Text Generation
Transformers
Safetensors
English
qwen3
text-generation-inference
unsloth
conversational
Instructions to use CompassioninMachineLearning/qwen-compassion-part1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use CompassioninMachineLearning/qwen-compassion-part1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="CompassioninMachineLearning/qwen-compassion-part1") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("CompassioninMachineLearning/qwen-compassion-part1") model = AutoModelForCausalLM.from_pretrained("CompassioninMachineLearning/qwen-compassion-part1", 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=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use CompassioninMachineLearning/qwen-compassion-part1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "CompassioninMachineLearning/qwen-compassion-part1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "CompassioninMachineLearning/qwen-compassion-part1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/CompassioninMachineLearning/qwen-compassion-part1
- SGLang
How to use CompassioninMachineLearning/qwen-compassion-part1 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 "CompassioninMachineLearning/qwen-compassion-part1" \ --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": "CompassioninMachineLearning/qwen-compassion-part1", "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 "CompassioninMachineLearning/qwen-compassion-part1" \ --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": "CompassioninMachineLearning/qwen-compassion-part1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Unsloth Desktop
- Docker Model Runner
How to use CompassioninMachineLearning/qwen-compassion-part1 with Docker Model Runner:
docker model run hf.co/CompassioninMachineLearning/qwen-compassion-part1
add run_manifest.json
Browse files- run_manifest.json +86 -0
run_manifest.json
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_name": "qwen-compassion-part1",
|
| 3 |
+
"git_sha": "6950fbbc6f29902d7fc5b26c1aef8d053966e9a3",
|
| 4 |
+
"argv": [
|
| 5 |
+
"pilot/train_unsloth.py",
|
| 6 |
+
"--data",
|
| 7 |
+
"/workspace/anchor/qwen-compassion/train_qwen-compassion_replay50.jsonl",
|
| 8 |
+
"--base-model",
|
| 9 |
+
"CompassioninMachineLearning/Qwen3-8b-compassion-cleaned-10k-CPT-best-step750",
|
| 10 |
+
"--no-4bit",
|
| 11 |
+
"--lora-r",
|
| 12 |
+
"64",
|
| 13 |
+
"--lora-alpha",
|
| 14 |
+
"128",
|
| 15 |
+
"--seed",
|
| 16 |
+
"7",
|
| 17 |
+
"--epochs",
|
| 18 |
+
"1",
|
| 19 |
+
"--eval-steps",
|
| 20 |
+
"50",
|
| 21 |
+
"--early-stopping-patience",
|
| 22 |
+
"2",
|
| 23 |
+
"--snapshot-frac",
|
| 24 |
+
"0",
|
| 25 |
+
"--replay-chat",
|
| 26 |
+
"--replay-token-frac",
|
| 27 |
+
"0.25",
|
| 28 |
+
"--run-name",
|
| 29 |
+
"qwen-compassion-part1",
|
| 30 |
+
"--push-to-hub-merged",
|
| 31 |
+
"CompassioninMachineLearning/qwen-compassion-part1-build",
|
| 32 |
+
"--output-dir",
|
| 33 |
+
"/workspace/anchor/qwen-compassion/train"
|
| 34 |
+
],
|
| 35 |
+
"base_model": "CompassioninMachineLearning/Qwen3-8b-compassion-cleaned-10k-CPT-best-step750",
|
| 36 |
+
"template_source": "CompassioninMachineLearning/Qwen3-8b-compassion-cleaned-10k-CPT-best-step750",
|
| 37 |
+
"modules_to_save": [
|
| 38 |
+
"embed_tokens",
|
| 39 |
+
"lm_head"
|
| 40 |
+
],
|
| 41 |
+
"data_files": [
|
| 42 |
+
"/workspace/anchor/qwen-compassion/train_qwen-compassion_replay50.jsonl"
|
| 43 |
+
],
|
| 44 |
+
"data_sha256": {
|
| 45 |
+
"/workspace/anchor/qwen-compassion/train_qwen-compassion_replay50.jsonl": "1f22b1da2039471bb1932e866f1a1138a08f3e8c1eee8e94dfb2dc19a3b9b2e8"
|
| 46 |
+
},
|
| 47 |
+
"rows_raw": 18500,
|
| 48 |
+
"rows_kept": 10076,
|
| 49 |
+
"by_source": {
|
| 50 |
+
"replay:compassion_12185_cleaned": 1259,
|
| 51 |
+
"recovery": 3750,
|
| 52 |
+
"apigen": 4567,
|
| 53 |
+
"efficiency": 500
|
| 54 |
+
},
|
| 55 |
+
"train_examples": 9824,
|
| 56 |
+
"eval_examples": 252,
|
| 57 |
+
"hyperparams": {
|
| 58 |
+
"lora_r": 64,
|
| 59 |
+
"lora_alpha": 128,
|
| 60 |
+
"lora_dropout": 0.05,
|
| 61 |
+
"lr_adapters": 0.0002,
|
| 62 |
+
"lr_embeddings": 2e-05,
|
| 63 |
+
"epochs": 1.0,
|
| 64 |
+
"max_steps": -1,
|
| 65 |
+
"max_seq_len": 8192,
|
| 66 |
+
"effective_batch": 16,
|
| 67 |
+
"per_device_batch": 2,
|
| 68 |
+
"grad_accum": 8,
|
| 69 |
+
"warmup_ratio": 0.05,
|
| 70 |
+
"schedule": "cosine",
|
| 71 |
+
"seed": 7,
|
| 72 |
+
"snapshot_frac": 0.0
|
| 73 |
+
},
|
| 74 |
+
"versions": {
|
| 75 |
+
"torch": "2.11.0+cu128",
|
| 76 |
+
"transformers": "4.57.6",
|
| 77 |
+
"unsloth": "2026.9.2",
|
| 78 |
+
"trl": "0.22.2",
|
| 79 |
+
"datasets": "5.0.1",
|
| 80 |
+
"peft": "0.20.0",
|
| 81 |
+
"accelerate": "1.14.0",
|
| 82 |
+
"bitsandbytes": "0.50.2"
|
| 83 |
+
},
|
| 84 |
+
"gpu": "NVIDIA H100 NVL",
|
| 85 |
+
"started_at": "2026-10-11T23:26:44Z"
|
| 86 |
+
}
|