", n_gpu_layers=33, n_ctx=16384)
|
185 |
response = llm.create_chat_completion(
|
186 |
+
temperature = 0.0,
|
187 |
+
repeat_penalty = 1.1,
|
188 |
messages = [
|
189 |
{
|
190 |
"role": "user",
|
|
|
217 |
print(json.loads(response["choices"][0]["text"]))
|
218 |
|
219 |
print(llm.create_chat_completion(
|
220 |
+
temperature = 0.0,
|
221 |
+
repeat_penalty = 1.1,
|
222 |
messages = [
|
223 |
{
|
224 |
"role": "user",
|