Instructions to use PKU-Alignment/ProgressGym-HistLlama3-70B-C018-pretrain-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-C018-pretrain-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-C018-pretrain-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-C018-pretrain-v0.1")
model = AutoModelForCausalLM.from_pretrained("PKU-Alignment/ProgressGym-HistLlama3-70B-C018-pretrain-v0.1", 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 PKU-Alignment/ProgressGym-HistLlama3-70B-C018-pretrain-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-C018-pretrain-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-C018-pretrain-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/PKU-Alignment/ProgressGym-HistLlama3-70B-C018-pretrain-v0.1

SGLang

How to use PKU-Alignment/ProgressGym-HistLlama3-70B-C018-pretrain-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-C018-pretrain-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-C018-pretrain-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-C018-pretrain-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-C018-pretrain-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use PKU-Alignment/ProgressGym-HistLlama3-70B-C018-pretrain-v0.1 with Docker Model Runner:
```
docker model run hf.co/PKU-Alignment/ProgressGym-HistLlama3-70B-C018-pretrain-v0.1
```

ProgressGym-HistLlama3-70B-C018-pretrain-v0.1 / trainer_log.jsonl

TianyiQ

Upload folder using huggingface_hub

be87820 verified over 2 years ago

Raw

History Blame Contribute Delete

5.37 kB

	{"current_steps": 1, "total_steps": 57, "loss": 2.0802, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0, "epoch": 0.008771929824561403, "percentage": 1.75, "elapsed_time": "0:05:52", "remaining_time": "5:28:49"}
	{"current_steps": 3, "total_steps": 57, "loss": 2.0494, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0, "epoch": 0.02631578947368421, "percentage": 5.26, "elapsed_time": "0:16:18", "remaining_time": "4:53:24"}
	{"current_steps": 6, "total_steps": 57, "loss": 2.0771, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.000000000000001e-07, "epoch": 0.05263157894736842, "percentage": 10.53, "elapsed_time": "0:32:12", "remaining_time": "4:33:44"}
	{"current_steps": 9, "total_steps": 57, "loss": 2.0245, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8e-06, "epoch": 0.07894736842105263, "percentage": 15.79, "elapsed_time": "0:47:59", "remaining_time": "4:15:56"}
	{"current_steps": 12, "total_steps": 57, "loss": 2.0568, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9432692307692307e-06, "epoch": 0.10526315789473684, "percentage": 21.05, "elapsed_time": "1:03:29", "remaining_time": "3:58:05"}
	{"current_steps": 15, "total_steps": 57, "loss": 2.0326, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.7730769230769233e-06, "epoch": 0.13157894736842105, "percentage": 26.32, "elapsed_time": "1:18:59", "remaining_time": "3:41:10"}
	{"current_steps": 18, "total_steps": 57, "loss": 2.0396, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6028846153846155e-06, "epoch": 0.15789473684210525, "percentage": 31.58, "elapsed_time": "1:34:38", "remaining_time": "3:25:03"}
	{"current_steps": 21, "total_steps": 57, "loss": 2.0336, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4326923076923077e-06, "epoch": 0.18421052631578946, "percentage": 36.84, "elapsed_time": "1:50:22", "remaining_time": "3:09:12"}
	{"current_steps": 24, "total_steps": 57, "loss": 1.9571, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2625e-06, "epoch": 0.21052631578947367, "percentage": 42.11, "elapsed_time": "2:05:56", "remaining_time": "2:53:10"}
	{"current_steps": 27, "total_steps": 57, "loss": 1.9792, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.092307692307692e-06, "epoch": 0.23684210526315788, "percentage": 47.37, "elapsed_time": "2:21:30", "remaining_time": "2:37:13"}
	{"current_steps": 30, "total_steps": 57, "loss": 2.015, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9221153846153848e-06, "epoch": 0.2631578947368421, "percentage": 52.63, "elapsed_time": "2:37:00", "remaining_time": "2:21:18"}
	{"current_steps": 33, "total_steps": 57, "loss": 1.9845, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7519230769230768e-06, "epoch": 0.2894736842105263, "percentage": 57.89, "elapsed_time": "2:52:28", "remaining_time": "2:05:26"}
	{"current_steps": 36, "total_steps": 57, "loss": 1.974, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.581730769230769e-06, "epoch": 0.3157894736842105, "percentage": 63.16, "elapsed_time": "3:07:55", "remaining_time": "1:49:37"}
	{"current_steps": 39, "total_steps": 57, "loss": 1.9848, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4115384615384616e-06, "epoch": 0.34210526315789475, "percentage": 68.42, "elapsed_time": "3:23:22", "remaining_time": "1:33:51"}
	{"current_steps": 42, "total_steps": 57, "loss": 1.9453, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2413461538461538e-06, "epoch": 0.3684210526315789, "percentage": 73.68, "elapsed_time": "3:38:55", "remaining_time": "1:18:11"}
	{"current_steps": 45, "total_steps": 57, "loss": 1.9987, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.071153846153846e-06, "epoch": 0.39473684210526316, "percentage": 78.95, "elapsed_time": "3:54:16", "remaining_time": "1:02:28"}
	{"current_steps": 48, "total_steps": 57, "loss": 2.0054, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.009615384615385e-07, "epoch": 0.42105263157894735, "percentage": 84.21, "elapsed_time": "4:10:10", "remaining_time": "0:46:54"}
	{"current_steps": 51, "total_steps": 57, "loss": 1.9706, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.307692307692307e-07, "epoch": 0.4473684210526316, "percentage": 89.47, "elapsed_time": "4:25:47", "remaining_time": "0:31:16"}
	{"current_steps": 54, "total_steps": 57, "loss": 1.9874, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.740384615384617e-07, "epoch": 0.47368421052631576, "percentage": 94.74, "elapsed_time": "4:41:08", "remaining_time": "0:15:37"}
	{"current_steps": 57, "total_steps": 57, "loss": 2.0058, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.038461538461539e-07, "epoch": 0.5, "percentage": 100.0, "elapsed_time": "4:56:45", "remaining_time": "0:00:00"}
	{"current_steps": 57, "total_steps": 57, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.5, "percentage": 100.0, "elapsed_time": "4:56:45", "remaining_time": "0:00:00"}