Spaces:
Paused
Paused
feat(seed): Random seed for reproducibility.
Browse files- run-llama.sh +1 -0
- run-sailor.sh +1 -0
run-llama.sh
CHANGED
@@ -21,6 +21,7 @@ printf "Running meta-llama/Llama-3.2-3B-Instruct using vLLM OpenAI compatible AP
|
|
21 |
python -u /app/openai_compatible_api_server.py \
|
22 |
--model meta-llama/Llama-3.2-3B-Instruct \
|
23 |
--revision 0cb88a4f764b7a12671c53f0838cd831a0843b95 \
|
|
|
24 |
--host 0.0.0.0 \
|
25 |
--port 7860 \
|
26 |
--max-num-batched-tokens 32768 \
|
|
|
21 |
python -u /app/openai_compatible_api_server.py \
|
22 |
--model meta-llama/Llama-3.2-3B-Instruct \
|
23 |
--revision 0cb88a4f764b7a12671c53f0838cd831a0843b95 \
|
24 |
+
--seed 42 \
|
25 |
--host 0.0.0.0 \
|
26 |
--port 7860 \
|
27 |
--max-num-batched-tokens 32768 \
|
run-sailor.sh
CHANGED
@@ -23,6 +23,7 @@ printf "Running sail/Sailor-4B-Chat using vLLM OpenAI compatible API Server at p
|
|
23 |
python -u /app/openai_compatible_api_server.py \
|
24 |
--model sail/Sailor-4B-Chat \
|
25 |
--revision 89a866a7041e6ec023dd462adeca8e28dd53c83e \
|
|
|
26 |
--host 0.0.0.0 \
|
27 |
--port 7860 \
|
28 |
--max-num-batched-tokens 32768 \
|
|
|
23 |
python -u /app/openai_compatible_api_server.py \
|
24 |
--model sail/Sailor-4B-Chat \
|
25 |
--revision 89a866a7041e6ec023dd462adeca8e28dd53c83e \
|
26 |
+
--seed 42 \
|
27 |
--host 0.0.0.0 \
|
28 |
--port 7860 \
|
29 |
--max-num-batched-tokens 32768 \
|