Instructions to use lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample with Transformers:

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

pipe = pipeline("text-generation", model="lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample")
model = AutoModelForCausalLM.from_pretrained("lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample")
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 lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample

SGLang

How to use lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample 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 "lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample" \
    --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": "lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample",
		"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 "lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample" \
        --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": "lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample with Docker Model Runner:
```
docker model run hf.co/lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample
```

C2_12k_random_sample

This model is a fine-tuned version of Qwen/Qwen2.5-7B-Instruct on the C2_12k_random_sample dataset. It achieves the following results on the evaluation set:

Loss: 0.2589

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: 5e-06
train_batch_size: 1
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 4
total_train_batch_size: 4
total_eval_batch_size: 4
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.3904	0.0418	100	0.3497
0.3128	0.0836	200	0.3307
0.2616	0.1255	300	0.3212
0.2342	0.1673	400	0.3136
0.2543	0.2091	500	0.3083
0.3405	0.2509	600	0.3062
0.2475	0.2928	700	0.3003
0.3254	0.3346	800	0.2890
0.2794	0.3764	900	0.2863
0.2511	0.4182	1000	0.2890
0.2998	0.4601	1100	0.2855
0.2563	0.5019	1200	0.2773
0.2902	0.5437	1300	0.2755
0.2236	0.5855	1400	0.2724
0.2059	0.6274	1500	0.2706
0.207	0.6692	1600	0.2668
0.261	0.7110	1700	0.2655
0.2599	0.7528	1800	0.2637
0.2684	0.7946	1900	0.2624
0.3109	0.8365	2000	0.2608
0.2679	0.8783	2100	0.2598
0.2271	0.9201	2200	0.2586
0.2401	0.9619	2300	0.2591

Framework versions

Transformers 4.46.1
Pytorch 2.6.0+cu124
Datasets 3.1.0
Tokenizers 0.20.3

Downloads last month: 1

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for lemonhat/Qwen2.5-7B-Instruct-C2_12k_random_sample

Base model

Qwen/Qwen2.5-7B

Finetuned

Qwen/Qwen2.5-7B-Instruct

Finetuned

(2636)

this model