Instructions to use ifable/gemma-2-Ifable-9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ifable/gemma-2-Ifable-9B with Transformers:

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

pipe = pipeline("text-generation", model="ifable/gemma-2-Ifable-9B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ifable/gemma-2-Ifable-9B")
model = AutoModelForCausalLM.from_pretrained("ifable/gemma-2-Ifable-9B", 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 ifable/gemma-2-Ifable-9B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ifable/gemma-2-Ifable-9B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ifable/gemma-2-Ifable-9B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ifable/gemma-2-Ifable-9B

SGLang

How to use ifable/gemma-2-Ifable-9B 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 "ifable/gemma-2-Ifable-9B" \
    --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": "ifable/gemma-2-Ifable-9B",
		"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 "ifable/gemma-2-Ifable-9B" \
        --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": "ifable/gemma-2-Ifable-9B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ifable/gemma-2-Ifable-9B with Docker Model Runner:
```
docker model run hf.co/ifable/gemma-2-Ifable-9B
```

gemma-2-Ifable-9B / all_results.json

linzaiyun

Upload folder using huggingface_hub

eea2579 verified about 2 years ago

Raw History Blame Contribute Delete

1.34 kB

	{
	"before_init_mem_cpu": 3802071040,
	"before_init_mem_gpu": 22016,
	"epoch": 0.9807355516637478,
	"eval_logits/chosen": -12.004097938537598,
	"eval_logits/rejected": -17.047502517700195,
	"eval_logps/chosen": -2.168222427368164,
	"eval_logps/rejected": -4.787535667419434,
	"eval_loss": 1.0162526369094849,
	"eval_mem_cpu_alloc_delta": 466944,
	"eval_mem_cpu_peaked_delta": 0,
	"eval_mem_gpu_alloc_delta": 0,
	"eval_mem_gpu_peaked_delta": 25220711424,
	"eval_rewards/accuracies": 0.9166666865348816,
	"eval_rewards/chosen": -21.682226181030273,
	"eval_rewards/margins": 26.193130493164062,
	"eval_rewards/rejected": -47.875362396240234,
	"eval_runtime": 9.9413,
	"eval_samples_per_second": 9.456,
	"eval_sft_loss": 0.01844729855656624,
	"eval_steps_per_second": 1.207,
	"init_mem_cpu_alloc_delta": 364544,
	"init_mem_cpu_peaked_delta": 0,
	"init_mem_gpu_alloc_delta": 0,
	"init_mem_gpu_peaked_delta": 0,
	"total_flos": 39867492466688.0,
	"train_loss": 3.085822834287371,
	"train_mem_cpu_alloc_delta": 5213659136,
	"train_mem_cpu_peaked_delta": 22737326080,
	"train_mem_gpu_alloc_delta": 16267848704,
	"train_mem_gpu_peaked_delta": 36029468160,
	"train_runtime": 1628.7465,
	"train_samples_per_second": 2.805,
	"train_steps_per_second": 0.021
	}