Instructions to use JZSAWYER/rcgrpo_llama_wmft_long with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use JZSAWYER/rcgrpo_llama_wmft_long with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="JZSAWYER/rcgrpo_llama_wmft_long")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("JZSAWYER/rcgrpo_llama_wmft_long")
model = AutoModelForCausalLM.from_pretrained("JZSAWYER/rcgrpo_llama_wmft_long", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use JZSAWYER/rcgrpo_llama_wmft_long with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "JZSAWYER/rcgrpo_llama_wmft_long"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "JZSAWYER/rcgrpo_llama_wmft_long",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/JZSAWYER/rcgrpo_llama_wmft_long

SGLang

How to use JZSAWYER/rcgrpo_llama_wmft_long with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "JZSAWYER/rcgrpo_llama_wmft_long" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "JZSAWYER/rcgrpo_llama_wmft_long",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "JZSAWYER/rcgrpo_llama_wmft_long" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "JZSAWYER/rcgrpo_llama_wmft_long",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use JZSAWYER/rcgrpo_llama_wmft_long with Docker Model Runner:
```
docker model run hf.co/JZSAWYER/rcgrpo_llama_wmft_long
```

wm-aligned-long

This model is a fine-tuned version of /inspire/hdd/project/chemicalreaction/dijixiu-CZXS25220051/models/Llama-3.1-8B-Instruct on the bfcl_wm_aligned_train dataset. It achieves the following results on the evaluation set:

Loss: 0.0777

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-05
train_batch_size: 1
eval_batch_size: 2
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 2
total_train_batch_size: 16
total_eval_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 5.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.1963	0.2778	25	0.1500
0.1651	0.5556	50	0.1128
0.0959	0.8333	75	0.0995
0.07	1.1111	100	0.0939
0.0747	1.3889	125	0.0807
0.0444	1.6667	150	0.0837
0.0372	1.9444	175	0.0822
0.0321	2.2222	200	0.0777
0.033	2.5	225	0.0751
0.0346	2.7778	250	0.0729
0.0151	3.0556	275	0.0740
0.02	3.3333	300	0.0752
0.0188	3.6111	325	0.0745
0.0178	3.8889	350	0.0735
0.0136	4.1667	375	0.0753
0.0108	4.4444	400	0.0779
0.0115	4.7222	425	0.0782
0.0106	5.0	450	0.0782

Framework versions

Transformers 4.57.1
Pytorch 2.9.1+cu128
Datasets 4.0.0
Tokenizers 0.22.1

Downloads last month: 28

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for JZSAWYER/rcgrpo_llama_wmft_long

Base model

meta-llama/Llama-3.1-8B

Finetuned

meta-llama/Llama-3.1-8B-Instruct

Finetuned

(2916)

this model