Upload LoRA adapter - README.md

README.md

@@ -0,0 +1,163 @@
+---
+library_name: peft
+license: apache-2.0
+base_model: Qwen/Qwen3-4B-Instruct-2507
+tags:
+- axolotl
+- base_model:adapter:Qwen/Qwen3-4B-Instruct-2507
+- lora
+- transformers
+datasets:
+- custom
+pipeline_tag: text-generation
+model-index:
+- name: workspace/checkpoints
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
+<details><summary>See axolotl config</summary>
+
+axolotl version: `0.12.2`
+```yaml
+# 基础模型配置
+base_model: Qwen/Qwen3-4B-Instruct-2507
+load_in_8bit: false
+load_in_4bit: false  # QLoRA才需要4bit
+
+# LoRA 适配器配置 - 这是关键部分
+adapter: lora  # 明确指定使用LoRA
+lora_model_dir:  # 如果有预训练的LoRA权重可以在这里指定
+
+# LoRA 具体参数
+lora_r: 64
+lora_alpha: 128
+lora_dropout: 0.05
+lora_target_modules:  # Qwen3模型的关键模块
+  - q_proj
+  - k_proj
+  - v_proj
+  - o_proj
+  - gate_proj
+  - up_proj
+  - down_proj
+lora_target_linear: true  # 自动找到所有线性层
+lora_fan_in_fan_out: false
+
+# 数据集设置
+chat_template: qwen3
+datasets:
+  - path: /workspace/workspace/tool_data.json
+    type: chat_template
+    roles_to_train: ["assistant"]
+    field_messages: messages
+    message_property_mappings:
+      role: role
+      content: content
+
+dataset_prepared_path:
+val_set_size: 0.05
+output_dir: /workspace/checkpoints
+
+# 序列长度设置
+sequence_len: 10000
+pad_to_sequence_len: true
+sample_packing: false
+eval_sample_packing: false
+group_by_length: true  # 启用长度分组，提高效率
+
+# 训练超参数
+num_epochs: 3
+micro_batch_size: 4  # H100显存大
+gradient_accumulation_steps: 4  # 8卡LoRA不需要太大的累积
+eval_batch_size: 8
+
+# 优化器设置
+optimizer: adamw_torch_fused
+lr_scheduler: cosine_with_restarts
+cosine_restarts: 2  # 每个epoch重启一次
+learning_rate: 4e-5
+warmup_ratio: 0.03
+weight_decay: 0.05
+
+# 精度设置
+bf16: auto  # H100支持bf16
+tf32: true
+gradient_checkpointing: true  # 节省显存
+flash_attention: true
+
+# 日志和保存
+logging_steps: 30
+evals_per_epoch: 1
+saves_per_epoch: 1
+save_total_limit: 3  # 只保留最新的3个checkpoint
+
+# 多卡训练配置 - 使用DeepSpeed而不是FSDP
+deepspeed: /workspace/workspace/zero2.json # 或者直接内联配置
+
+# 其他优化
+ddp_timeout: 3600  # DDP超时设置
+ddp_find_unused_parameters: false  # LoRA通常不需要
+```
+
+</details><br>
+
+# workspace/checkpoints
+
+This model is a fine-tuned version of [Qwen/Qwen3-4B-Instruct-2507](https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507) on the /workspace/workspace/tool_data.json dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0482
+- Memory/max Mem Active(gib): 123.28
+- Memory/max Mem Allocated(gib): 123.28
+- Memory/device Mem Reserved(gib): 124.72
+
+## 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: 4e-05
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 128
+- total_eval_batch_size: 64
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine_with_restarts
+- lr_scheduler_warmup_steps: 39
+- training_steps: 1316
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Mem Active(gib) | Mem Allocated(gib) | Mem Reserved(gib) |
+|:-------------:|:-----:|:----:|:---------------:|:---------------:|:------------------:|:-----------------:|
+| No log        | 0     | 0    | 1.1193          | 123.25          | 123.25             | 124.05            |
+| 0.0509        | 1.0   | 439  | 0.0503          | 123.28          | 123.28             | 124.72            |
+| 0.0461        | 2.0   | 878  | 0.0482          | 123.28          | 123.28             | 124.72            |
+
+
+### Framework versions
+
+- PEFT 0.17.0
+- Transformers 4.55.2
+- Pytorch 2.6.0+cu126
+- Datasets 4.0.0
+- Tokenizers 0.21.4