phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-31 23:30:02 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Update Strategy in SamplingParams to be a union

Browse source
This commit is contained in:
Hardik Shah 2025-01-14 15:56:02 -08:00 • committed by Ashwin Bharambe
parent 300e6e2702
commit dea575c994
28 changed files with 600 additions and 377 deletions
Download patch file Download diff file Expand all files Collapse all files

16
docs/source/benchmark_evaluations/index.md
View file

@ -56,9 +56,10 @@ response = client.eval.evaluate_rows(
"type": "model",
"model": "meta-llama/Llama-3.2-90B-Vision-Instruct",
"sampling_params": {
"temperature": 0.0,
"strategy": {
"type": "greedy",
},
"max_tokens": 4096,
"top_p": 0.9,
"repeat_penalty": 1.0,
},
"system_message": system_message
@ -113,9 +114,10 @@ response = client.eval.evaluate_rows(
"type": "model",
"model": "meta-llama/Llama-3.2-90B-Vision-Instruct",
"sampling_params": {
"temperature": 0.0,
"strategy": {
"type": "greedy",
},
"max_tokens": 4096,
"top_p": 0.9,
"repeat_penalty": 1.0,
},
}
@ -134,9 +136,9 @@ agent_config = {
"model": "meta-llama/Llama-3.1-405B-Instruct",
"instructions": "You are a helpful assistant",
"sampling_params": {
"strategy": "greedy",
"temperature": 0.0,
"top_p": 0.95,
"strategy": {
"type": "greedy",
},
},
"tools": [
{

6
docs/source/building_applications/index.md
View file

@ -189,7 +189,11 @@ agent_config = AgentConfig(
# Control the inference loop
max_infer_iters=5,
sampling_params={
"temperature": 0.7,
"strategy": {
"type": "top_p",
"temperature": 0.7,
"top_p": 0.95
},
"max_tokens": 2048
}
)

30
docs/source/references/evals_reference/index.md
View file

@ -92,9 +92,10 @@ response = client.eval.evaluate_rows(
"type": "model",
"model": "meta-llama/Llama-3.2-90B-Vision-Instruct",
"sampling_params": {
"temperature": 0.0,
"strategy": {
"type": "greedy",
},
"max_tokens": 4096,
"top_p": 0.9,
"repeat_penalty": 1.0,
},
"system_message": system_message
@ -149,9 +150,10 @@ response = client.eval.evaluate_rows(
"type": "model",
"model": "meta-llama/Llama-3.2-90B-Vision-Instruct",
"sampling_params": {
"temperature": 0.0,
"strategy": {
"type": "greedy",
},
"max_tokens": 4096,
"top_p": 0.9,
"repeat_penalty": 1.0,
},
}
@ -170,9 +172,9 @@ agent_config = {
"model": "meta-llama/Llama-3.1-405B-Instruct",
"instructions": "You are a helpful assistant",
"sampling_params": {
"strategy": "greedy",
"temperature": 0.0,
"top_p": 0.95,
"strategy": {
"type": "greedy",
},
},
"tools": [
{
@ -318,10 +320,9 @@ The `EvalTaskConfig` are user specified config to define:
"type": "model",
"model": "Llama3.2-3B-Instruct",
"sampling_params": {
"strategy": "greedy",
"temperature": 0,
"top_p": 0.95,
"top_k": 0,
"strategy": {
"type": "greedy",
},
"max_tokens": 0,
"repetition_penalty": 1.0
}
@ -337,10 +338,9 @@ The `EvalTaskConfig` are user specified config to define:
"type": "model",
"model": "Llama3.1-405B-Instruct",
"sampling_params": {
"strategy": "greedy",
"temperature": 0,
"top_p": 0.95,
"top_k": 0,
"strategy": {
"type": "greedy",
},
"max_tokens": 0,
"repetition_penalty": 1.0
}

1
docs/source/references/llama_cli_reference/index.md
View file

@ -214,7 +214,6 @@ llama model describe -m Llama3.2-3B-Instruct
| | } |
+-----------------------------+----------------------------------+
| Recommended sampling params | { |
| | "strategy": "top_p", |
| | "temperature": 1.0, |
| | "top_p": 0.9, |
| | "top_k": 0 |

7
docs/source/references/llama_stack_client_cli_reference.md
View file

@ -200,10 +200,9 @@ Example eval_task_config.json:
"type": "model",
"model": "Llama3.1-405B-Instruct",
"sampling_params": {
"strategy": "greedy",
"temperature": 0,
"top_p": 0.95,
"top_k": 0,
"strategy": {
"type": "greedy"
},
"max_tokens": 0,
"repetition_penalty": 1.0
}