Instructions to use PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1 with Transformers:

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

pipe = pipeline("text-generation", model="PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1")
model = AutoModelForCausalLM.from_pretrained("PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1")
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 PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1

SGLang

How to use PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1 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 "PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1" \
    --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": "PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1",
		"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 "PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1" \
        --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": "PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1 with Docker Model Runner:
```
docker model run hf.co/PKU-Alignment/ProgressGym-HistLlama3-70B-C020-instruct-v0.1
```

ProgressGym-HistLlama3-70B-C020-instruct-v0.1 / trainer_log.jsonl

TianyiQ

Upload folder using huggingface_hub

a051f4e verified about 2 years ago

Raw

History Blame Contribute Delete

3.44 kB

	{"current_steps": 1, "total_steps": 24, "loss": 0.7774, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0, "epoch": 0.041666666666666664, "percentage": 4.17, "elapsed_time": "0:03:08", "remaining_time": "1:12:10"}
	{"current_steps": 2, "total_steps": 24, "loss": 0.7324, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0, "epoch": 0.08333333333333333, "percentage": 8.33, "elapsed_time": "0:05:43", "remaining_time": "1:02:57"}
	{"current_steps": 4, "total_steps": 24, "loss": 0.873, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0, "epoch": 0.16666666666666666, "percentage": 16.67, "elapsed_time": "0:10:57", "remaining_time": "0:54:48"}
	{"current_steps": 6, "total_steps": 24, "loss": 0.8202, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0, "epoch": 0.25, "percentage": 25.0, "elapsed_time": "0:16:21", "remaining_time": "0:49:04"}
	{"current_steps": 8, "total_steps": 24, "loss": 0.8623, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0, "epoch": 0.3333333333333333, "percentage": 33.33, "elapsed_time": "0:21:37", "remaining_time": "0:43:15"}
	{"current_steps": 10, "total_steps": 24, "loss": 0.8277, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3e-06, "epoch": 0.4166666666666667, "percentage": 41.67, "elapsed_time": "0:26:51", "remaining_time": "0:37:36"}
	{"current_steps": 12, "total_steps": 24, "loss": 0.7238, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.7318181818181818e-06, "epoch": 0.5, "percentage": 50.0, "elapsed_time": "0:32:06", "remaining_time": "0:32:06"}
	{"current_steps": 14, "total_steps": 24, "loss": 0.6967, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4636363636363635e-06, "epoch": 0.5833333333333334, "percentage": 58.33, "elapsed_time": "0:37:25", "remaining_time": "0:26:44"}
	{"current_steps": 16, "total_steps": 24, "loss": 0.6387, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1954545454545456e-06, "epoch": 0.6666666666666666, "percentage": 66.67, "elapsed_time": "0:42:42", "remaining_time": "0:21:21"}
	{"current_steps": 18, "total_steps": 24, "loss": 0.5893, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9272727272727273e-06, "epoch": 0.75, "percentage": 75.0, "elapsed_time": "0:47:47", "remaining_time": "0:15:55"}
	{"current_steps": 20, "total_steps": 24, "loss": 0.5832, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.659090909090909e-06, "epoch": 0.8333333333333334, "percentage": 83.33, "elapsed_time": "0:52:57", "remaining_time": "0:10:35"}
	{"current_steps": 22, "total_steps": 24, "loss": 0.6296, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.390909090909091e-06, "epoch": 0.9166666666666666, "percentage": 91.67, "elapsed_time": "0:58:08", "remaining_time": "0:05:17"}
	{"current_steps": 24, "total_steps": 24, "loss": 0.673, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1227272727272728e-06, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "1:03:26", "remaining_time": "0:00:00"}
	{"current_steps": 24, "total_steps": 24, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "1:03:26", "remaining_time": "0:00:00"}