chore: 项目已迁移至 Gitee/GitHub，仅保留迁移说明 (tombstone)

- api_correction.py: 嵌入 Astron key, max_workers=5, 修复 batch_convert/multi_temp 返回空列表bug
- scripts/api_correction.py: 同步更新
- train_max.py: 两阶段训练(预训练+角色对齐), Warmup+Cosine, 分层LR, AMP
- distill_daemon.py: 持续蒸馏守护脚本, 15大类种子话题, 5并发

.gitattributes

@@ -1,2 +0,0 @@
-*.json filter=lfs diff=lfs merge=lfs -text
-llm_model_* filter=lfs diff=lfs merge=lfs -text

README.en.md

@@ -1,36 +0,0 @@
-# xiaoli-llm-chat
-
-#### Description
-小狸AI - 自研中文对话大模型
-
-#### Software Architecture
-Software architecture description
-
-#### Installation
-
-1.  xxxx
-2.  xxxx
-3.  xxxx
-
-#### Instructions
-
-1.  xxxx
-2.  xxxx
-3.  xxxx
-
-#### Contribution
-
-1.  Fork the repository
-2.  Create Feat_xxx branch
-3.  Commit your code
-4.  Create Pull Request
-
-
-#### Gitee Feature
-
-1.  You can use Readme\_XXX.md to support different languages, such as Readme\_en.md, Readme\_zh.md
-2.  Gitee blog [blog.gitee.com](https://blog.gitee.com)
-3.  Explore open source project [https://gitee.com/explore](https://gitee.com/explore)
-4.  The most valuable open source project [GVP](https://gitee.com/gvp)
-5.  The manual of Gitee [https://gitee.com/help](https://gitee.com/help)
-6.  The most popular members  [https://gitee.com/gitee-stars/](https://gitee.com/gitee-stars/)

README.md

@@ -1,37 +1,17 @@
-# xiaoli-llm-chat
+# 项目已迁移 ✨
 
-#### 介绍
-小狸AI - 自研中文对话大模型
+你好呀～这个仓库已经搬到更好用的地方啦。
 
-#### 软件架构
-软件架构说明
+为了让代码管理更清爽、多平台同步更省心，我们把原先分散的镜像统一收拢到主仓维护，这里就只保留这张"迁移小纸条"咯。
 
+📦 新家在这里：
+- Gitee：https://gitee.com/shuiyu1123/xiaoli-llm-chat
+- GitHub：https://github.com/shuiyu67/xiaoli-llm-chat
 
-#### 安装教程
+📋 内部评估：
+- 镜像分散在多个平台，重复维护成本高、容易漏更；
+- 自托管 Gitea 实例为个人非管理员环境，长期可用性与备份缺乏保障；
+- 代码已稳定收拢至 Gitee / GitHub 主仓，原镜像不再承载更新职责；
+- 统一主仓后便于协作、CI 与备份，减少多端同步负担。
 
-1.  xxxx
-2.  xxxx
-3.  xxxx
-
-#### 使用说明
-
-1.  xxxx
-2.  xxxx
-3.  xxxx
-
-#### 参与贡献
-
-1.  Fork 本仓库
-2.  新建 Feat_xxx 分支
-3.  提交代码
-4.  新建 Pull Request
-
-
-#### 特技
-
-1.  使用 Readme\_XXX.md 来支持不同的语言，例如 Readme\_en.md, Readme\_zh.md
-2.  Gitee 官方博客 [blog.gitee.com](https://blog.gitee.com)
-3.  你可以 [https://gitee.com/explore](https://gitee.com/explore) 这个地址来了解 Gitee 上的优秀开源项目
-4.  [GVP](https://gitee.com/gvp) 全称是 Gitee 最有价值开源项目，是综合评定出的优秀开源项目
-5.  Gitee 官方提供的使用手册 [https://gitee.com/help](https://gitee.com/help)
-6.  Gitee 封面人物是一档用来展示 Gitee 会员风采的栏目 [https://gitee.com/gitee-stars/](https://gitee.com/gitee-stars/)
+旧镜像不再更新，但代码都好好的在新家继续生长。感谢你一路的关注与陪伴 💛

all_data_cleaned.json

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:136863dba34ed2300a56611600b9cd758aa6e01cdd7e3dcdd35d0371be32d41c
-size 43711

chat_server.py

@@ -1,143 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-小狸聊天 Web 服务 v2 - 支持多轮对话
-"""
-import sys, os, json, re
-sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
-
-from flask import Flask, render_template, request, jsonify
-import torch
-from llm_chat_improved import (
-    CharTokenizer, build_vocabulary, LLMModel, LLMChat, load_training_data
-)
-
-app = Flask(__name__)
-
-# 全局模型
-llm_chat = None
-vocab = None
-
-def clean_text(text):
-    """清洗文本"""
-    text = text.replace('\n', '')
-    text = re.sub(r'\*\*([^*]+)\*\*', r'\1', text)
-    text = re.sub(r'\([^)]*[ω｡ﾉ♡^][^)]*\)', '', text)
-    emoji_pattern = re.compile(
-        '[\U0001F600-\U0001F64F\U0001F300-\U0001F5FF\U0001F680-\U0001F6FF\U0001F1E0-\U0001F1FF\U00002702-\U000027B0\U0001f926-\U0001f937\u2640-\u2642\u2600-\u2B55\u200d\u23cf\u23e9\u231a\ufe0f\u3030\u2728\u2714\u2716\u27a1\u2b05\u2b06\u2b07\u2b1b\u2b1c\u2b50\u2b55\u2934\u2935\u25aa\u25ab\u25b6\u25c0\u25fb\u25fc\u25fd\u25fe]+',
-        flags=re.UNICODE
-    )
-    text = emoji_pattern.sub('', text)
-    text = re.sub(r'[ \t]+', ' ', text)
-    return text.strip()
-
-def load_model():
-    """加载训练好的模型"""
-    global llm_chat, vocab
-    
-    print("🐱 正在加载小狸模型...")
-    
-    device = 'cpu'
-    
-    # 直接从模型文件加载 vocab 和配置 (v4大模型)
-    checkpoint = torch.load('xiaoli_final_v4.pth', map_location=device)
-    vocab = checkpoint['vocab']
-    config = checkpoint.get('model_config', {})
-    
-    print(f"   词汇表大小: {len(vocab)}")
-    
-    # 创建模型 - 使用保存的配置（v4大模型带MoE+Thinking+Memory）
-    model = LLMModel(
-        vocab_size=config.get('vocab_size', len(vocab)),
-        d_model=config.get('d_model', 384),
-        n_heads=config.get('n_heads', 8),
-        n_layers=config.get('n_layers', 6),
-        d_ff=config.get('d_ff', 1536),
-        max_len=config.get('max_len', 128),
-        dropout=0.0,
-        use_memory=config.get('use_memory', True),
-        use_thinking=config.get('use_thinking', True),
-        use_moe=config.get('use_moe', True),
-        num_experts=config.get('num_experts', 4),
-        top_k=config.get('top_k', 2),
-    )
-    
-    # 加载权重
-    model.load_state_dict(checkpoint['model_state_dict'])
-    
-    llm_chat = LLMChat(model, vocab, device, 128)
-    print("✅ 小狸已就绪！")
-
-# 存储会话历史
-session_history = {}
-
-@app.route('/')
-def index():
-    return render_template('chat.html')
-
-@app.route('/chat', methods=['POST'])
-def chat():
-    data = request.json
-    message = data.get('message', '').strip()
-    session_id = data.get('session_id', 'default')
-    
-    if not message:
-        return jsonify({'response': '请输入内容哦~'})
-    
-    if llm_chat is None:
-        return jsonify({'response': '模型加载中，请稍后再试'})
-    
-    try:
-        # 获取会话历史
-        history = session_history.get(session_id, [])
-        
-        # 构建带上下文的输入（使用与训练一致的格式）
-        # 训练格式: <SOS>问题<SEP><START>
-        # 多轮格式: <SOS>历史问题1<SEP>历史回答1<SEP>历史问题2<SEP>历史回答2<SEP>当前问题<SEP><START>
-        if history and len(history) >= 2:
-            # 只保留最近2轮对话（4条记录）
-            recent_history = history[-4:]
-            # 提取问题和答案
-            context_parts = []
-            for i in range(0, len(recent_history), 2):
-                if i + 1 < len(recent_history):
-                    q = recent_history[i].replace("用户：", "")
-                    a = recent_history[i+1].replace("小狸：", "")
-                    context_parts.append(f"{q}<SEP>{a}")
-            
-            if context_parts:
-                context = "<SEP>".join(context_parts)
-                full_input = f"{context}<SEP>{message}"
-            else:
-                full_input = message
-        else:
-            full_input = message
-        
-        # 限制输入长度（字符数）
-        if len(full_input) > 80:
-            full_input = full_input[-80:]
-        
-        response = llm_chat.generate_response(full_input, max_length=50, temperature=0.3)
-        
-        # 更新历史
-        history.append(f"用户：{message}")
-        history.append(f"小狸：{response}")
-        # 只保留最近6轮（12条记录）
-        session_history[session_id] = history[-12:]
-        
-        return jsonify({'response': response})
-    except Exception as e:
-        import traceback
-        return jsonify({'response': f'出错了: {str(e)}\n{traceback.format_exc()}'})
-
-@app.route('/clear', methods=['POST'])
-def clear_history():
-    data = request.json
-    session_id = data.get('session_id', 'default')
-    session_history[session_id] = []
-    return jsonify({'status': 'ok'})
-
-if __name__ == '__main__':
-    load_model()
-    print("\n🌐 启动 Web 服务: http://0.0.0.0:5000")
-    app.run(host='0.0.0.0', port=5000, debug=False)

multiturn_data.json

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:7c4d05531958838c1c6750216a98fc781ea13085b278a0154eb1378680959fe9
-size 10038

templates/chat.html

@@ -1,309 +0,0 @@
-<!DOCTYPE html>
-<html lang="zh-CN">
-<head>
-    <meta charset="UTF-8">
-    <meta name="viewport" content="width=device-width, initial-scale=1.0">
-    <title>小狸 - AI 聊天</title>
-    <style>
-        * {
-            margin: 0;
-            padding: 0;
-            box-sizing: border-box;
-        }
-        body {
-            font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, sans-serif;
-            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
-            height: 100vh;
-            display: flex;
-            justify-content: center;
-            align-items: center;
-        }
-        .chat-container {
-            width: 90%;
-            max-width: 500px;
-            height: 80vh;
-            background: white;
-            border-radius: 20px;
-            box-shadow: 0 20px 60px rgba(0,0,0,0.3);
-            display: flex;
-            flex-direction: column;
-            overflow: hidden;
-        }
-        .chat-header {
-            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
-            color: white;
-            padding: 20px;
-            text-align: center;
-        }
-        .chat-header h1 {
-            font-size: 24px;
-            margin-bottom: 5px;
-        }
-        .chat-header p {
-            font-size: 12px;
-            opacity: 0.9;
-        }
-        .chat-messages {
-            flex: 1;
-            overflow-y: auto;
-            padding: 20px;
-            background: #f5f5f5;
-        }
-        .message {
-            margin-bottom: 15px;
-            display: flex;
-            align-items: flex-start;
-        }
-        .message.user {
-            justify-content: flex-end;
-        }
-        .message.bot {
-            justify-content: flex-start;
-        }
-        .message-content {
-            max-width: 70%;
-            padding: 12px 16px;
-            border-radius: 18px;
-            font-size: 14px;
-            line-height: 1.5;
-            word-wrap: break-word;
-        }
-        .message.user .message-content {
-            background: #667eea;
-            color: white;
-            border-bottom-right-radius: 4px;
-        }
-        .message.bot .message-content {
-            background: white;
-            color: #333;
-            border-bottom-left-radius: 4px;
-            box-shadow: 0 2px 5px rgba(0,0,0,0.1);
-        }
-        .avatar {
-            width: 36px;
-            height: 36px;
-            border-radius: 50%;
-            display: flex;
-            align-items: center;
-            justify-content: center;
-            font-size: 18px;
-            margin: 0 8px;
-        }
-        .chat-input {
-            padding: 15px 20px;
-            background: white;
-            border-top: 1px solid #eee;
-            display: flex;
-            gap: 10px;
-        }
-        .chat-input input {
-            flex: 1;
-            padding: 12px 16px;
-            border: 2px solid #e0e0e0;
-            border-radius: 25px;
-            font-size: 14px;
-            outline: none;
-            transition: border-color 0.3s;
-        }
-        .chat-input input:focus {
-            border-color: #667eea;
-        }
-        .chat-input button {
-            padding: 12px 24px;
-            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
-            color: white;
-            border: none;
-            border-radius: 25px;
-            font-size: 14px;
-            cursor: pointer;
-            transition: transform 0.2s, box-shadow 0.2s;
-        }
-        .chat-input button:hover {
-            transform: translateY(-2px);
-            box-shadow: 0 5px 15px rgba(102, 126, 234, 0.4);
-        }
-        .chat-input button:disabled {
-            opacity: 0.6;
-            cursor: not-allowed;
-            transform: none;
-        }
-        .typing {
-            display: flex;
-            gap: 4px;
-            padding: 12px 16px;
-        }
-        .typing span {
-            width: 8px;
-            height: 8px;
-            background: #999;
-            border-radius: 50%;
-            animation: typing 1.4s infinite;
-        }
-        .typing span:nth-child(2) { animation-delay: 0.2s; }
-        .typing span:nth-child(3) { animation-delay: 0.4s; }
-        @keyframes typing {
-            0%, 60%, 100% { transform: translateY(0); }
-            30% { transform: translateY(-10px); }
-        }
-        .start-overlay {
-            position: fixed;
-            top: 0; left: 0; right: 0; bottom: 0;
-            background: rgba(0,0,0,0.6);
-            display: flex;
-            justify-content: center;
-            align-items: center;
-            z-index: 1000;
-        }
-        .start-overlay.hidden {
-            display: none;
-        }
-        .start-card {
-            background: white;
-            border-radius: 24px;
-            padding: 50px 40px;
-            text-align: center;
-            box-shadow: 0 20px 60px rgba(0,0,0,0.3);
-            animation: popIn 0.5s ease;
-        }
-        @keyframes popIn {
-            0% { transform: scale(0.8); opacity: 0; }
-            100% { transform: scale(1); opacity: 1; }
-        }
-        .start-avatar {
-            font-size: 64px;
-            margin-bottom: 15px;
-        }
-        .start-card h2 {
-            color: #333;
-            font-size: 28px;
-            margin-bottom: 10px;
-        }
-        .start-card p {
-            color: #666;
-            font-size: 16px;
-            margin-bottom: 30px;
-        }
-        .start-btn {
-            padding: 16px 40px;
-            font-size: 18px;
-            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
-            color: white;
-            border: none;
-            border-radius: 30px;
-            cursor: pointer;
-            transition: transform 0.2s, box-shadow 0.2s;
-        }
-        .start-btn:hover {
-            transform: translateY(-3px);
-            box-shadow: 0 10px 25px rgba(102, 126, 234, 0.5);
-        }
-    </style>
-</head>
-<body>
-    <div class="chat-container">
-        <div class="chat-header">
-            <h1>🐱 小狸</h1>
-            <p>你的 AI 小伙伴</p>
-        </div>
-        <div class="chat-messages" id="messages">
-            <div class="message bot">
-                <div class="avatar">🐱</div>
-                <div class="message-content">你好！我是小狸，很高兴见到你！有什么想聊的吗？</div>
-            </div>
-        </div>
-        <div class="chat-input">
-            <input type="text" id="userInput" placeholder="输入消息..." maxlength="100">
-            <button id="sendBtn" onclick="sendMessage()">发送</button>
-        </div>
-        <div class="start-overlay" id="startOverlay">
-            <div class="start-card">
-                <div class="start-avatar">🐱</div>
-                <h2>小狸已就绪！</h2>
-                <p>你的 AI 小伙伴正在等你聊天~</p>
-                <button class="start-btn" onclick="startChat()">👉 点击开始聊天</button>
-            </div>
-        </div>
-    </div>
-
-    <script>
-        const messagesDiv = document.getElementById('messages');
-        const userInput = document.getElementById('userInput');
-        const sendBtn = document.getElementById('sendBtn');
-        const sessionId = 'session_' + Math.random().toString(36).substr(2, 9);
-
-        function addMessage(text, isUser) {
-            const div = document.createElement('div');
-            div.className = `message ${isUser ? 'user' : 'bot'}`;
-            div.innerHTML = `
-                ${isUser ? '' : '<div class="avatar">🐱</div>'}
-                <div class="message-content">${escapeHtml(text)}</div>
-                ${isUser ? '<div class="avatar">👤</div>' : ''}
-            `;
-            messagesDiv.appendChild(div);
-            messagesDiv.scrollTop = messagesDiv.scrollHeight;
-        }
-
-        function showTyping() {
-            const div = document.createElement('div');
-            div.className = 'message bot';
-            div.id = 'typing-indicator';
-            div.innerHTML = `
-                <div class="avatar">🐱</div>
-                <div class="message-content">
-                    <div class="typing">
-                        <span></span><span></span><span></span>
-                    </div>
-                </div>
-            `;
-            messagesDiv.appendChild(div);
-            messagesDiv.scrollTop = messagesDiv.scrollHeight;
-        }
-
-        function hideTyping() {
-            const typing = document.getElementById('typing-indicator');
-            if (typing) typing.remove();
-        }
-
-        function escapeHtml(text) {
-            const div = document.createElement('div');
-            div.textContent = text;
-            return div.innerHTML;
-        }
-
-        async function sendMessage() {
-            const text = userInput.value.trim();
-            if (!text) return;
-
-            addMessage(text, true);
-            userInput.value = '';
-            sendBtn.disabled = true;
-            showTyping();
-
-            try {
-                const response = await fetch('/chat', {
-                    method: 'POST',
-                    headers: { 'Content-Type': 'application/json' },
-                    body: JSON.stringify({ message: text, session_id: sessionId })
-                });
-                const data = await response.json();
-                hideTyping();
-                addMessage(data.response, false);
-            } catch (error) {
-                hideTyping();
-                addMessage('抱歉，我出错了，请稍后再试~', false);
-            }
-
-            sendBtn.disabled = false;
-        }
-
-        userInput.addEventListener('keypress', (e) => {
-            if (e.key === 'Enter') sendMessage();
-        });
-
-        function startChat() {
-            document.getElementById('startOverlay').classList.add('hidden');
-            userInput.focus();
-        }
-    </script>
-</body>
-</html>

xiaoli_data.json

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:c0c0848f74cc7604d41699134a5f34f54d5958779a6653e27973ef495790c29c
-size 5267