---
title: Qwen3-14B-QLoRA-SoulChat-R1
canonical_url: "https://www.modelscope.cn/models/wuwukaka/Qwen3-14B-QLoRA-SoulChat-R1"
md_url: "https://www.modelscope.cn/models/wuwukaka/Qwen3-14B-QLoRA-SoulChat-R1.md"
repository: wuwukaka/Qwen3-14B-QLoRA-SoulChat-R1
chinese_name: "智能心理对话模型_Qwen3-14B-QLoRA-SoulChat-R1"
last_updated: 2025-11-26
license: "Apache License 2.0"
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - qwen3
architectures:
  - Qwen3ForCausalLM
base_model:
  - Qwen/Qwen3-14B
base_model_relation: finetune
parameters: 14.8B
tensor_type:
  - BF16
library_name:
  - safetensors
  - gguf
  - pytorch
frameworks:
  - Pytorch
inference_backends:
  - "deploy_task text/emb"
  - "lmdeploy_turbomind 0.9.1"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 189
stars: 3
tags:
  - gguf
---

# Qwen3-14B-QLoRA-SoulChat-R1

> Qwen3-14B-QLoRA-SoulChat-R1 - wuwukaka 在 ModelScope 开源的模型。Qwen3-14B-QLoRA-SoulChat-R1

wuwukaka/Qwen3-14B-QLoRA-SoulChat-R1 是 ModelScope 魔搭社区上的 14.8B 参数text-generation模型，采用 Apache License 2.0 许可，基于 Qwen/Qwen3-14B 构建，可用 deploy_task text/emb、lmdeploy_turbomind 0.9.1、sglang 0.5.2 部署。

- **Repository**: wuwukaka/Qwen3-14B-QLoRA-SoulChat-R1
- **License**: Apache License 2.0
- **Tasks**: text-generation
- **Parameters**: 14.8B
- **Base model**: Qwen/Qwen3-14B
- **Inference backends**: deploy_task text/emb, lmdeploy_turbomind 0.9.1, sglang 0.5.2, vllm 0.9.2
- **Tags**: gguf
- **Downloads**: 189
- **Stars**: 3
- **Last updated**: 2025-11-26

Source: https://www.modelscope.cn/models/wuwukaka/Qwen3-14B-QLoRA-SoulChat-R1

---

# Qwen3-14B-QLoRA-SoulChat-R1

##模型介绍

##本模型基于 **Qwen3-14B**，使用 **QLoRA** 方法在 `SoulChat-R1` 数据集上进行微调，专注于提升模型在心理咨询对话方面的能力。通过 8-bit 量化与 LoRA 低秩适配技术，在保持轻量级的同时显著提升任务表现。
 ##本模型也用于目前尚未完工的智能心理对话项目LumiChat

##微调配置
- top.booster: liger_kernel
- top.finetuning_type: lora
- top.model_name: Qwen3-14B
- top.quantization_bit: '8'
- top.quantization_method: bnb
- top.rope_scaling: none
- top.template: qwen3
- train.additional_target: ''
- train.apollo_rank: 16
- train.apollo_scale: 32
- train.apollo_target: all
- train.apollo_update_interval: 200
- train.badam_mode: layer
- train.badam_switch_interval: 50
- train.badam_switch_mode: ascending
- train.badam_update_ratio: 0.05
- train.batch_size: 3
- train.compute_type: bf16
- train.create_new_adapter: false
- train.cutoff_len: 4096
- train.dataset: SoulChat-R1
- train.dataset_dir: data
- train.ds_offload: false
- train.ds_stage: none
- train.enable_thinking: true
- train.extra_args: '{"optim": "adamw_torch"}'
- train.freeze_extra_modules: ''
- train.freeze_language_model: false
- train.freeze_multi_modal_projector: true
- train.freeze_trainable_layers: 2
- train.freeze_trainable_modules: all
- train.freeze_vision_tower: true
- train.galore_rank: 16
- train.galore_scale: 2
- train.galore_target: all
- train.galore_update_interval: 200
- train.gradient_accumulation_steps: 6
- train.learning_rate: 5e-5
- train.logging_steps: 5
- train.lora_alpha: 16
- train.lora_dropout: 0
- train.lora_rank: 8
- train.lora_target: ''
- train.loraplus_lr_ratio: 0
- train.lr_scheduler_type: cosine
- train.mask_history: false
- train.max_grad_norm: '1.0'
- train.max_samples: '100000'
- train.neat_packing: false
- train.neftune_alpha: 0
- train.num_train_epochs: '8.0'
- train.packing: false
- train.ppo_score_norm: false
- train.ppo_whiten_rewards: false
- train.pref_beta: 0.1
- train.pref_ftx: 0
- train.pref_loss: sigmoid
- train.report_to: none
- train.resize_vocab: false
- train.reward_model: []
- train.save_steps: 100
- train.train_on_prompt: false
- train.training_stage: Supervised Fine-Tuning
- train.use_apollo: false
- train.use_badam: false
- train.use_dora: false
- train.use_galore: false
- train.use_llama_pro: false
- train.use_pissa: false
- train.use_rslora: false
- train.val_size: 0.07
- train.warmup_steps: 0

##训练损失

![](./training_loss.png)

#### 您可以通过如下git clone命令，或者ModelScope SDK来下载模型

SDK下载
```bash
#安装ModelScope
pip install modelscope
```
```python
#SDK模型下载
from modelscope import snapshot_download
model_dir = snapshot_download('wuwukaka/Qwen3-14B-QLoRA-SoulChat-R1')
```
Git下载
```
#Git模型下载
git clone https://www.modelscope.cn/wuwukaka/Qwen3-14B-QLoRA-SoulChat-R1.git
```

<p style="color: lightgrey;">如果您是本模型的贡献者，我们邀请您根据<a href="https://modelscope.cn/docs/ModelScope%E6%A8%A1%E5%9E%8B%E6%8E%A5%E5%85%A5%E6%B5%81%E7%A8%8B%E6%A6%82%E8%A7%88" style="color: lightgrey; text-decoration: underline;">模型贡献文档</a>，及时完善模型卡片内容。</p>