Instructions to use mlfoundations-dev/ds_no_offload_liger_packing_zero2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mlfoundations-dev/ds_no_offload_liger_packing_zero2 with Transformers:

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

pipe = pipeline("text-generation", model="mlfoundations-dev/ds_no_offload_liger_packing_zero2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mlfoundations-dev/ds_no_offload_liger_packing_zero2")
model = AutoModelForCausalLM.from_pretrained("mlfoundations-dev/ds_no_offload_liger_packing_zero2")
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

vLLM

How to use mlfoundations-dev/ds_no_offload_liger_packing_zero2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mlfoundations-dev/ds_no_offload_liger_packing_zero2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlfoundations-dev/ds_no_offload_liger_packing_zero2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mlfoundations-dev/ds_no_offload_liger_packing_zero2

SGLang

How to use mlfoundations-dev/ds_no_offload_liger_packing_zero2 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 "mlfoundations-dev/ds_no_offload_liger_packing_zero2" \
    --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": "mlfoundations-dev/ds_no_offload_liger_packing_zero2",
		"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 "mlfoundations-dev/ds_no_offload_liger_packing_zero2" \
        --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": "mlfoundations-dev/ds_no_offload_liger_packing_zero2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mlfoundations-dev/ds_no_offload_liger_packing_zero2 with Docker Model Runner:
```
docker model run hf.co/mlfoundations-dev/ds_no_offload_liger_packing_zero2
```

ds_no_offload_liger_packing_zero2 / trainer_log.jsonl

sedrickkeh

Training in progress, epoch 3

5891021 verified about 1 year ago

raw

history blame contribute delete

3.42 kB

	{"current_steps": 1, "total_steps": 18, "loss": 1.2885, "lr": 5e-06, "epoch": 0.16666666666666666, "percentage": 5.56, "elapsed_time": "0:01:14", "remaining_time": "0:21:09"}
	{"current_steps": 2, "total_steps": 18, "loss": 1.3273, "lr": 1e-05, "epoch": 0.3333333333333333, "percentage": 11.11, "elapsed_time": "0:02:18", "remaining_time": "0:18:31"}
	{"current_steps": 3, "total_steps": 18, "loss": 1.2085, "lr": 9.903926402016153e-06, "epoch": 0.5, "percentage": 16.67, "elapsed_time": "0:03:23", "remaining_time": "0:16:55"}
	{"current_steps": 4, "total_steps": 18, "loss": 1.1297, "lr": 9.619397662556434e-06, "epoch": 0.6666666666666666, "percentage": 22.22, "elapsed_time": "0:04:27", "remaining_time": "0:15:35"}
	{"current_steps": 5, "total_steps": 18, "loss": 1.1569, "lr": 9.157348061512728e-06, "epoch": 0.8333333333333334, "percentage": 27.78, "elapsed_time": "0:05:31", "remaining_time": "0:14:22"}
	{"current_steps": 6, "total_steps": 18, "loss": 1.0759, "lr": 8.535533905932739e-06, "epoch": 1.0, "percentage": 33.33, "elapsed_time": "0:06:26", "remaining_time": "0:12:52"}
	{"current_steps": 7, "total_steps": 18, "loss": 1.037, "lr": 7.777851165098012e-06, "epoch": 1.1666666666666667, "percentage": 38.89, "elapsed_time": "0:08:45", "remaining_time": "0:13:45"}
	{"current_steps": 8, "total_steps": 18, "loss": 1.0009, "lr": 6.913417161825449e-06, "epoch": 1.3333333333333333, "percentage": 44.44, "elapsed_time": "0:09:49", "remaining_time": "0:12:16"}
	{"current_steps": 9, "total_steps": 18, "loss": 0.9742, "lr": 5.975451610080643e-06, "epoch": 1.5, "percentage": 50.0, "elapsed_time": "0:10:53", "remaining_time": "0:10:53"}
	{"current_steps": 10, "total_steps": 18, "loss": 0.9315, "lr": 5e-06, "epoch": 1.6666666666666665, "percentage": 55.56, "elapsed_time": "0:11:57", "remaining_time": "0:09:33"}
	{"current_steps": 11, "total_steps": 18, "loss": 0.9095, "lr": 4.02454838991936e-06, "epoch": 1.8333333333333335, "percentage": 61.11, "elapsed_time": "0:13:01", "remaining_time": "0:08:17"}
	{"current_steps": 12, "total_steps": 18, "loss": 0.9142, "lr": 3.0865828381745515e-06, "epoch": 2.0, "percentage": 66.67, "elapsed_time": "0:13:55", "remaining_time": "0:06:57"}
	{"current_steps": 13, "total_steps": 18, "loss": 0.8895, "lr": 2.2221488349019903e-06, "epoch": 2.1666666666666665, "percentage": 72.22, "elapsed_time": "0:16:17", "remaining_time": "0:06:15"}
	{"current_steps": 14, "total_steps": 18, "loss": 0.87, "lr": 1.4644660940672628e-06, "epoch": 2.3333333333333335, "percentage": 77.78, "elapsed_time": "0:17:21", "remaining_time": "0:04:57"}
	{"current_steps": 15, "total_steps": 18, "loss": 0.8943, "lr": 8.426519384872733e-07, "epoch": 2.5, "percentage": 83.33, "elapsed_time": "0:18:25", "remaining_time": "0:03:41"}
	{"current_steps": 16, "total_steps": 18, "loss": 0.8858, "lr": 3.8060233744356634e-07, "epoch": 2.6666666666666665, "percentage": 88.89, "elapsed_time": "0:19:29", "remaining_time": "0:02:26"}
	{"current_steps": 17, "total_steps": 18, "loss": 0.8833, "lr": 9.607359798384785e-08, "epoch": 2.8333333333333335, "percentage": 94.44, "elapsed_time": "0:20:33", "remaining_time": "0:01:12"}
	{"current_steps": 18, "total_steps": 18, "loss": 0.8752, "lr": 0.0, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "0:21:28", "remaining_time": "0:00:00"}
	{"current_steps": 18, "total_steps": 18, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "0:23:48", "remaining_time": "0:00:00"}