初始化项目，由ModelHub XC社区提供模型

Model: AIR-hl/Qwen2.5-1.5B-ultrachat200k Source: Original Platform
2026-09-13 14:28:42 +08:00
commit 4faf52f712
16 changed files with 307839 additions and 0 deletions
--- a/README.md
+++ b/README.md
@@ -0,0 +1,135 @@
+---
+license: apache-2.0
+datasets:
+- HuggingFaceH4/ultrachat_200k
+base_model:
+- Qwen/Qwen2.5-1.5B
+pipeline_tag: text-generation
+tags:
+- trl
+- qwen
+- sft
+- alignment
+- transformers
+- custome
+- chat
+---
+# Qwen2.5-1.5B-ultrachat200k
+
+
+## Model Details
+
+- **Model type:** sft model
+- **License:** Apache license 2.0
+- **Finetuned from model:** [Qwen/Qwen2.5-1.5B](https://huggingface.co/Qwen/Qwen2.5-1.5B)
+- **Training data:** [HuggingFaceH4/ultrachat_200k](https://huggingface.co/datasets/HuggingFaceH4/ultrachat_200k)
+- **Training framework:** [trl](https://github.com/huggingface/trl)
+
+## Training Details
+
+### Training Hyperparameters
+`attn_implementation`: flash_attention_2 \
+`bf16`: True \
+`learning_rate`: 5e-5 \
+`lr_scheduler_type`: cosine \
+`per_device_train_batch_size`: 2 \
+`gradient_accumulation_steps`: 16 \
+`torch_dtype`: bfloat16 \
+`num_train_epochs`: 1 \
+`max_seq_length`: 2048 \
+`warmup_ratio`: 0.1
+
+### Results
+
+`init_train_loss`: 1.421 \
+`final_train_loss`: 1.192 \
+`eval_loss`: 1.2003
+
+### Training script
+
+```python
+import multiprocessing
+
+from datasets import load_dataset
+from tqdm.rich import tqdm
+from transformers import AutoTokenizer, AutoModelForCausalLM
+from trl import (
+    ModelConfig,
+    SFTTrainer,
+    get_peft_config,
+    get_quantization_config,
+    get_kbit_device_map,
+    SFTConfig,
+    ScriptArguments
+)
+from trl.commands.cli_utils import TrlParser
+
+tqdm.pandas()
+
+if __name__ == "__main__":
+    parser = TrlParser((ScriptArguments, SFTConfig, ModelConfig))
+    args, training_args, model_config = parser.parse_args_and_config()
+
+    quantization_config = get_quantization_config(model_config)
+    model_kwargs = dict(
+        revision=model_config.model_revision,
+        trust_remote_code=model_config.trust_remote_code,
+        attn_implementation=model_config.attn_implementation,
+        torch_dtype=model_config.torch_dtype,
+        use_cache=False if training_args.gradient_checkpointing else True,
+        device_map=get_kbit_device_map() if quantization_config is not None else None,
+        quantization_config=quantization_config,
+    )
+
+    model = AutoModelForCausalLM.from_pretrained(model_config.model_name_or_path,
+                                                 **model_kwargs)
+    tokenizer = AutoTokenizer.from_pretrained(
+        model_config.model_name_or_path, trust_remote_code=model_config.trust_remote_code, use_fast=True
+    )
+    tokenizer.pad_token = tokenizer.eos_token
+
+    train_dataset = load_dataset(args.dataset_name,
+                                 split=args.dataset_train_split,
+                                 num_proc=multiprocessing.cpu_count())
+
+    trainer = SFTTrainer(
+        model=model,
+        args=training_args,
+        train_dataset=train_dataset,
+        processing_class=tokenizer,
+        peft_config=get_peft_config(model_config),
+    )
+
+    trainer.train()
+
+    trainer.save_model(training_args.output_dir)
+```
+
+### Test Script
+```python
+from vllm import LLM
+from datasets import load_dataset
+from vllm.sampling_params import SamplingParams
+from transformers import AutoTokenizer
+
+MODEL_PATH = "autodl-tmp/saves/Qwen2.5-1.5B-ultrachat200k"
+
+model = LLM(MODEL_PATH,
+            tensor_parallel_size=1,
+            dtype='bfloat16')
+tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH)
+
+input = tokenizer.apply_chat_template([{"role": "user", "content": "Which province is Shenyang in?"}],
+                                    tokenize=False,
+                                    add_generation_prompt=True)
+sampling_params = SamplingParams(max_tokens=1024,
+                                 temperature=0.7,
+                                 logprobs=1,
+                                 stop_token_ids=[tokenizer.eos_token_id])
+
+vllm_generations = model.generate(input,
+                                  sampling_params)
+
+print(vllm_generations[0].outputs[0].text)
+# print result: Shenyang is in Liaoning province, China.
+```