{"behavior":"degenerate_repetition","label_filter":null,"limit":10,"next_offset":7414,"offset":7404,"previous_offset":null,"rows":[{"assistant_initiates":false,"asst_text":"\u4f60\u9047\u5230\u7684\u9519\u8bef\u662f\u56e0\u4e3a\u5728 Python \u4e2d\u5c1d\u8bd5\u5c06\u4e00\u4e2a\u6574\u6570 (`len(context)`) \u4e0e\u4e00\u4e2a\u5b57\u7b26\u4e32\uff08`\"\\n\"`) \u8fdb\u884c\u5b57\u7b26\u4e32\u62fc\u63a5\u3002\u8fd9\u662f\u4e00\u4e2a\u7c7b\u578b\u9519\u8bef\uff0c\u65e0\u6cd5\u76f4\u63a5\u8fdb\u884c\u8fd9\u79cd\u64cd\u4f5c\u3002\n\n\u4f60\u53ef\u4ee5\u901a\u8fc7\u5c06\u6574\u6570\u8f6c\u6362\u4e3a\u5b57\u7b26\u4e32\u6765\u89e3\u51b3\u8fd9\u4e2a\u95ee\u9898\u3002\n\n### \u4fee\u6539\u4ee3\u7801\n\n\u4f60\u7684\u4ee3\u7801\u793a\u4f8b\u5982\u4e0b\uff1a\n```python\nprint(len(context) + \"\\n\")\n```\n\n\u53ef\u4ee5\u4fee\u6539\u4e3a\u4ee5\u4e0b\u4efb\u610f\u4e00\u79cd\u65b9\u5f0f\uff1a\n\n#### \u4f7f\u7528\u5b57\u7b26\u4e32\u683c\u5f0f\u5316\u51fd\u6570 `str()`\n```python\nprint(str(len(context)) + \"\\n\")\n```\n\n#### \u4f7f\u7528 f-string\uff08Python 3.6 \u53ca\u4ee5\u4e0a\u7248\u672c\uff09\n```python\nprint(f\"{len(context)}\\n\")\n```\n\n#### \u4f7f\u7528 `format` \u65b9\u6cd5\n```python\nprint(\"{}\\n\".format(len(context)))\n```\n\n#### \u4f7f\u7528\u9017\u53f7\u5206\u9694\u53c2\u6570\n\u8fd9\u79cd\u65b9\u5f0f\u4e0d\u9700\u8981\u663e\u5f0f\u8f6c\u6362\uff0c\u4f1a\u5728\u4e0d\u540c\u7c7b\u578b\u4e4b\u95f4\u63d2\u5165\u4e00\u4e2a\u7a7a\u683c\uff1a\n```python\nprint(len(context), \"\\n\")\n```\n\n### \u9009\u62e9\u9002\u5408\u7684\u65b9\u5f0f\n\n\u5047\u8bbe `context` \u662f\u4f60\u5728\u7a0b\u5e8f\u4e2d\u4f20\u9012\u7684\u4e0a\u4e0b\u6587\u5185\u5bb9\uff0c\u793a\u4f8b\u5982\u4e0b\uff1a\n\n```python\ndef print_paragraphs(context):\n print(f\"The length of the context is: {len(context)}\\n\")\n # \u4f60\u5176\u4ed6\u7684\u4ee3\u7801\n\n# \u793a\u4f8b\u8c03\u7528\ncontext = \"\u8fd9\u662f\u4e00\u4e2a\u793a\u4f8b\u4e0a\u4e0b\u6587\u3002\" # \u4f8b\u5982\u66ff\u6362\u4e3a\u5b9e\u9645\u7684\u4e0a\u4e0b\u6587\u5185\u5bb9\nprint_paragraphs(context)\n```\n\n### \u793a\u4f8b\u5b8c\u6574\u51fd\u6570\n\n\u5982\u679c\u4f60\u60f3\u786e\u4fdd\u5728\u4e0d\u540c\u4e0a\u4e0b\u6587\u4e2d\u90fd\u80fd\u6b63\u5e38\u5de5\u4f5c\uff0c\u4ee5\u4e0b\u662f\u4e00\u4e2a\u5b8c\u6574\u793a\u4f8b\uff1a\n\n```python\ndef print_paragraphs(context):\n # \u8f93\u51fa\u4e0a\u4e0b\u6587\u957f\u5ea6\n print(f\"The length of the context is: {len(context)}\\n\")\n\n # \u793a\u4f8b\uff1a\u8f93\u51fa\u4e0a\u4e0b\u6587\n print(\"Context:\")\n print(context)\n\n# \u793a\u4f8b\u8c03\u7528\ncontext = \"\u8fd9\u662f\u4e00\u4e2a\u793a\u4f8b\u4e0a\u4e0b\u6587\u3002\" # \u4f8b\u5982\u66ff\u6362\u4e3a\u5b9e\u9645\u7684\u4e0a\u4e0b\u6587\u5185\u5bb9\nprint_paragraphs(context)\n```\n\n\u5728\u8fd9\u4e2a\u4f8b\u5b50\u4e2d\uff0c\u4f7f\u7528\u4e86 `f-string` \u6765\u683c\u5f0f\u5316\u5b57\u7b26\u4e32\uff0c\u65b9\u4fbf\u4e14\u76f4\u89c2\u5730\u5c06\u4e0a\u4e0b\u6587\u7684\u957f\u5ea6\u5d4c\u5165\u5b57\u7b26\u4e32\u4e2d\u3002\n\n\u901a\u8fc7\u4e0a\u8ff0\u4fee\u6539\uff0c\u4f60\u5e94\u8be5\u80fd\u591f\u6210\u529f\u8fd0\u884c\u5e76\u8f93\u51fa `context` \u7684\u957f\u5ea6\u3002","behavior":"degenerate_repetition","conversation_hash":"44b305e7e7991de62b21af46e7a54afe","conversation_redacted":false,"conversation_toxic":false,"country":"United States","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Chinese","model":"gpt-4o-2024-05-13","n_chars":1316,"n_turns_in_conversation":16,"n_words":125,"source_row_offset":7404,"source_shard":"turns-00026.parquet","timestamp":"Sat, 27 Jul 2024 08:10:57 GMT","turn_id":"5f08d2172f3fb8d050dadf72","turn_index":9,"user_text":" File \"/g5500-nas/960G/srt/py/cmsplit.py\", line 68, in print_paragraphs\n print(len(context) + \"\\n\")\n ~~~~~~~~~~~~~^~~~~~\nTypeError: unsupported operand type(s) for +: 'int' and 'str'\uff0c\u5982\u4f55\u4fee\u6539\uff1f"},{"assistant_initiates":false,"asst_text":"\u6839\u636e\u4f60\u63d0\u4f9b\u7684\u4fe1\u606f\uff0cOllama API \u8c03\u7528\u4e2d\u7684\u65e5\u5fd7\u4fe1\u606f `INFO [update_slots] input truncated` \u8868\u793a\u5728\u591a\u6b21 API \u8c03\u7528\u8fc7\u7a0b\u4e2d\uff0c\u4e0a\u4e0b\u6587\uff08context\uff09\u8d85\u8fc7\u4e86\u6a21\u578b\u7684\u6700\u5927\u4e0a\u4e0b\u6587\u957f\u5ea6\uff084096 tokens\uff09\uff0c\u56e0\u6b64\u88ab\u622a\u65ad\u3002\n\n### Context \u884c\u4e3a\u89e3\u91ca\n\n\u4f60\u7684\u95ee\u9898\u6d89\u53ca\u4e0a\u4e0b\u6587\u5728\u591a\u6b21\u8c03\u7528\u4e2d\u7684\u7d2f\u79ef\u884c\u4e3a\u3002\u901a\u5e38\u60c5\u51b5\u4e0b\uff0cAPI \u4f1a\u6309\u7167\u5982\u4e0b\u65b9\u5f0f\u7ba1\u7406\u4e0a\u4e0b\u6587\uff1a\n\n- **\u7d2f\u52a0\u4e0a\u4e0b\u6587**\uff1a\u6bcf\u6b21\u8c03\u7528\u65f6\u7684\u4e0a\u4e0b\u6587\u4e0d\u4ec5\u5305\u542b\u5f53\u524d\u8f93\u5165\uff0c\u8fd8\u5305\u62ec\u4e4b\u524d\u8c03\u7528\u8fd4\u56de\u7684\u5185\u5bb9\u3002\u8fd9\u610f\u5473\u7740\u4e0a\u4e0b\u6587\u4f1a\u9010\u6b65\u7d2f\u79ef\uff0c\u9010\u6e10\u53d8\u5f97\u8d8a\u6765\u8d8a\u957f\u3002\n- **\u8d85\u8fc7\u9650\u5236\u622a\u65ad**\uff1a\u4e00\u65e6\u4e0a\u4e0b\u6587\u7684\u957f\u5ea6\u8d85\u8fc7\u6a21\u578b\u7684\u6700\u5927\u9650\u5236\uff08\u4f8b\u5982 4096 tokens\uff09\uff0c\u7cfb\u7edf\u4f1a\u4fdd\u7559\u6700\u65b0\u7684\u4e0a\u4e0b\u6587\u5e76\u622a\u65ad\u591a\u4f59\u7684\u90e8\u5206\u3002\n\n### \u4e0a\u4e0b\u6587\u7ba1\u7406\u7b56\u7565\n\n\u4e3a\u4e86\u9632\u6b62\u4e0a\u4e0b\u6587\u8fc7\u957f\u5bfc\u81f4\u7684\u622a\u65ad\uff0c\u4f60\u9700\u8981\u5728\u6bcf\u6b21\u8c03\u7528\u4e4b\u524d\u5bf9\u4e0a\u4e0b\u6587\u8fdb\u884c\u5408\u7406\u7ba1\u7406\u3002\u8fd9\u901a\u5e38\u6d89\u53ca\u4ee5\u4e0b\u51e0\u4e2a\u6b65\u9aa4\uff1a\n\n1. **\u62c6\u5206\u8f93\u5165**\uff1a\u907f\u514d\u4e00\u6b21\u6027\u8f93\u5165\u8fc7\u957f\u6587\u672c\uff0c\u4f7f\u7528\u6ed1\u52a8\u7a97\u53e3\u6216\u5206\u6bb5\u65b9\u6cd5\u5904\u7406\u957f\u6587\u672c\u3002\n2. **\u9650\u5236\u4e0a\u4e0b\u6587\u957f\u5ea6**\uff1a\u5728\u6bcf\u6b21\u8c03\u7528\u4e4b\u524d\u786e\u4fdd\u4e0a\u4e0b\u6587\u7684\u603b\u957f\u5ea6\u4e0d\u8d85\u8fc7\u6700\u5927\u9650\u5236\u3002\n3. **\u91cd\u7528\u4e0a\u4e0b\u6587**\uff1a\u4eceAPI\u8fd4\u56de\u7684\u4e0a\u4e0b\u6587\u4e2d\u4fdd\u7559\u6700\u8fd1\u7684\u90e8\u5206\uff0c\u7136\u540e\u5728\u4e0b\u6b21\u8c03\u7528\u65f6\u7ee7\u7eed\u4f7f\u7528\u3002\n\n### \u793a\u4f8b\u4ee3\u7801\n\n\u5047\u8bbe\u4f60\u6709\u4e00\u4e2a\u957f\u6587\u672c `input_texts` \u9700\u8981\u9010\u6bb5\u5904\u7406\uff0c\u5e76\u4e14\u4fdd\u7559\u4e0a\u4e0b\u6587\u8fdb\u884c\u8fde\u7eed\u5bf9\u8bdd\uff1a\n\n```python\nimport requests\n\nAPI_URL = \"https://api.ollama.com/your-endpoint\"\nAPI_KEY = \"YOUR_API_KEY\"\n\nheaders = {\n \"Authorization\": f\"Bearer {API_KEY}\"\n}\n\ndef query(payload):\n response = requests.post(API_URL, headers=headers, json=payload)\n return response.json()\n\n# \u521d\u59cb\u5316\u4e0a\u4e0b\u6587\nfull_context = \"\"\n\ndef update_context(new_text, max_length=4096):\n global full_context\n full_context += \" \" + new_text # \u7d2f\u79ef\u4e0a\u4e0b\u6587\n tokens = full_context.split()\n if len(tokens) > max_length:\n full_context = ' '.join(tokens[-max_length:]) # \u622a\u65ad\u6700\u65e9\u7684\u90e8\u5206\n return full_context\n\n# \u793a\u4f8b: \u591a\u6b21\u8c03\u7528 API \u5e76\u7ba1\u7406\u4e0a\u4e0b\u6587\ninput_texts = [\"Your first input text here\", \"Your second input text here\", \"Your third input text here\"]\n\nfor input_text in input_texts:\n # \u66f4\u65b0\u4e0a\u4e0b\u6587\n context = update_context(input_text)\n payload = {\"inputs\": context}\n response = query(payload)\n \n # \u5047\u8bbe\u4f60\u9700\u8981\u4eceresponse\u4e2d\u63d0\u53d6\u4e00\u4e9b\u8fd4\u56de\u7684\u4e0a\u4e0b\u6587\u4ee5\u4fbf\u4e8e\u4e0b\u4e00\u6b21\u8c03\u7528\n new_text_from_response = response.get(\"new_text\", \"\")\n \n # \u6dfb\u52a0\u65b0\u7684\u8fd4\u56de\u5230\u4e0a\u4e0b\u6587\u4e2d\n context = update_context(new_text_from_response)\n \n print(response)\n```\n\n### \u5177\u4f53\u89e3\u91ca\n\n1. **\u62c6\u5206\u8f93\u5165**\uff1a\u5c06 `input_texts` \u4f5c\u4e3a\u4e00\u4e2a\u5217\u8868\uff0c\u9010\u6bb5\u5904\u7406\u3002\n2. **\u66f4\u65b0\u4e0a\u4e0b\u6587**\uff1a\u901a\u8fc7 `update_context` \u51fd\u6570\u7ba1\u7406\u4e0a\u4e0b\u6587\u957f\u5ea6\uff0c\u786e\u4fdd\u4e0d\u8d85\u8fc7 4096 \u4e2a tokens\u3002\n3. **\u8c03\u7528 API**\uff1a\u5c06\u5f53\u524d\u4e0a\u4e0b\u6587\u53d1\u9001\u7ed9 API \u5e76\u83b7\u53d6\u54cd\u5e94\u3002\n4. **\u5904\u7406\u54cd\u5e94**\uff1a\u4ece\u54cd\u5e94\u4e2d\u63d0\u53d6\u65b0\u7684\u6587\u672c\uff0c\u5e76\u66f4\u65b0\u4e0a\u4e0b\u6587\uff0c\u4ee5\u4fbf\u5728\u4e0b\u4e00\u6b21\u8c03\u7528\u65f6\u4f7f\u7528\u3002\n\n### \u603b\u7ed3\n\n\u901a\u8fc7\u5408\u7406\u7ba1\u7406\u4e0a\u4e0b\u6587\u957f\u5ea6\uff0c\u53ef\u4ee5\u6709\u6548\u907f\u514d\u56e0\u4e0a\u4e0b\u6587\u8fc7\u957f\u5bfc\u81f4\u88ab\u622a\u65ad\u7684\u95ee\u9898\u3002\u8bf7\u6839\u636e\u5b9e\u9645\u7684 API \u6587\u6863\u548c\u884c\u4e3a\u9a8c\u8bc1\u5177\u4f53\u7ec6\u8282\uff0c\u5e76\u5fc5\u8981\u65f6\u8054\u7cfb\u6280\u672f\u652f\u6301\u83b7\u53d6\u5e2e\u52a9\u3002","behavior":"degenerate_repetition","conversation_hash":"44b305e7e7991de62b21af46e7a54afe","conversation_redacted":false,"conversation_toxic":false,"country":"United States","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Chinese","model":"gpt-4o-2024-05-13","n_chars":2127,"n_turns_in_conversation":16,"n_words":183,"source_row_offset":7405,"source_shard":"turns-00026.parquet","timestamp":"Sat, 27 Jul 2024 08:10:57 GMT","turn_id":"8d4098dda4f4be863b9294d9","turn_index":10,"user_text":"\u591a\u6b21\u8c03\u7528ollama api\u65f6\u4ea7\u751fINFO [update_slots] input truncated | n_ctx=4096 n_erase=4993 n_keep=4 n_left=4092 n_shift=2046 tid=\u201c125829096280064\u201d timestamp=1722057101\u4fe1\u606f\uff0c\u56e0\u4e3acontext\u662f\u6839\u636e\u4e0a\u6b21\u8c03\u7528\u8fd4\u56de\u7684\u503c\uff0c\u90a3\u4e48\u591a\u6b21\u8c03\u7528\u8fd4\u56de\u7684\u662f\u4e4b\u524d\u6240\u6709\u8c03\u7528\u7684\u7d2f\u52a0\uff0c\u8fd8\u662f\u53ea\u662f\u4e0a\u4e00\u6b21\u8c03\u7528\u6240\u4ea7\u751f\u7684\uff1f "},{"assistant_initiates":false,"asst_text":"`full_context.split()` \u662f\u4e00\u4e2a\u6807\u51c6\u7684 Python \u5b57\u7b26\u4e32\u65b9\u6cd5\uff0c\u7528\u4e8e\u5c06\u5b57\u7b26\u4e32\u5206\u5272\u6210\u5217\u8868\u3002\u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c`split()` \u65b9\u6cd5\u4f1a\u6839\u636e\u7a7a\u683c\u5c06\u5b57\u7b26\u4e32\u62c6\u5206\u6210\u591a\u4e2a\u5b50\u5b57\u7b26\u4e32\uff08\u5373\u201ctokens\u201d\uff09\u3002\u5728\u81ea\u7136\u8bed\u8a00\u5904\u7406\uff08NLP\uff09\u9886\u57df\uff0c\u6211\u4eec\u901a\u5e38\u5728\u8fd9\u79cd\u60c5\u51b5\u4e0b\u8c08\u8bba\u7684\u662f\u5c06\u53e5\u5b50\u5206\u5272\u6210\u5355\u8bcd\u6216\u4ee4\u724c\uff08tokens\uff09\u3002\n\n\u5728\u4f60\u7684\u4e0a\u4e0b\u6587\u4e2d\uff0c\u5b9e\u73b0 `tokens = full_context.split()` \u6709\u4ee5\u4e0b\u76ee\u7684\uff1a\n\n1. **\u5206\u5272\u5b57\u7b26\u4e32**\uff1a\u5c06 `full_context` \u5b57\u7b26\u4e32\u5206\u5272\u6210\u5355\u8bcd\u5217\u8868\uff08tokens\uff09\u3002\n2. **\u9002\u5e94\u4e0a\u4e0b\u6587\u957f\u5ea6\u9650\u5236**\uff1a\u8ba1\u7b97\u5f53\u524d\u4e0a\u4e0b\u6587\u4e2d\u5305\u542b\u7684\u4ee4\u724c\u6570\u91cf\uff0c\u4ee5\u4fbf\u4e4b\u540e\u8fdb\u884c\u957f\u5ea6\u7ba1\u7406\uff0c\u4f8b\u5982\u622a\u65ad\u6216\u8005\u8c03\u6574\u4e0a\u4e0b\u6587\u957f\u5ea6\u3002\n\n### \u793a\u4f8b\u89e3\u6790\n\n\u5047\u8bbe `full_context` \u662f\u4ee5\u4e0b\u5b57\u7b26\u4e32\uff1a\n```python\nfull_context = \"This is an example of context that could be too long for the model.\"\n```\n\u8c03\u7528 `full_context.split()` \u5c06\u4f1a\u6309\u7a7a\u683c\u5206\u5272\u6210\u4ee5\u4e0b\u5217\u8868\uff1a\n```python\ntokens = full_context.split()\n# tokens \u5217\u8868\u5185\u5bb9\u5c06\u662f\u4ee5\u4e0b\u5f62\u5f0f\uff1a\n# ['This', 'is', 'an', 'example', 'of', 'context', 'that', 'could', 'be', 'too', 'long', 'for', 'the', 'model.']\n```\n\n### \u8fdb\u4e00\u6b65\u5904\u7406\u793a\u4f8b\n\n\u4ee5\u4e0b\u662f\u5982\u4f55\u7ba1\u7406\u548c\u622a\u65ad\u4e0a\u4e0b\u6587\u7684\u5177\u4f53\u793a\u4f8b\u4ee3\u7801\uff0c\u5176\u4e2d\u5bf9 `tokens` \u8fdb\u884c\u5904\u7406\u4ee5\u786e\u4fdd\u957f\u5ea6\u4e0d\u8fc7\u957f\uff1a\n\n```python\n# \u5047\u8bbe\u6700\u5927\u4e0a\u4e0b\u6587\u957f\u5ea6\u4e3a 4096\nmax_length = 4096\n\n# \u7d2f\u52a0\u4e0a\u4e0b\u6587\nfull_context += \" \" + new_text\n\n# \u5c06\u4e0a\u4e0b\u6587\u5206\u5272\u6210 tokens \u5217\u8868\ntokens = full_context.split()\n\n# \u5982\u679c tokens \u7684\u6570\u91cf\u8d85\u8fc7\u4e86\u6700\u5927\u957f\u5ea6\nif len(tokens) > max_length:\n # \u4fdd\u7559\u6700\u540e\u7684 max_length \u4e2a tokens\n tokens = tokens[-max_length:]\n\n# \u91cd\u65b0\u7ec4\u5408 tokens \u5217\u8868\u4e3a\u5b57\u7b26\u4e32\nfull_context = ' '.join(tokens)\n```\n\n### \u603b\u4f53\u6d41\u7a0b\u89e3\u6790\n\n1. **\u7d2f\u52a0\u4e0a\u4e0b\u6587**\uff1a\u6bcf\u6b21\u65b0\u7684\u6587\u672c\u88ab\u6dfb\u52a0\u5230\u5f53\u524d\u4e0a\u4e0b\u6587\uff08`full_context`\uff09\u4e2d\u3002\n2. **\u5206\u5272\u6210 tokens**\uff1a`tokens = full_context.split()` \u5c06\u4e0a\u4e0b\u6587\u5206\u5272\u6210\u4e00\u4e2a\u4e2a\u5355\u8bcd\u6216\u4ee4\u724c\u3002\n3. **\u68c0\u67e5\u5e76\u8c03\u6574\u957f\u5ea6**\uff1a\u5982\u679c `tokens` \u5217\u8868\u7684\u957f\u5ea6\u8d85\u8fc7\u4e86\u6700\u5927\u4e0a\u4e0b\u6587\u957f\u5ea6\uff084096\uff09\uff0c\u5219\u4fdd\u7559\u6700\u540e\u7684 4096 \u4e2a tokens \u5e76\u820d\u5f03\u4e4b\u524d\u7684\u90e8\u5206\u3002\n4. **\u91cd\u65b0\u7ec4\u5408\u5b57\u7b26\u4e32**\uff1a\u4f7f\u7528 `' '.join(tokens)` \u5c06 tokens \u5217\u8868\u91cd\u65b0\u7ec4\u5408\u6210\u5b57\u7b26\u4e32\uff0c\u5f62\u6210\u65b0\u7684 `full_context`\u3002\n\n### \u5c0f\u7ed3\n\n\u901a\u8fc7\u4e0a\u8ff0\u65b9\u6cd5\uff0c\u53ef\u4ee5\u6709\u6548\u7ba1\u7406\u7d2f\u79ef\u7684\u4e0a\u4e0b\u6587\u957f\u5ea6\uff0c\u786e\u4fdd\u5176\u59cb\u7ec8\u5728\u6a21\u578b\u5141\u8bb8\u7684\u8303\u56f4\u5185\uff0c\u907f\u514d\u622a\u65ad\u4fe1\u606f\u4e0e\u6f5c\u5728\u9519\u8bef\u3002\u8fd9\u6837\u7684\u505a\u6cd5\u5728\u5904\u7406\u957f\u6587\u672c\u8f93\u5165\u65f6\u5c24\u5176\u6709\u7528\uff0c\u56e0\u4e3a\u5b83\u786e\u4fdd\u4e86\u4f60\u6bcf\u6b21\u8c03\u7528\u6a21\u578b\u65f6\u63d0\u4f9b\u7684\u4e0a\u4e0b\u6587\u90fd\u662f\u6700\u65b0\u4e14\u6709\u6548\u7684\u3002","behavior":"degenerate_repetition","conversation_hash":"44b305e7e7991de62b21af46e7a54afe","conversation_redacted":false,"conversation_toxic":false,"country":"United States","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Chinese","model":"gpt-4o-2024-05-13","n_chars":1476,"n_turns_in_conversation":16,"n_words":149,"source_row_offset":7406,"source_shard":"turns-00026.parquet","timestamp":"Sat, 27 Jul 2024 08:10:57 GMT","turn_id":"78dec4352398fc24c9d42a79","turn_index":11,"user_text":"tokens = full_context.split()\u8fd9\u4e2a\u51fd\u6570\u662f\u4ec0\u4e48\u610f\u601d\uff1f"},{"assistant_initiates":false,"asst_text":"\u5728 Python \u4e2d\uff0c\u5982\u679c\u4f60\u6709\u4e00\u4e2a\u5217\u8868\uff0c\u5e76\u4e14\u60f3\u8981\u8fd4\u56de\u8be5\u5217\u8868\u7684\u6700\u540e4096\u4e2a\u5143\u7d20\uff0c\u53ef\u4ee5\u4f7f\u7528\u5207\u7247\uff08slicing\uff09\u64cd\u4f5c\u3002\u8fd9\u662f\u4e00\u4e2a\u975e\u5e38\u65b9\u4fbf\u548c\u9ad8\u6548\u7684\u65b9\u6cd5\uff0c\u53ef\u4ee5\u8f7b\u677e\u5730\u4ece\u5217\u8868\u4e2d\u63d0\u53d6\u6240\u9700\u7684\u90e8\u5206\u3002\n\n### \u793a\u4f8b\u4ee3\u7801\n\n\u5047\u8bbe\u4f60\u7684\u5217\u8868\u540d\u5b57\u662f `my_list`\uff0c\u4f60\u53ef\u4ee5\u7528\u4e0b\u9762\u7684\u4ee3\u7801\u6765\u83b7\u53d6\u6700\u540e4096\u4e2a\u5143\u7d20\uff1a\n\n```python\n# \u5047\u8bbe my_list \u662f\u4f60\u7684\u5217\u8868\nmy_list = [0, 1, 2, ..., 'n'] # \u793a\u4f8b\u5217\u8868\n\n# \u83b7\u53d6\u6700\u540e4096\u4e2a\u5143\u7d20\nlast_4096_elements = my_list[-4096:]\n```\n\n### \u8be6\u7ec6\u89e3\u91ca\n\n1. **my_list[-4096:]**\uff1a\n - `-4096` \u8fd9\u4e2a\u8d1f\u7d22\u5f15\u8868\u793a\u4ece\u5217\u8868\u7684\u5012\u6570\u7b2c4096\u4e2a\u5143\u7d20\u5f00\u59cb\u3002\n - `:` \u540e\u9762\u6ca1\u6709\u5199\u4efb\u4f55\u4e1c\u897f\uff0c\u8fd9\u610f\u5473\u7740\u5207\u7247\u5c06\u4e00\u76f4\u8fdb\u884c\u5230\u5217\u8868\u7684\u6700\u540e\u4e00\u4e2a\u5143\u7d20\u3002\n\n\u5f53\u5217\u8868\u957f\u5ea6\u5c0f\u4e8e 4096 \u65f6\uff0c\u8fd9\u4e2a\u5207\u7247\u5c06\u8fd4\u56de\u6574\u4e2a\u5217\u8868\u800c\u4e0d\u4f1a\u5f15\u53d1\u9519\u8bef\u3002\n\n### \u4e3e\u4f8b\n\n\u8ba9\u6211\u4eec\u770b\u4e00\u4e2a\u5177\u4f53\u7684\u4f8b\u5b50\uff1a\n\n```python\n# \u521b\u5efa\u4e00\u4e2a\u793a\u4f8b\u5217\u8868\uff0c\u5047\u8bbe\u67095000\u4e2a\u5143\u7d20\nmy_list = list(range(5000))\n\n# \u83b7\u53d6\u6700\u540e4096\u4e2a\u5143\u7d20\nlast_4096_elements = my_list[-4096:]\n\nprint(len(last_4096_elements)) # \u8f93\u51fa\uff1a4096\nprint(last_4096_elements) # \u8f93\u51fa\uff1a\u6700\u540e4096\u4e2a\u5143\u7d20\n```\n\n\u5982\u679c `my_list` \u7684\u957f\u5ea6\u5c11\u4e8e 4096 \u4e2a\u5143\u7d20\uff0c\u4f8b\u5982 3000 \u4e2a\u5143\u7d20\uff1a\n\n```python\n# \u521b\u5efa\u4e00\u4e2a\u793a\u4f8b\u5217\u8868\uff0c\u5047\u8bbe\u67093000\u4e2a\u5143\u7d20\nmy_list = list(range(3000))\n\n# \u83b7\u53d6\u6700\u540e4096\u4e2a\u5143\u7d20\uff08\u5b9e\u9645\u4e0a\u662f\u5168\u90e83000\u4e2a\u5143\u7d20\uff09\nlast_4096_elements = my_list[-4096:]\n\nprint(len(last_4096_elements)) # \u8f93\u51fa\uff1a3000\nprint(last_4096_elements) # \u8f93\u51fa\uff1a\u6574\u4e2a\u5217\u8868\uff0c\u56e0\u4e3a\u603b\u957f\u5ea6\u5c0f\u4e8e4096\n```\n\n### \u5728\u5b9e\u9645\u4ee3\u7801\u4e2d\u7684\u5e94\u7528\n\n\u5047\u8bbe\u4f60\u6b63\u5728\u5904\u7406\u4e0a\u4e0b\u6587\u7d2f\u79ef\u5e76\u9700\u8981\u786e\u4fdd\u5176\u957f\u5ea6\u4e0d\u8d85\u8fc74096\u4e2a\u5143\u7d20\uff0c\u53ef\u4ee5\u8fd9\u6837\u7ed3\u5408\u5207\u7247\u64cd\u4f5c\uff1a\n\n```python\n# \u5b9a\u4e49\u4e00\u4e2a\u51fd\u6570\u4ee5\u786e\u4fdd\u5217\u8868\u6700\u5927\u957f\u5ea6\u4e3a4096\ndef ensure_max_length(lst, max_length=4096):\n return lst[-max_length:]\n\n# \u521d\u59cb\u5316\u5217\u8868\nfull_context = []\n\n# \u793a\u4f8b\uff1a\u591a\u6b21\u8ffd\u52a0\u5185\u5bb9\u5e76\u4fdd\u6301\u6700\u5927\u957f\u5ea6\u4e0d\u8d85\u8fc74096\nnew_data_list = [\"some new data\"] * 5000 # \u793a\u4f8b\u65b0\u6570\u636e\nfor new_data in new_data_list:\n full_context.append(new_data)\n full_context = ensure_max_length(full_context)\n\nprint(len(full_context)) # \u5c06\u4f1a\u8f93\u51fa\uff1a4096\n```\n\n\u5728\u8fd9\u4e2a\u793a\u4f8b\u4e2d\uff0c\u65e0\u8bba `new_data_list` \u4e2d\u6709\u591a\u5c11\u6570\u636e\uff0c`full_context` \u6700\u7ec8\u90fd\u4f1a\u9650\u5236\u5728\u6700\u540e4096\u4e2a\u5143\u7d20\u3002\n\n\u901a\u8fc7\u8fd9\u79cd\u65b9\u6cd5\uff0c\u4f60\u53ef\u4ee5\u5728\u5904\u7406\u72b6\u6001\u6216\u4e0a\u4e0b\u6587\u65f6\u6709\u6548\u5730\u63a7\u5236\u5217\u8868\u7684\u957f\u5ea6\uff0c\u907f\u514d\u6570\u636e\u8fc7\u591a\u5bfc\u81f4\u8d85\u9650\u95ee\u9898\u3002","behavior":"degenerate_repetition","conversation_hash":"44b305e7e7991de62b21af46e7a54afe","conversation_redacted":false,"conversation_toxic":false,"country":"United States","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Chinese","model":"gpt-4o-2024-05-13","n_chars":1486,"n_turns_in_conversation":16,"n_words":131,"source_row_offset":7407,"source_shard":"turns-00026.parquet","timestamp":"Sat, 27 Jul 2024 08:10:57 GMT","turn_id":"63e29e0c5b15ade6bf3839a7","turn_index":12,"user_text":"\u5982\u679c\u4e00\u4e2a\u53d8\u91cf\u662flist\uff0c\u5982\u4f55\u8fd4\u56delist\u7684\u6700\u540e4096\u4e2a\u5143\u7d20\uff1f"},{"assistant_initiates":false,"asst_text":"Tesla P100 \u8ba1\u7b97\u5361\u7684 CPU \u8d1f\u8377\u5f88\u9ad8\u4f46\u529f\u8017\u8f83\u4f4e\uff08\u5982\u53ea\u6709\u7ea6 130W\uff0c\u5c3d\u7ba1\u4e0a\u9650\u662f 250W\uff09\uff0c\u8fd9\u79cd\u60c5\u51b5\u53ef\u80fd\u51fa\u4e8e\u4ee5\u4e0b\u51e0\u79cd\u539f\u56e0\uff1a\n\n### 1. \u6570\u636e\u4f20\u8f93\u74f6\u9888\n\u5982\u679c CPU \u548c GPU \u4e4b\u95f4\u7684\u6570\u636e\u4f20\u8f93\u901f\u5ea6\u8f83\u6162\uff0c\u90a3\u4e48 GPU \u53ef\u80fd\u4f1a\u9891\u7e41\u7b49\u5f85\u6570\u636e\u4f20\u8f93\u5b8c\u6210\uff0c\u5bfc\u81f4 GPU \u4f7f\u7528\u7387\u4e0d\u9ad8\uff0c\u529f\u8017\u8f83\u4f4e\uff0c\u800c CPU \u8d1f\u8377\u5374\u5f88\u9ad8\u3002\u5178\u578b\u7684\u4f8b\u5b50\u662f\uff1a\n- **PCIe \u5e26\u5bbd**\uff1aPCIe \u5e26\u5bbd\u4e0d\u591f\u9ad8\uff0c\u6570\u636e\u4f20\u8f93\u53d8\u6162\u3002\n- **Memory Copy**\uff1a\u6570\u636e\u5728\u4e3b\u673a\u5185\u5b58\uff08CPU\uff09\u548c\u8bbe\u5907\u5185\u5b58\uff08GPU\uff09\u4e4b\u95f4\u9891\u7e41\u62f7\u8d1d\uff0c\u5bfc\u81f4 CPU \u8d1f\u8377\u9ad8\u3002\n\n### 2. \u8ba1\u7b97\u4efb\u52a1\u7279\u70b9\n\u67d0\u4e9b\u8ba1\u7b97\u4efb\u52a1\u53ef\u80fd\u5e76\u4e0d\u9002\u5408\u5728 GPU \u4e0a\u9ad8\u5ea6\u5e76\u884c\u5316\uff0c\u6216\u8005 GPU \u5229\u7528\u7387\u4e0d\u9ad8\u3002\u4f8b\u5982\uff1a\n- **CPU-bound \u4efb\u52a1**\uff1a\u67d0\u4e9b\u64cd\u4f5c\u4e3b\u8981\u4f9d\u8d56 CPU \u8ba1\u7b97\uff0c\u800c\u975e\u5e76\u884c\u7684 GPU \u8ba1\u7b97\u3002\n- **\u4f4e\u5e76\u884c\u5ea6\u4efb\u52a1**\uff1a\u4efb\u52a1\u5e76\u884c\u5ea6\u8f83\u4f4e\uff0cGPU \u8d44\u6e90\u672a\u88ab\u5145\u5206\u5229\u7528\u3002\n\n### 3. \u4ee3\u7801\u5b9e\u73b0\u95ee\u9898\n\u4ee3\u7801\u5728\u5b9e\u73b0\u65f6\u6ca1\u6709\u5145\u5206\u5229\u7528 GPU \u7684\u6027\u80fd\uff0c\u53ef\u80fd\u6709\u4ee5\u4e0b\u95ee\u9898\uff1a\n- **\u4e0d\u5145\u5206\u7684\u5e76\u884c\u5316**\uff1a\u672a\u80fd\u5c06\u4efb\u52a1\u5145\u5206\u5e76\u884c\u5316\uff0c\u5bfc\u81f4 GPU \u5927\u91cf\u7a7a\u95f2\u3002\n- **\u5185\u5b58\u5206\u914d**\uff1a\u9891\u7e41\u7684\u5185\u5b58\u5206\u914d\u548c\u91ca\u653e\uff0c\u5bfc\u81f4 GPU \u6027\u80fd\u672a\u80fd\u5145\u5206\u53d1\u6325\u3002\n- **\u540c\u6b65\u64cd\u4f5c**\uff1a\u9891\u7e41\u7684 CPU-GPU \u540c\u6b65\u64cd\u4f5c\uff0c\u4f7f\u5f97 GPU \u5728\u7b49\u5f85 CPU \u547d\u4ee4\uff0c\u5bfc\u81f4 GPU \u5229\u7528\u7387\u4f4e\u3002\n\n### 4. GPU \u9650\u5236\u6216\u8282\u80fd\u6a21\u5f0f\nGPU \u53ef\u80fd\u88ab\u914d\u7f6e\u5728\u67d0\u79cd\u8282\u80fd\u6a21\u5f0f\u6216\u529f\u8017\u9650\u5236\u6a21\u5f0f\u4e0b\uff1a\n- **\u529f\u8017\u9650\u5236\uff08Power Cap\uff09**\uff1aGPU \u529f\u8017\u88ab\u9650\u5236\u5728\u4f4e\u4e8e\u6700\u5927\u503c\u7684\u67d0\u4e2a\u6c34\u5e73\uff0c\u4ee5\u51cf\u5c11\u529f\u8017\u6216\u53d1\u70ed\u3002\n- **\u8282\u80fd\u6a21\u5f0f**\uff1aGPU \u5728\u4f4e\u8d1f\u8377\u4e0b\u81ea\u52a8\u8fdb\u5165\u67d0\u79cd\u8282\u80fd\u6a21\u5f0f\uff0c\u964d\u4f4e\u529f\u8017\u3002\n\n### \u8bca\u65ad\u4e0e\u89e3\u51b3\u65b9\u6848\n\n#### 1. **\u76d1\u63a7\u6570\u636e\u4f20\u8f93**\n\u4f7f\u7528\u50cf `nvidia-smi` \u548c\u5176\u4ed6 GPU \u76d1\u63a7\u5de5\u5177\u6765\u76d1\u63a7 GPU \u7684\u6570\u636e\u4f20\u8f93\u548c\u5229\u7528\u7387\uff0c\u68c0\u67e5 PCIe \u5e26\u5bbd\uff0c\u5185\u5b58\u4f7f\u7528\u60c5\u51b5\u7b49\u3002\n```sh\nnvidia-smi\n```\n\n#### 2. **\u4f18\u5316\u4ee3\u7801\u5e76\u884c\u6027**\n\u786e\u4fdd\u4ee3\u7801\u5df2\u5145\u5206\u5e76\u884c\u5316\uff0c\u5c3d\u91cf\u51cf\u5c11 CPU-GPU \u540c\u6b65\u64cd\u4f5c\uff0c\u786e\u4fdd\u4efb\u52a1\u9002\u5408\u5e76\u884c\u5904\u7406\u3002\u67e5\u770b\u548c\u4f7f\u7528 CUDA Profiling \u5de5\u5177\uff08\u4f8b\u5982 NVIDIA Nsight\uff09\u6765\u5206\u6790\u548c\u4f18\u5316\u4ee3\u7801\u3002\n\n#### 3. **\u907f\u514d\u6570\u636e\u4f20\u8f93\u74f6\u9888**\n\u5c3d\u91cf\u51cf\u5c11\u6570\u636e\u5728 CPU \u548c GPU \u4e4b\u95f4\u7684\u4f20\u8f93\u6b21\u6570\uff0c\u5c3d\u91cf\u5c06\u6570\u636e\u4fdd\u5b58\u5728 GPU \u4e0a\u8fdb\u884c\u8ba1\u7b97\u3002\n\n#### 4. **\u68c0\u67e5\u5e76\u8c03\u6574 GPU \u8bbe\u7f6e**\n\u68c0\u67e5 GPU \u662f\u5426\u6709\u529f\u8017\u9650\u5236\u6216\u5904\u4e8e\u8282\u80fd\u6a21\u5f0f\uff0c\u6839\u636e\u9700\u8981\u8fdb\u884c\u8c03\u6574\u3002\n```sh\n# \u89e3\u9664\u529f\u8017\u9650\u5236\u793a\u4f8b\nnvidia-smi -pm ENABLED\nnvidia-smi -pl 250 # \u5c06\u529f\u8017\u4e0a\u9650\u8bbe\u7f6e\u4e3a 250W\n```\n\n\u901a\u8fc7\u5408\u7406\u7684\u4ee3\u7801\u4f18\u5316\u548c\u786c\u4ef6\u8bbe\u7f6e\uff0c\u53ef\u4ee5\u66f4\u6709\u6548\u5730\u5229\u7528 Tesla P100 \u8ba1\u7b97\u5361\u7684\u8ba1\u7b97\u80fd\u529b\uff0c\u63d0\u9ad8\u5176\u529f\u8017\u5229\u7528\u7387\u3002","behavior":"degenerate_repetition","conversation_hash":"44b305e7e7991de62b21af46e7a54afe","conversation_redacted":false,"conversation_toxic":false,"country":"United States","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Chinese","model":"gpt-4o-2024-05-13","n_chars":1263,"n_turns_in_conversation":16,"n_words":151,"source_row_offset":7408,"source_shard":"turns-00026.parquet","timestamp":"Sat, 27 Jul 2024 08:10:57 GMT","turn_id":"9ba208df0101636b78fb7130","turn_index":13,"user_text":"\u4e3a\u4ec0\u4e48 telsa P100\u8ba1\u7b97\u5361\u7684CPU\u8d1f\u8377\u5f88\u9ad8\uff0c\u4f46\u662f\u529f\u8017\u5374\u4e0d\u9ad8\uff0c\u6700\u9ad8\u53ea\u6709130w\u5de6\u53f3\uff0c\u4e0a\u9650\u662f250w"},{"assistant_initiates":false,"asst_text":"\u5982\u679c Tesla P100 GPU \u7684\u8d1f\u8377\u5f88\u9ad8\uff08\u4f8b\u5982\u901a\u8fc7 `nvidia-smi` \u67e5\u770b\uff0c\u5229\u7528\u7387\u7ecf\u5e38\u8d85\u8fc7 95%\uff09\uff0c\u4f46\u662f\u529f\u8017\u76f8\u5bf9\u8f83\u4f4e\uff08\u4f8b\u5982\u53ea\u6709\u7ea6 130W\uff0c\u5c3d\u7ba1\u4e0a\u9650\u662f 250W\uff09\uff0c\u53ef\u80fd\u6709\u51e0\u79cd\u60c5\u51b5\u5bfc\u81f4\u8fd9\u79cd\u73b0\u8c61\u3002\u4ee5\u4e0b\u662f\u53ef\u80fd\u7684\u539f\u56e0\u53ca\u8bca\u65ad\u65b9\u6cd5\uff1a\n\n### 1. \u5de5\u4f5c\u8d1f\u8f7d\u7279\u6027\n\u67d0\u4e9b\u8ba1\u7b97\u4efb\u52a1\u867d\u7136\u4f7f\u5f97 GPU \u5229\u7528\u7387\u5f88\u9ad8\uff0c\u4f46\u5e76\u4e0d\u9700\u8981\u5927\u91cf\u529f\u8017\u3002\u8fd9\u4e9b\u4efb\u52a1\u53ef\u80fd\u662f\uff1a\n- **\u5185\u5b58\u5bc6\u96c6\u578b\u4efb\u52a1**\uff1a\u8fd9\u4e9b\u4efb\u52a1\u4e3b\u8981\u6d89\u53ca\u5927\u91cf\u7684\u5185\u5b58\u8bbf\u95ee\uff0c\u800c\u4e0d\u662f\u5927\u91cf\u7684\u8ba1\u7b97\uff0c\u56e0\u800c\u529f\u8017\u8f83\u4f4e\u3002\n- **\u4f4e\u8fd0\u7b97\u590d\u6742\u5ea6**\uff1a\u67d0\u4e9b\u7b80\u5355\u7684\u8ba1\u7b97\u64cd\u4f5c\uff0c\u5373\u4f7f\u9ad8\u5229\u7528\u7387\u4f46\u4e5f\u4e0d\u9700\u8981\u5f88\u591a\u529f\u8017\u3002\n\n### 2. \u529f\u8017\u9650\u5236\u6216\u8282\u80fd\u6a21\u5f0f\nGPU \u53ef\u80fd\u88ab\u914d\u7f6e\u5728\u67d0\u79cd\u8282\u80fd\u6a21\u5f0f\u6216\u529f\u8017\u9650\u5236\u6a21\u5f0f\u4e0b\uff1a\n- **\u529f\u8017\u9650\u5236\uff08Power Cap\uff09\u8bbe\u7f6e**\uff1a\u68c0\u67e5\u662f\u5426\u6709\u8bbe\u7f6e\u9650\u5236\u4e86 GPU \u7684\u529f\u8017\u3002\n- **\u81ea\u52a8\u8282\u80fd\u6a21\u5f0f**\uff1aGPU \u53ef\u80fd\u52a8\u6001\u8c03\u6574\u9891\u7387\u548c\u529f\u8017\uff0c\u4ee5\u9002\u5e94\u8d1f\u8f7d\u9700\u6c42\u3002\n\n### 3. \u6563\u70ed\u548c\u6e29\u5ea6\u9650\u5236\nGPU \u7684\u529f\u8017\u548c\u6027\u80fd\u53ef\u80fd\u53d7\u5230\u6e29\u5ea6\u7ba1\u7406\u7b56\u7565\u7684\u5f71\u54cd\uff1a\n- **\u6e29\u5ea6\u8282\u6d41\uff08Thermal Throttling\uff09**\uff1a\u5f53 GPU \u6e29\u5ea6\u63a5\u8fd1\u5176\u4e0a\u9650\u65f6\uff0c\u53ef\u80fd\u4f1a\u81ea\u52a8\u964d\u4f4e\u9891\u7387\u4ee5\u63a7\u5236\u6e29\u5ea6\uff0c\u8fd9\u4f1a\u964d\u4f4e\u529f\u8017\u3002\n\n### 4. \u7535\u6e90\u7ba1\u7406\u7b56\u7565\n\u6709\u65f6 GPU \u4f1a\u6839\u636e\u7cfb\u7edf\u7535\u6e90\u7b56\u7565\u81ea\u52a8\u8c03\u8282\u529f\u8017\u3002\n\n### \u8bca\u65ad\u4e0e\u89e3\u51b3\u65b9\u6848\n\n#### 1. \u68c0\u67e5\u529f\u8017\u8bbe\u7f6e\n\u4f7f\u7528 `nvidia-smi` \u68c0\u67e5\u5e76\u8c03\u6574\u529f\u8017\u8bbe\u7f6e\uff1a\n```sh\n# \u68c0\u67e5\u5f53\u524d\u529f\u8017\u9650\u5236\nnvidia-smi -q | grep \"Power Management\"\nnvidia-smi -q | grep \"Power Draw\"\n\n# \u89e3\u9664\u6216\u8c03\u6574\u529f\u8017\u9650\u5236\n# \u542f\u7528\u6301\u4e45\u6a21\u5f0f\uff0c\u786e\u4fdd\u529f\u8017\u8bbe\u7f6e\u80fd\u751f\u6548\nsudo nvidia-smi -pm ENABLED\n\n# \u8c03\u6574\u529f\u8017\u4e0a\u9650\u4e3a 250W\uff08\u5177\u4f53\u503c\u6839\u636e\u578b\u53f7\u548c\u786c\u4ef6\u89c4\u683c\uff09\nsudo nvidia-smi -pl 250\n```\n\n#### 2. \u76d1\u63a7\u6e29\u5ea6\u548c\u9891\u7387\n\u901a\u8fc7 `nvidia-smi` \u67e5\u770b GPU \u7684\u6e29\u5ea6\u548c\u9891\u7387\uff0c\u4ee5\u786e\u5b9a\u662f\u5426\u53d7\u6e29\u5ea6\u8282\u6d41\u5f71\u54cd\uff1a\n```sh\nnvidia-smi -q -d TEMPERATURE,CLOCK\n```\n\u67e5\u770b\u6e29\u5ea6\u4fe1\u606f\u548c\u65f6\u949f\u9891\u7387\uff0c\u5982\u679c\u6e29\u5ea6\u63a5\u8fd1\u4e0a\u9650\uff0c\u53ef\u80fd\u4f1a\u5f71\u54cd\u529f\u8017\u4f7f\u7528\u60c5\u51b5\u3002\n\n#### 3. \u68c0\u67e5\u4efb\u52a1\u7c7b\u578b\n\u5206\u6790\u4f60\u7684\u8ba1\u7b97\u4efb\u52a1\uff0c\u770b\u770b\u662f\u5426\u4e3b\u8981\u4ee5\u5185\u5b58\u8bbf\u95ee\u4e3a\u4e3b\uff0c\u800c\u4e0d\u662f\u8ba1\u7b97\u5bc6\u96c6\u578b\u5de5\u4f5c\u8d1f\u8f7d\u3002\u53ef\u4ee5\u4f7f\u7528 NVIDIA Nsight \u7b49\u5de5\u5177\u8fdb\u884c\u8be6\u7ec6\u7684\u6027\u80fd\u5256\u6790\uff1a\n```sh\nnvprof your_application\n```\n\u6216\u8005\u4f7f\u7528 Nsight Compute \u548c Nsight Systems \u8fdb\u884c\u66f4\u8be6\u7ec6\u7684\u5206\u6790\u3002\n\n### \u8bbe\u5b9a\u6301\u4e45\u5316\u6a21\u5f0f\u548c\u6027\u80fd\u6a21\u5f0f\n\n\u5bf9\u4e8e\u5b9e\u9a8c\u6027\u548c\u4e34\u65f6\u6027\u7684\u8c03\u8bd5\uff0c\u4f60\u53ef\u4ee5\u5c06 GPU \u8bbe\u4e3a\u6700\u5927\u6027\u80fd\u6a21\u5f0f\uff0c\u4ee5\u786e\u4fdd\u5176\u4e0d\u8fdb\u884c\u529f\u8017\u8282\u80fd\u64cd\u4f5c\uff1a\n\n```sh\n# \u542f\u7528\u6301\u4e45\u5316\u6a21\u5f0f\nsudo nvidia-smi -pm ENABLED\n\n# \u5c06 GPU \u8bbe\u7f6e\u4e3a\u6301\u4e45\u7684\u6700\u9ad8\u6027\u80fd\u6a21\u5f0f\nsudo nvidia-smi -lgc 0,0\n```\n\n### \u6700\u540e\u603b\u7ed3\n\u6839\u636e\u4f60\u7684 GPU \u9ad8\u8d1f\u8377\u4f46\u4f4e\u529f\u8017\u73b0\u8c61\uff0c\u6700\u53ef\u80fd\u7684\u539f\u56e0\u662f\u8ba1\u7b97\u4efb\u52a1\u7279\u70b9\uff08\u5982\u5185\u5b58\u5bc6\u96c6\u578b\u4efb\u52a1\uff09\u3001\u529f\u8017\u9650\u5236\u8bbe\u7f6e\u6216\u6e29\u5ea6\u8282\u6d41\u5f15\u8d77\u7684\u6027\u80fd\u9650\u5236\u3002\u901a\u8fc7 `nvidia-smi` \u6765\u68c0\u67e5\u548c\u8c03\u6574\u529f\u8017\u9650\u5236\u3001\u76d1\u63a7\u6e29\u5ea6\u548c\u9891\u7387\u3001\u5206\u6790\u4efb\u52a1\u7279\u6027\uff0c\u80fd\u591f\u5e2e\u52a9\u4f60\u66f4\u597d\u5730\u7406\u89e3\u548c\u4f18\u5316 GPU \u7684\u5229\u7528\u60c5\u51b5\u3002\n\n\u8fdb\u4e00\u6b65\u7684\u8be6\u7ec6\u6027\u80fd\u5206\u6790\u53ef\u80fd\u9700\u8981\u7528\u5230\u5982 Nsight \u7cfb\u5217\u5de5\u5177\uff0c\u4ee5\u4fbf\u66f4\u7cbe\u786e\u5730\u4e86\u89e3\u8d1f\u8f7d\u548c\u529f\u8017\u7279\u6027\u3002","behavior":"degenerate_repetition","conversation_hash":"44b305e7e7991de62b21af46e7a54afe","conversation_redacted":false,"conversation_toxic":false,"country":"United States","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Chinese","model":"gpt-4o-2024-05-13","n_chars":1546,"n_turns_in_conversation":16,"n_words":157,"source_row_offset":7409,"source_shard":"turns-00026.parquet","timestamp":"Sat, 27 Jul 2024 08:10:57 GMT","turn_id":"7553157b7037cea6f32e7887","turn_index":14,"user_text":"\u6211\u8bf4\u7684\u6709\u95ee\u9898\uff0c\u5e94\u8be5\u662f\u4e3a\u4ec0\u4e48 telsa P100\u8ba1\u7b97\u5361\u7684GPU\u8d1f\u8377\u5f88\u9ad8\uff08\u7528nvidia_smi\u67e5\u770b\u65f6GPU\u8d1f\u8377\u7ecf\u5e38\u8d85\u8fc795%\uff09\uff0c\u4f46\u662f\u529f\u8017\u5374\u4e0d\u9ad8\uff0c\u6700\u9ad8\u53ea\u6709130w\u5de6\u53f3\uff0c\u4e0a\u9650\u662f250w "},{"assistant_initiates":false,"asst_text":"\u5bf9\u4e8e Ollama \u662f\u5426\u652f\u6301\u591a\u7ebf\u7a0b\u8fd0\u884c\u7684\u95ee\u9898\uff0c\u5177\u4f53\u7b54\u6848\u9700\u8981\u4f9d\u636e\u5176\u5b98\u65b9\u6587\u6863\u6216\u652f\u6301\u6e20\u9053\u63d0\u4f9b\u7684\u7ec6\u8282\u3002\u7136\u800c\uff0c\u4ece\u5e38\u89c1\u7684\u673a\u5668\u5b66\u4e60\u670d\u52a1\u548c API \u7684\u89d2\u5ea6\u6765\u8bf4\uff0c\u591a\u7ebf\u7a0b\u8fd0\u884c\u6709\u5982\u4e0b\u51e0\u79cd\u53ef\u80fd\u6027\uff1a\n\n### 1. \u5ba2\u6237\u7aef\u4ee3\u7801\u591a\u7ebf\u7a0b\n\n\u5982\u679c\u4f60\u6709\u591a\u4e2a\u72ec\u7acb\u7684\u8bf7\u6c42\u9700\u8981\u5e76\u884c\u5904\u7406\uff0c\u4f60\u53ef\u4ee5\u5728\u5ba2\u6237\u7aef\u4ee3\u7801\u4e2d\u4f7f\u7528\u591a\u7ebf\u7a0b\u6216\u591a\u8fdb\u7a0b\u8fdb\u884c\u5e76\u884c\u8c03\u7528\u3002\u4f8b\u5982\uff0c\u4f7f\u7528 Python \u7684 `threading` \u6216 `concurrent.futures` \u6a21\u5757\u3002\n\n#### \u4f7f\u7528 `threading` \u6a21\u5757\n\n\u4ee5\u4e0b\u662f\u4e00\u4e2a\u4f7f\u7528 `threading` \u5e76\u884c\u5904\u7406\u591a\u4e2a API \u8bf7\u6c42\u7684\u793a\u4f8b\uff1a\n\n```python\nimport threading\nimport requests\n\nAPI_URL = \"https://api-inference.huggingface.co/models/your-api-model\"\nAPI_KEY = \"YOUR_API_KEY\"\n\nheaders = {\n \"Authorization\": f\"Bearer {API_KEY}\"\n}\n\ndef query(payload):\n response = requests.post(API_URL, headers=headers, json=payload)\n print(response.json())\n\ninput_texts = [\"Your first input text here\", \"Your second input text here\", \"Your third input text here\"]\n\nthreads = []\nfor text in input_texts:\n payload = {\"inputs\": text}\n thread = threading.Thread(target=query, args=(payload,))\n thread.start()\n threads.append(thread)\n\n# \u7b49\u5f85\u6240\u6709\u7ebf\u7a0b\u5b8c\u6210\nfor thread in threads:\n thread.join()\n```\n\n#### \u4f7f\u7528 `concurrent.futures` \u6a21\u5757\n\n\u4ee5\u4e0b\u662f\u4e00\u4e2a\u4f7f\u7528 `concurrent.futures` \u6a21\u5757\u7684\u793a\u4f8b\uff1a\n\n```python\nimport concurrent.futures\nimport requests\n\nAPI_URL = \"https://api-inference.huggingface.co/models/your-api-model\"\nAPI_KEY = \"YOUR_API_KEY\"\n\nheaders = {\n \"Authorization\": f\"Bearer {API_KEY}\"\n}\n\ndef query(payload):\n response = requests.post(API_URL, headers=headers, json=payload)\n return response.json()\n\ninput_texts = [\"Your first input text here\", \"Your second input text here\", \"Your third input text here\"]\n\nwith concurrent.futures.ThreadPoolExecutor() as executor:\n futures = [executor.submit(query, {\"inputs\": text}) for text in input_texts]\n for future in concurrent.futures.as_completed(futures):\n print(future.result())\n```\n\n### 2. \u670d\u52a1\u5668\u7aef\u5e76\u884c\u5316\n\n\u5982\u679c Ollama API \u652f\u6301\u5728\u670d\u52a1\u5668\u7aef\u8fdb\u884c\u591a\u7ebf\u7a0b\u6216\u5e76\u884c\u5904\u7406\uff0c\u4f60\u53ef\u4ee5\u914d\u7f6e\u670d\u52a1\u5668\uff08\u5982\u679c\u4f60\u6709\u7ba1\u7406\u6743\u9650\uff09\u6765\u5229\u7528\u591a\u7ebf\u7a0b\u6765\u5904\u7406\u8bf7\u6c42\u3002\u4e0d\u8fc7\uff0c\u8fd9\u7c7b\u914d\u7f6e\u901a\u5e38\u9700\u8981\u670d\u52a1\u5668\u7aef\u7684\u5177\u4f53\u6587\u6863\u548c\u652f\u6301\u3002\n\n### 3. \u4f7f\u7528\u5f02\u6b65\u7f16\u7a0b\n\n\u5bf9\u4e8e Python\uff0c\u4f60\u8fd8\u53ef\u4ee5\u4f7f\u7528 `asyncio` \u6a21\u5757\u8fdb\u884c\u5f02\u6b65\u7f16\u7a0b\u6765\u5904\u7406\u5e76\u53d1\u8bf7\u6c42\uff1a\n\n```python\nimport aiohttp\nimport asyncio\n\nAPI_URL = \"https://api-inference.huggingface.co/models/your-api-model\"\nAPI_KEY = \"YOUR_API_KEY\"\n\nheaders = {\n \"Authorization\": f\"Bearer {API_KEY}\"\n}\n\nasync def query(session, payload):\n async with session.post(API_URL, json=payload, headers=headers) as response:\n return await response.json()\n\nasync def main():\n input_texts = [\"Your first input text here\", \"Your second input text here\", \"Your third input text here\"]\n async with aiohttp.ClientSession() as session:\n tasks = [query(session, {\"inputs\": text}) for text in input_texts]\n results = await asyncio.gather(*tasks)\n for result in results:\n print(result)\n\n# \u8fd0\u884c\u4e3b\u51fd\u6570\nasyncio.run(main())\n```\n\n### \u786e\u8ba4 API \u652f\u6301\n\n\u8981\u4e86\u89e3 Ollama \u662f\u5426\u652f\u6301\u548c\u5982\u4f55\u652f\u6301\u591a\u7ebf\u7a0b\u5904\u7406\uff0c\u5efa\u8bae\u67e5\u9605\u4ee5\u4e0b\u8d44\u6e90\uff1a\n\n1. **\u5b98\u65b9\u6587\u6863**\uff1a\u901a\u5e38 API \u6587\u6863\u4f1a\u8bf4\u660e\u662f\u5426\u652f\u6301\u591a\u7ebf\u7a0b\u6216\u5e76\u53d1\u8bf7\u6c42\u4ee5\u53ca\u5982\u4f55\u8bbe\u7f6e\u3002\n2. **\u6280\u672f\u652f\u6301**\uff1a\u8054\u7cfb Ollama \u7684\u6280\u672f\u652f\u6301\u6216\u5ba2\u6237\u670d\u52a1\u56e2\u961f\u4ee5\u83b7\u53d6\u66f4\u591a\u5177\u4f53\u4fe1\u606f\u3002\n3. **\u793e\u533a\u8d44\u6e90**\uff1a\u67e5\u770b\u6216\u8be2\u95ee\u76f8\u5173\u5f00\u53d1\u8005\u793e\u533a\uff0c\u770b\u770b\u662f\u5426\u6709\u76f8\u4f3c\u7684\u4f7f\u7528\u6848\u4f8b\u548c\u6700\u4f73\u5b9e\u8df5\u3002\n\n\u901a\u8fc7\u5408\u9002\u7684\u5ba2\u6237\u7aef\u5e76\u53d1\u5904\u7406\u65b9\u5f0f\uff0c\u53ef\u4ee5\u5927\u5927\u63d0\u5347 API \u8c03\u7528\u7684\u6548\u7387\u548c\u541e\u5410\u91cf\u3002\u6839\u636e\u7cfb\u7edf\u548c API \u7684\u5177\u4f53\u884c\u4e3a\uff0c\u9009\u62e9\u6700\u9002\u5408\u7684\u5e76\u884c\u6216\u5f02\u6b65\u5904\u7406\u65b9\u5f0f\u3002","behavior":"degenerate_repetition","conversation_hash":"44b305e7e7991de62b21af46e7a54afe","conversation_redacted":false,"conversation_toxic":false,"country":"United States","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Chinese","model":"gpt-4o-2024-05-13","n_chars":2834,"n_turns_in_conversation":16,"n_words":278,"source_row_offset":7410,"source_shard":"turns-00026.parquet","timestamp":"Sat, 27 Jul 2024 08:10:57 GMT","turn_id":"0a3b6caf917027036a77674f","turn_index":15,"user_text":"ollama\u662f\u5426\u53ef\u4ee5\u6307\u5b9a\u591a\u7ebf\u7a0b\u8fd0\u884c\uff1f"},{"assistant_initiates":false,"asst_text":"\n/imagine prompt: Gold Saint Seiya, shining armor reflecting light, intricate details on the helmet and armor, powerful stance with raised sword, standing in a mythical arena, surrounded by celestial motifs and glowing energy, capturing the essence of valor and heroism, rendered in a realistic 3D model, capturing the character's iconic look and fierce demeanor, --ar 16:9 --v 5\n\n/imagine prompt: Gold Saint Seiya, gleaming golden armor with intricate engravings, dynamic pose with sword raised high, amidst a celestial galaxy backdrop, stars and galaxies swirling around, embodying a sense of cosmic power and strength, composed in a dramatic and epic style, emphasizing the character's mythical essence and warrior spirit, --ar 9:16 --v 5\n\n/imagine prompt: Gold Saint Seiya, radiant gold armor with ornate details, in a mythical temple setting, pillars and statues adorned with intricate carvings, mystical symbols floating in the air, creating a sacred and divine atmosphere, captured in a realistic 3D model with attention to textures and lighting, conveying a sense of ancient mysticism and sacred power, --ar 1:1 --niji\n\n/imagine prompt: Gold Saint Seiya, majestic stance with sword held high, surrounded by cosmic energy and celestial light rays, standing on a celestial platform floating in space, overlooking a vast universe filled with stars and nebulae, evoking a sense of cosmic grandeur and heroism, rendered in a detailed and immersive 3D model, capturing the character's divine presence and awe-inspiring aura, --ar 9:16 --v 5.","behavior":"degenerate_repetition","conversation_hash":"c50bb1ee159ec02848d3bc5088123aad","conversation_redacted":false,"conversation_toxic":false,"country":"United States","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"English","model":"gpt-3.5-turbo-0125","n_chars":5427,"n_turns_in_conversation":1,"n_words":699,"source_row_offset":7411,"source_shard":"turns-00026.parquet","timestamp":"Sat, 27 Jul 2024 08:12:55 GMT","turn_id":"46ca3181ebbf368a3ae59554","turn_index":0,"user_text":"\n As a prompt generator for a generative AI called \"Midjourney\", you will create image prompts for the AI to visualize. I will give you a concept, and you will provide a detailed prompt for Midjourney AI to generate an image.\n \n Please adhere to the structure and formatting below, and follow these guidelines:\n \n Do not use the words \"description\" or \":\" in any form.\n Do not place a comma between [ar] and [v].\n Write each prompt in one line without using return.\n Structure:\n [1] = Image of Gold Saint Seiya from Saint Seiya comics, 3d model, white background, realistic effect\n [2] = a detailed description of [1] with specific imagery details.\n [3] = a detailed description of the scene's environment.\n [4] = a detailed description of the compositions.\n [5] = a detailed description of the scene's mood, feelings, and atmosphere.\n [6] = A style (e.g. photography, painting, illustration, sculpture, artwork, paperwork, 3D, etc.) for [1].\n [7] = a detailed description of the scene's mood, feelings, and atmosphere.\n [ar] = Use \"--ar 16:9\" for horizontal images, \"--ar 9:16\" for vertical images, or \"--ar 1:1\" for square images.\n [v] = Use \"--niji\" for Japanese art style, or \"--v 5\" for other styles.\n \n \n Formatting:\n Follow this prompt structure: \"/imagine prompt: [1], [2], [3], [4], [5], [6], [7], [ar] [v]\".\n \n Your task: Create 4 distinct prompts for each concept [1], varying in details description, environment,compositions,atmosphere, and realization.\n \n Write your prompts in english.\n Do not describe unreal concepts as \"real\" or \"photographic\".\n Include one realistic photographic style prompt with lens type and size.\n Separate different prompts with two new lines.\n Example Prompts:\n \n /imagine prompt: cute dog, fluffy fur, wagging tail, playful expression, sitting on a grassy field, under a clear blue sky, with a colorful collar, in a natural and vibrant setting, by a lake, captured with a Nikon D750 camera, 50mm lens, shallow depth of field, composition focused on the dog's face, capturing its joyful spirit, in a style reminiscent of William Wegman's iconic dog portraits. --ar 1:1 --v 5.2\n /imagine prompt: beautiful women in the coffee shop, elegant and sophisticated, sipping a cup of steaming coffee, natural sunlight streaming through the window, soft and warm color tones, vintage decor with cozy armchairs and wooden tables, a bookshelf filled with classic novels, delicate porcelain teacups, a hint of aromatic coffee beans in the air, captured by a Leica M10 camera, 35mm lens, capturing the essence of timeless beauty, composition focused on the woman's face and hands, reminiscent of a painting by Leonardo da Vinci. --ar 1:1 --v 5.2\n /imagine prompt: A captivating Halo Reach landscape with a Spartan amidst a battlefield, fallen enemies around, smoke and fire in the background, emphasizing the Spartan's determination and bravery, detailed environment blending chaos and beauty, Illustration, digital art, --ar 16:9 --v 5\n "},{"assistant_initiates":false,"asst_text":"Posh\u793e\u4ea4\u6d3b\u52a8\u5e94\u7528\u83b72200\u4e07\u7f8e\u5143\u878d\u8d44\uff0c\u76ee\u6807\u5e02\u573aC\u7aef\u7528\u6237\uff0c\u501f\u9274TikTok\u8bbe\u8ba1\uff0c\u7b80\u5316\u6d41\u7a0b\u63d0\u5347\u6d3b\u52a8\u53c2\u4e0e\u4f53\u9a8c\uff0c\u4e3b\u8981\u9760\u552e\u7968\u4f63\u91d1\u76c8\u5229\uff0c\u672a\u6765\u8ba1\u5212\u6269\u5c55\u9879\u76ee\u7c7b\u522b\u548c\u52a0\u5f3a\u7528\u6237\u4e92\u52a8\u3002","behavior":"degenerate_repetition","conversation_hash":"446b121f0db7657870be1ddea309265f","conversation_redacted":false,"conversation_toxic":false,"country":"United States","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Chinese","model":"gpt-4o-2024-05-13","n_chars":3483,"n_turns_in_conversation":1,"n_words":75,"source_row_offset":7412,"source_shard":"turns-00026.parquet","timestamp":"Sat, 27 Jul 2024 08:13:15 GMT","turn_id":"14d9b9e9d5210bb16d0ba087","turn_index":0,"user_text":"\u6574\u7406\u5e76\u603b\u7ed3\u4e0b\u9762\u7684\u6587\u7ae0\uff0c\u8fd4\u56de50\u5b57\u6458\u8981

\u66fe\u51e0\u4f55\u65f6\uff0c\u56fd\u5185\u8fd8\u662f\u9760\u501f\u9274\u7f8e\u56fd\u7684\u5e94\u7528\u53d1\u5c55\u3002\u73b0\u5728\uff0c\u6211\u4eec\u5df2\u7ecf\u5f00\u59cb\u53cd\u5411\u8f93\u51fa\u2014\u2014\u7f8e\u7248\u201c\u5927\u9ea6\u201d\u793e\u4ea4\u6d3b\u52a8\u5e94\u7528Posh\u83b72200\u4e07\u7f8e\u5143\u878d\u8d44\uff0c\u8fd9\u4e2a\u4ea7\u54c1\u5982\u4f55\uff1f\u6211\u4eec\u6765\u770b\u770b\u4f5c\u8005\u7684\u5206\u6790\u3002

\n

\n

\u7ebf\u4e0a\u793e\u4ea4\u5df2\u7ecf\u4e0d\u80fd\u6ee1\u8db3\u4eba\u4eec\u7684\u9700\u6c42\uff0c\u7ebf\u4e0b\u793e\u4ea4\u6d3b\u52a8\u7684\u98ce\u522e\u5f97\u6b63\u731b\u3002

\n

7\u670823\u65e5\uff0c\u6d77\u5916\u793e\u4ea4\u6d3b\u52a8\u5e73\u53f0Posh\u5ba3\u5e03\u83b7\u5f972200\u4e07\u7f8e\u5143\u7684A\u8f6e\u878d\u8d44\uff0c\u7531Goodwater Capital\u9886\u6295\uff0cCompanyon Ventures\u3001Epic Ventures\u3001Day One Ventures\u548cPareto Holdings \u7b49\u8ddf\u6295\uff0c\u672c\u8f6e\u878d\u8d44\u8fc7\u540e\uff0cPosh\u7684\u603b\u878d\u8d44\u91d1\u989d\u8fbe\u5230\u4e863100\u4e07\u7f8e\u5143\u3002

\n

Posh\u662f2020\u5e74\u63a8\u51fa\u7684\u5e73\u53f0\uff0c\u8d77\u521d\u8be5\u5e73\u53f0\u6d89\u53ca\u8f6f\u4ef6\u670d\u52a1\uff0c\u76ee\u6807\u5ba2\u6237\u662f\u4f01\u4e1a\u3002\u4e4b\u540e\uff0c\u8fd9\u4e2a\u5e73\u53f0\u5c06\u5e02\u573a\u5bf9\u51c6C\u7aef\u6d88\u8d39\u8005\uff0c\u4e3a\u666e\u901a\u7528\u6237\u63d0\u4f9b\u793e\u4ea4\u6d3b\u52a8\u670d\u52a1\u3002

\n

\u53bb\u5e7410\u6708\uff0cPosh\u7684\u793e\u4ea4\u6d3b\u52a8\u529f\u80fd\u6b63\u5f0f\u95ee\u4e16\uff0c\u5728\u9875\u9762\u5c55\u73b0\u5f62\u5f0f\u4e0a\uff0cPosh\u501f\u9274\u4e86TikTok\u7b49\u5e94\u7528\u7a0b\u5e8f\u7684\u9875\u9762\u8bbe\u8ba1\uff0c\u901a\u8fc7\u6ed1\u52a8\u754c\u9762\u7684\u8fc7\u7a0b\uff0c\u7528\u6237\u53ef\u4ee5\u770b\u5230Posh\u4e0a\u9762\u6709\u54ea\u4e9b\u65b0\u6d3b\u52a8\uff0c\u53d1\u73b0\u597d\u73a9\u6709\u8da3\u7684\u793e\u4ea4\u6d3b\u52a8\uff0c\u6574\u4e2a\u8fc7\u7a0b\u5c31\u50cf\u5728TikTok\u4e0a\u6d4f\u89c8\u89c6\u9891\u3002

\n

Posh\u5bf9\u65b0\u624b\u6d3b\u52a8\u7ec4\u7ec7\u65b9\u66f4\u53cb\u597d\uff0c\u56e0\u4e3a\u5728\u8fd9\u4e2a\u7a0b\u5e8f\u4e0a\uff0cPosh\u53ef\u4ee5\u4e3a\u65b0\u624b\u6d3b\u52a8\u7ec4\u7ec7\u8005\u63d0\u4f9b\u66f4\u516c\u5e73\u7684\u4fe1\u606f\u5c55\u73b0\u6e20\u9053\u3002

\n

\u800c\u4ece\u6d3b\u52a8\u53c2\u4e0e\u8005\u7684\u89d2\u5ea6\u6765\u8bf4\uff0c\u4ed6\u4eec\u5219\u66f4\u5e0c\u671b\u6709\u4e00\u4e2a\u6d41\u7a0b\u7b80\u5355\u7684\u6d3b\u52a8\u53c2\u4e0e\u8fc7\u7a0b\u3002

\n

Posh\u5c31\u662f\u6293\u4f4f\u8fd9\u4e00\u75db\u70b9\uff0c\u5728\u6d3b\u52a8\u6d41\u7a0b\u7684\u8bbe\u8ba1\u65b9\u9762\u5c3d\u53ef\u80fd\u7b80\u5316\uff0c\u628a\u793e\u4ea4\u6d3b\u52a8\u5c55\u73b0\u573a\u666f\u8bbe\u8ba1\u7684\u66f4\u52a0\u4eba\u6027\u5316\uff0c\u7279\u522b\u662f\u975e\u5e38\u6ce8\u91cd\u5bf9\u827a\u672f\u5bb6\u54c1\u724c\u7684\u5ba3\u4f20\uff0c\u8fd9\u6837\u80fd\u591f\u8ba9\u6d3b\u52a8\u53c2\u4e0e\u8005\u66f4\u76f4\u89c2\u66f4\u5feb\u901f\u7684\u5bf9\u6d3b\u52a8\u6027\u8d28\u6709\u6240\u4e86\u89e3\u3002

\n

\u56e0\u4e3a\u662f\u5e73\u53f0\u5c5e\u6027\uff0cPosh\u7684\u670d\u52a1\u662f\u9488\u5bf9\u6d3b\u52a8\u7ec4\u7ec7\u8005\u548c\u6d3b\u52a8\u53c2\u4e0e\u8005\u53cc\u65b9\u7684\u3002

\n
\n

\u9664\u4e86\u7ed9\u6d3b\u52a8\u7ec4\u7ec7\u8005\u63d0\u4f9b\u4fe1\u606f\u5c55\u73b0\u4e4b\u5916\uff0cPosh\u8fd8\u5e2e\u52a9\u4ed6\u4eec\u5bfb\u627e\u573a\u5730\u4ee5\u53ca\u6444\u5f71\u5e08\u7b49\u7b49\uff0c\u76f8\u5f53\u4e8e\u5c06\u4e1a\u52a1\u8303\u56f4\u4ecb\u5165\u5230\u6d3b\u52a8\u7ec4\u7ec7\u73af\u8282\u4e2d\u53bb\uff0c\u8fd9\u80cc\u540e\u7684\u5e02\u573a\u673a\u4f1a\u4e5f\u662f\u5927\u5927\u589e\u52a0\u3002

\n

\u548cTikTok\u4e00\u6837\uff0cPosh\u5728\u7528\u6237\u7aef\u975e\u5e38\u6ce8\u91cd\u504f\u597d\u63a8\u8350\uff0c\u6b64\u4e3e\u65e8\u5728\u4fc3\u8fdb\u6d3b\u52a8\u7684\u591a\u6837\u6027\uff0c\u5e76\u4e14\u5e2e\u52a9\u548c\u5f15\u5bfc\u6d3b\u52a8\u53c2\u4e0e\u8005\u4e4b\u95f4\u5efa\u7acb\u66f4\u7262\u56fa\u7684\u793e\u4ea4\u5173\u7cfb\u3002

\n

\u4e0e\u6b64\u540c\u65f6\uff0c\u5bf9\u4e8e\u4e00\u4e9b\u4e0d\u9002\u5408\u81ea\u5df1\u6216\u8005\u4e0d\u611f\u5174\u8da3\u7684\u6d3b\u52a8\uff0c\u7528\u6237\u53ef\u4ee5\u9009\u62e9\u6027\u7684\u8fdb\u884c\u8fc7\u6ee4\uff0c\u8fd9\u4e5f\u6709\u52a9\u4e8e\u4fdd\u62a4\u81ea\u5df1\u7684\u9690\u79c1\u548c\u6570\u636e\u5b89\u5168\uff0c\u63d0\u9ad8\u4f7f\u7528Posh\u7684\u8212\u9002\u5ea6\u3002

\n

\"\u793e\u4ea4\u6d3b\u52a8\u5e94\u7528Posh\u83b72200\u4e07\u7f8e\u5143\u878d\u8d44\uff0c\u7f8e\u7248\u201c\u5927\u9ea6\u201d\u9760\u8d5a\u95e8\u7968\u4f63\u91d1\u62a2\u5360\u56db\u5206\u4e4b\u4e00\u5e02\u573a\"

\n

\u6253\u5f00Posh\u53ef\u4ee5\u53d1\u73b0\uff0c\u5176\u9875\u9762\u975e\u5e38\u7b80\u6d01\uff0c\u7528\u6237\u53ef\u4ee5\u9009\u62e9\u201c\u4eca\u65e5\u201d\u201c\u672c\u5468\u201d\u4ee5\u53ca\u201c\u672c\u6708\u201d\u7684\u6d3b\u52a8\uff0c\u8fd9\u4e9b\u6d3b\u52a8\u8986\u76d6\u9886\u57df\u975e\u5e38\u5e7f\u6cdb\uff0c\u65e0\u8bba\u662f\u793e\u4ea4\u9152\u4f1a\uff0c\u6447\u6eda\u6d3e\u5bf9\u8fd8\u662f\u6237\u5916\u6d3b\u52a8\u7b49\u90fd\u53ef\u4ee5\u5728\u4e0a\u9762\u627e\u5230\u5fd7\u540c\u9053\u5408\u7684\u201c\u642d\u5b50\u201d\uff0c\u8fd9\u79cd\u5f62\u5f0f\uff0c\u770b\u8d77\u6765\u5f88\u50cf\u6211\u4eec\u56fd\u5185\u7b80\u5316\u7248\u7684\u201c\u5927\u9ea6\u7f51\u201d\u3002

\n

\u6570\u636e\u8bc1\u660e\uff0cPosh\u8fd9\u7c7b\u5e73\u53f0\u786e\u5b9e\u6709\u5f88\u5927\u7684\u5e02\u573a\u9700\u6c42\u3002

\n

\u4ece\u5f00\u59cb\u8fd0\u8425\u793e\u4ea4\u5e73\u53f0\u81f3\u4eca\uff0cPosh\u7684\u7528\u6237\u6570\u5df2\u7ecf\u7a81\u7834200\u4e07\uff0c\u5728\u5339\u914d\u6d3b\u52a8\u53c2\u4e0e\u8005\u65b9\u9762\u7684\u6570\u636e\u589e\u957f\u4e861400%\u3002

\n

Posh\u7684\u4e3b\u8981\u8425\u6536\u6e20\u9053\u6765\u81ea\u9500\u552e\u95e8\u7968\u83b7\u5f97\u7684\u4f63\u91d1\uff0c\u76ee\u524d\uff0c\u5176\u9500\u552e\u7684\u6bcf\u5f20\u95e8\u7968\u90fd\u4f1a\u6536\u53d610%\u7684\u4f63\u91d1+ 0.99\u7f8e\u5143\u3002

\n

\u5728Posh\u4e0a\uff0c\u6bcf\u6708\u4e3e\u529e\u7ea65000\u573a\u6d3b\u52a8\uff0c\u4e3b\u8981\u96c6\u4e2d\u5728\u591c\u751f\u6d3b\u548c\u793e\u4ea4\u6d3b\u52a8\uff0c\u76ee\u524d\uff0c\u5176\u901a\u8fc7\u9500\u552e\u95e8\u7968\u83b7\u5f97\u7684\u6536\u5165\u8d859500\u4e07\u7f8e\u5143\u3002

\n

\u4e0d\u8fc7\uff0cPosh\u81f3\u4eca\u5c1a\u672a\u76c8\u5229\uff0c\u800c\u5176\u56e2\u961f\u73b0\u9636\u6bb5\u7684\u7ecf\u8425\u4fa7\u91cd\u70b9\u4e5f\u5728\u4e8e\u7528\u6237\u589e\u957f\u65b9\u9762\uff0c\u800c\u4e0d\u662f\u6536\u5165\u548c\u76c8\u5229\u3002

\n

\u5728\u516d\u4e2a\u6708\u7684\u65f6\u95f4\u5185\uff0cPosh\u7684\u5e02\u573a\u5360\u6709\u4efd\u989d\u5df2\u7ecf\u4ece6%\u8dc3\u5347\u81f312%\uff0c\u4e5f\u6070\u6070\u662f\u56e0\u4e3a\u6b64\uff0cPosh\u7684\u5e02\u573a\u65b9\u5411\u5f97\u5230\u4e86\u9a8c\u8bc1\uff0c\u5176\u56e2\u961f\u8bbe\u5b9a\u7684\u76ee\u6807\u662f\uff0c\u52302024\u5e74\u5e95\uff0c\u62ff\u4e0b25%\u7684\u5e02\u573a\u4efd\u989d\u3002

\n

\u5546\u4e1a\u6a21\u5f0f\u8fd8\u662f\u597d\u7684\uff0c\u76ee\u524d\uff0cPosh\u7684\u590d\u8d2d\u7528\u6237\u5927\u6982\u670930\u4e07\uff0c\u4e3a\u4e86\u63d0\u9ad8\u590d\u8d2d\uff0cPosh\u65b9\u9762\u8ba4\u4e3a\uff0c\u4e2a\u6027\u5316\u63a8\u8350\u662f\u4e00\u4e2a\u5fc5\u8981\u7684\u73af\u8282\u3002

\n

\u5230\u4eca\u5e74\u7b2c\u4e09\u5b63\u5ea6\uff0cPosh\u4f1a\u63a8\u51fa\u4e13\u95e8\u7684\u201cFor You\u201d\u9009\u9879\uff0c\u5728\u8fd9\u4e2a\u9009\u9879\u4e2d\uff0cPosh\u53ef\u4ee5\u9488\u5bf9\u7528\u6237\u559c\u597d\u63d0\u4f9b\u66f4\u4e2a\u6027\u5316\u7684\u6d3b\u52a8\u5efa\u8bae\uff0c\u5f53\u7136\uff0c\u8fd9\u4f9d\u8d56\u4e8e\u66f4\u9ad8\u7ea7\u7684\u7b97\u6cd5\u3002

\n

\u56f4\u7ed5\u793e\u4ea4\u5c5e\u6027\uff0cPosh\u8fd8\u8ba1\u5212\u5c06\u5e73\u53f0\u7684\u6d3b\u52a8\u9879\u76ee\u6269\u5c55\u5230\u5065\u8eab\u3001\u827a\u672f\u548c\u7f8e\u98df\u4f53\u9a8c\u7b49\u65b0\u7c7b\u522b\u3002

\n

\u800c\u4f5c\u4e3a\u5e73\u53f0\uff0c\u867d\u7136Posh\u63d0\u4f9b\u7684\u662f\u7ebf\u4e0a\u6e20\u9053\u63a8\u5e7f\u7ebf\u4e0b\u6d3b\u52a8\u4ea4\u6d41\u673a\u4f1a\uff0c\u4f46\u62e5\u6709\u6570\u767e\u4e07\u7ebf\u4e0a\u7528\u6237\u8fd9\u6837\u7684\u8d44\u6e90\u4e5f\u4e0d\u80fd\u6d6a\u8d39\uff0cPosh\u8ba1\u5212\u5c06\u7ebf\u4e0a\u5e73\u53f0\u5f80\u64ad\u5ba2\u65b9\u5411\u8fdb\u884c\u8f6c\u5316\uff0c\u52a0\u5f3a\u4e0e\u7528\u6237\u7684\u4e92\u52a8\u6765\u589e\u5f3a\u7c98\u6027\uff0c\u63d0\u9ad8\u590d\u8d2d\u3002

\n

\"\u793e\u4ea4\u6d3b\u52a8\u5e94\u7528Posh\u83b72200\u4e07\u7f8e\u5143\u878d\u8d44\uff0c\u7f8e\u7248\u201c\u5927\u9ea6\u201d\u9760\u8d5a\u95e8\u7968\u4f63\u91d1\u62a2\u5360\u56db\u5206\u4e4b\u4e00\u5e02\u573a\"

\n

Posh\u7684CEO Avante Price\u66fe\u662f\u7ebd\u7ea6\u6d41\u884c\u4ff1\u4e50\u90e8\u7684\u8457\u540dDJ\uff0c\u53e6\u4e00\u4f4d\u521b\u59cb\u4eba\u5219\u7ecf\u8425\u7740\u4e00\u5bb6\u4e3a\u65f6\u5c1a\u754c\u548c\u97f3\u4e50\u754c\u540d\u4eba\u63d0\u4f9b\u6444\u5f71\u6444\u50cf\u670d\u52a1\u7684\u673a\u6784\u3002

\n

\u4e24\u4f4d\u521b\u59cb\u4eba\u5728\u6f14\u51fa\u548c\u7968\u52a1\u65b9\u9762\u90fd\u6bd4\u8f83\u6709\u7ecf\u9a8c\uff0c\u662f\u540d\u526f\u5176\u5b9e\u7684\u5185\u884c\u4eba\uff0c\u5305\u62ec\u76ee\u524dPosh\u6d89\u53ca\u5230\u7684\u4e3a\u970d\u603b\u7ec4\u7ec7\u8005\u5bfb\u627e\u6444\u50cf\u5e08\u8fd9\u6837\u7684\u670d\u52a1\uff0c\u5176\u521b\u59cb\u4eba\u4e5f\u6709\u7ecf\u9a8c\u548c\u8d44\u6e90\u652f\u6301\u3002

\n

\u4e30\u5bcc\u7684\u884c\u4e1a\u7ecf\u9a8c\u662fPosh\u80fd\u591f\u83b7\u5f97\u5e02\u573a\u548c\u8d44\u672c\u8ba4\u53ef\u7684\u91cd\u8981\u539f\u56e0\uff0cAvante Price \u8868\u793a\uff1a\u201c\u5728\u5f53\u4eca\u6570\u5b57\u5316\u65f6\u4ee3\uff0c\u4eba\u4e0e\u4eba\u4e4b\u95f4\u7684\u8054\u7cfb\u6b63\u5728\u51cf\u5f31\uff0c\u5728Posh\uff0c\u6211\u4eec\u76f8\u4fe1\u901a\u8fc7\u521b\u5efa\u4e00\u4e2a\u652f\u6301\u73b0\u5b9e\u4e16\u754c\u8054\u7cfb\u548c\u6709\u610f\u4e49\u4e92\u52a8\u7684\u5e73\u53f0\uff0c\u53ef\u4ee5\u626d\u8f6c\u8fd9\u4e00\u8d8b\u52bf\u3002\u201d

\n

\u4f5c\u8005\u4e28\u5468\u53ef \u516c\u4f17\u53f7\uff1a\u626c\u5e06\u51fa\u6d77

\n

\u672c\u6587\u7531 @\u626c\u5e06\u51fa\u6d77 \u539f\u521b\u53d1\u5e03\u4e8e\u4eba\u4eba\u90fd\u662f\u4ea7\u54c1\u7ecf\u7406\u3002\u672a\u7ecf\u4f5c\u8005\u8bb8\u53ef\uff0c\u7981\u6b62\u8f6c\u8f7d

\n

\u9898\u56fe\u6765\u81ea Unsplash\uff0c\u57fa\u4e8eCC0\u534f\u8bae

\n

\u8be5\u6587\u89c2\u70b9\u4ec5\u4ee3\u8868\u4f5c\u8005\u672c\u4eba\uff0c\u4eba\u4eba\u90fd\u662f\u4ea7\u54c1\u7ecf\u7406\u5e73\u53f0\u4ec5\u63d0\u4f9b\u4fe1\u606f\u5b58\u50a8\u7a7a\u95f4\u670d\u52a1

\n
"},{"assistant_initiates":false,"asst_text":"\u4ee5\u4e0b\u662f\u6574\u7406\u51fa\u7684\u5173\u4e8e\u4ea7\u54c1\u7ecf\u7406\u768410\u6761\u5efa\u8bae\uff1a\n\n1. **\u5173\u6ce8\u5e02\u573a\u9700\u6c42**\uff1a\u7ebf\u4e0a\u793e\u4ea4\u5df2\u7ecf\u4e0d\u80fd\u6ee1\u8db3\u4eba\u4eec\u7684\u9700\u6c42\uff0c\u7ebf\u4e0b\u793e\u4ea4\u6d3b\u52a8\u6709\u660e\u663e\u7684\u5e02\u573a\u673a\u4f1a\u3002\u4ea7\u54c1\u7ecf\u7406\u5e94\u65f6\u523b\u5173\u6ce8\u5e02\u573a\u53d8\u5316\uff0c\u4e86\u89e3\u7528\u6237\u65b0\u5174\u9700\u6c42\u3002\n\n2. **\u7b80\u5316\u7528\u6237\u6d41\u7a0b**\uff1a\u7528\u6237\u5e0c\u671b\u6709\u4e00\u4e2a\u6d41\u7a0b\u7b80\u5355\u7684\u6d3b\u52a8\u53c2\u4e0e\u8fc7\u7a0b\uff0c\u7b80\u5316\u6d41\u7a0b\u8bbe\u8ba1\u80fd\u63d0\u5347\u7528\u6237\u4f53\u9a8c\u3002\n\n3. **\u501f\u9274\u6210\u529f\u6848\u4f8b**\uff1aPosh\u501f\u9274\u4e86TikTok\u7684\u9875\u9762\u8bbe\u8ba1\uff0c\u901a\u8fc7\u6ed1\u52a8\u754c\u9762\u5c55\u793a\u6d3b\u52a8\uff0c\u4ea7\u54c1\u7ecf\u7406\u5e94\u5584\u4e8e\u5b66\u4e60\u548c\u501f\u9274\u6210\u529f\u4ea7\u54c1\u7684\u8bbe\u8ba1\u7406\u5ff5\u3002\n\n4. **\u53cc\u5411\u670d\u52a1**\uff1aPosh\u5bf9\u6d3b\u52a8\u7ec4\u7ec7\u8005\u548c\u6d3b\u52a8\u53c2\u4e0e\u8005\u90fd\u63d0\u4f9b\u4e86\u4fbf\u5229\uff0c\u4ea7\u54c1\u7ecf\u7406\u8981\u8003\u8651\u4ea7\u54c1\u5982\u4f55\u670d\u52a1\u597d\u5e73\u53f0\u7684\u4e0d\u540c\u7528\u6237\u7fa4\u4f53\u3002\n\n5. **\u6ce8\u91cd\u7528\u6237\u504f\u597d**\uff1a\u548cTikTok\u4e00\u6837\uff0cPosh\u5728\u7528\u6237\u7aef\u975e\u5e38\u6ce8\u91cd\u504f\u597d\u63a8\u8350\uff0c\u4e2a\u6027\u5316\u63a8\u8350\u53ef\u4ee5\u63d0\u9ad8\u7528\u6237\u9ecf\u6027\u548c\u6ee1\u610f\u5ea6\u3002\n\n6. **\u6570\u636e\u548c\u9690\u79c1\u4fdd\u62a4**\uff1a\u7528\u6237\u53ef\u4ee5\u9009\u62e9\u6027\u8fdb\u884c\u6d3b\u52a8\u8fc7\u6ee4\uff0c\u8fd9\u6709\u52a9\u4e8e\u4fdd\u62a4\u9690\u79c1\u548c\u6570\u636e\u5b89\u5168\uff0c\u4ea7\u54c1\u7ecf\u7406\u5e94\u91cd\u89c6\u7528\u6237\u6570\u636e\u7684\u5b89\u5168\u6027\u3002\n\n7. **\u591a\u6837\u5316\u670d\u52a1**\uff1aPosh\u4e0d\u4ec5\u63d0\u4f9b\u6d3b\u52a8\u4fe1\u606f\u5c55\u793a\uff0c\u8fd8\u6df1\u5165\u4e1a\u52a1\u73af\u8282\uff0c\u4e3a\u6d3b\u52a8\u7ec4\u7ec7\u8005\u63d0\u4f9b\u66f4\u591a\u670d\u52a1\uff08\u5982\u5bfb\u627e\u573a\u5730\u548c\u6444\u5f71\u5e08\uff09\uff0c\u62d3\u5c55\u5e02\u573a\u673a\u4f1a\u3002\n\n8. **\u52a0\u5f3a\u54c1\u724c\u5ba3\u4f20**\uff1a\u7279\u522b\u6ce8\u91cd\u5bf9\u827a\u672f\u5bb6\u54c1\u724c\u7684\u5ba3\u4f20\uff0c\u6709\u52a9\u4e8e\u7528\u6237\u66f4\u76f4\u89c2\u5730\u4e86\u89e3\u6d3b\u52a8\u6027\u8d28\uff0c\u4ea7\u54c1\u7ecf\u7406\u8981\u91cd\u89c6\u54c1\u724c\u8425\u9500\u3002\n\n9. **\u63a2\u7d22\u65b0\u7c7b\u522b**\uff1aPosh\u8ba1\u5212\u6269\u5c55\u5230\u5065\u8eab\u3001\u827a\u672f\u548c\u7f8e\u98df\u4f53\u9a8c\u7b49\u65b0\u7c7b\u522b\uff0c\u4ea7\u54c1\u7ecf\u7406\u5e94\u4e0d\u65ad\u63a2\u7d22\u65b0\u7684\u4e1a\u52a1\u65b9\u5411\uff0c\u6ee1\u8db3\u66f4\u591a\u7528\u6237\u9700\u6c42\u3002\n\n10. **\u5229\u7528\u521b\u59cb\u4eba\u7ecf\u9a8c**\uff1aPosh\u7684\u521b\u59cb\u4eba\u5728\u793e\u4ea4\u548c\u6d3b\u52a8\u7ec4\u7ec7\u9886\u57df\u6709\u4e30\u5bcc\u7ecf\u9a8c\uff0c\u4ea7\u54c1\u7ecf\u7406\u5e94\u5229\u7528\u56e2\u961f\u6210\u5458\u7684\u4e13\u4e1a\u80cc\u666f\u548c\u884c\u4e1a\u8d44\u6e90\uff0c\u63a8\u52a8\u4ea7\u54c1\u53d1\u5c55\u3002","behavior":"degenerate_repetition","conversation_hash":"c71d81c915545259e6199ba3ddf4bbfe","conversation_redacted":false,"conversation_toxic":false,"country":"United States","empty_output":false,"evidence_codes":[],"label":"absent","label_present":false,"label_qualified":false,"label_source":"Final dense release","language":"Chinese","model":"gpt-4o-2024-05-13","n_chars":4057,"n_turns_in_conversation":1,"n_words":95,"source_row_offset":7413,"source_shard":"turns-00026.parquet","timestamp":"Sat, 27 Jul 2024 08:13:22 GMT","turn_id":"0dc267acd6d3c438cb4bfd2c","turn_index":0,"user_text":"\u5c06\u4e0b\u9762\u7684\u6587\u7ae0\u6574\u7406\u621010\u6761\u5173\u4e8e\u4ea7\u54c1\u7ecf\u7406\u7684\u5efa\u8bae\uff1a

\u66fe\u51e0\u4f55\u65f6\uff0c\u56fd\u5185\u8fd8\u662f\u9760\u501f\u9274\u7f8e\u56fd\u7684\u5e94\u7528\u53d1\u5c55\u3002\u73b0\u5728\uff0c\u6211\u4eec\u5df2\u7ecf\u5f00\u59cb\u53cd\u5411\u8f93\u51fa\u2014\u2014\u7f8e\u7248\u201c\u5927\u9ea6\u201d\u793e\u4ea4\u6d3b\u52a8\u5e94\u7528Posh\u83b72200\u4e07\u7f8e\u5143\u878d\u8d44\uff0c\u8fd9\u4e2a\u4ea7\u54c1\u5982\u4f55\uff1f\u6211\u4eec\u6765\u770b\u770b\u4f5c\u8005\u7684\u5206\u6790\u3002

\n

\n

\u7ebf\u4e0a\u793e\u4ea4\u5df2\u7ecf\u4e0d\u80fd\u6ee1\u8db3\u4eba\u4eec\u7684\u9700\u6c42\uff0c\u7ebf\u4e0b\u793e\u4ea4\u6d3b\u52a8\u7684\u98ce\u522e\u5f97\u6b63\u731b\u3002

\n

7\u670823\u65e5\uff0c\u6d77\u5916\u793e\u4ea4\u6d3b\u52a8\u5e73\u53f0Posh\u5ba3\u5e03\u83b7\u5f972200\u4e07\u7f8e\u5143\u7684A\u8f6e\u878d\u8d44\uff0c\u7531Goodwater Capital\u9886\u6295\uff0cCompanyon Ventures\u3001Epic Ventures\u3001Day One Ventures\u548cPareto Holdings \u7b49\u8ddf\u6295\uff0c\u672c\u8f6e\u878d\u8d44\u8fc7\u540e\uff0cPosh\u7684\u603b\u878d\u8d44\u91d1\u989d\u8fbe\u5230\u4e863100\u4e07\u7f8e\u5143\u3002

\n

Posh\u662f2020\u5e74\u63a8\u51fa\u7684\u5e73\u53f0\uff0c\u8d77\u521d\u8be5\u5e73\u53f0\u6d89\u53ca\u8f6f\u4ef6\u670d\u52a1\uff0c\u76ee\u6807\u5ba2\u6237\u662f\u4f01\u4e1a\u3002\u4e4b\u540e\uff0c\u8fd9\u4e2a\u5e73\u53f0\u5c06\u5e02\u573a\u5bf9\u51c6C\u7aef\u6d88\u8d39\u8005\uff0c\u4e3a\u666e\u901a\u7528\u6237\u63d0\u4f9b\u793e\u4ea4\u6d3b\u52a8\u670d\u52a1\u3002

\n

\u53bb\u5e7410\u6708\uff0cPosh\u7684\u793e\u4ea4\u6d3b\u52a8\u529f\u80fd\u6b63\u5f0f\u95ee\u4e16\uff0c\u5728\u9875\u9762\u5c55\u73b0\u5f62\u5f0f\u4e0a\uff0cPosh\u501f\u9274\u4e86TikTok\u7b49\u5e94\u7528\u7a0b\u5e8f\u7684\u9875\u9762\u8bbe\u8ba1\uff0c\u901a\u8fc7\u6ed1\u52a8\u754c\u9762\u7684\u8fc7\u7a0b\uff0c\u7528\u6237\u53ef\u4ee5\u770b\u5230Posh\u4e0a\u9762\u6709\u54ea\u4e9b\u65b0\u6d3b\u52a8\uff0c\u53d1\u73b0\u597d\u73a9\u6709\u8da3\u7684\u793e\u4ea4\u6d3b\u52a8\uff0c\u6574\u4e2a\u8fc7\u7a0b\u5c31\u50cf\u5728TikTok\u4e0a\u6d4f\u89c8\u89c6\u9891\u3002

\n

Posh\u5bf9\u65b0\u624b\u6d3b\u52a8\u7ec4\u7ec7\u65b9\u66f4\u53cb\u597d\uff0c\u56e0\u4e3a\u5728\u8fd9\u4e2a\u7a0b\u5e8f\u4e0a\uff0cPosh\u53ef\u4ee5\u4e3a\u65b0\u624b\u6d3b\u52a8\u7ec4\u7ec7\u8005\u63d0\u4f9b\u66f4\u516c\u5e73\u7684\u4fe1\u606f\u5c55\u73b0\u6e20\u9053\u3002

\n

\u800c\u4ece\u6d3b\u52a8\u53c2\u4e0e\u8005\u7684\u89d2\u5ea6\u6765\u8bf4\uff0c\u4ed6\u4eec\u5219\u66f4\u5e0c\u671b\u6709\u4e00\u4e2a\u6d41\u7a0b\u7b80\u5355\u7684\u6d3b\u52a8\u53c2\u4e0e\u8fc7\u7a0b\u3002

\n

Posh\u5c31\u662f\u6293\u4f4f\u8fd9\u4e00\u75db\u70b9\uff0c\u5728\u6d3b\u52a8\u6d41\u7a0b\u7684\u8bbe\u8ba1\u65b9\u9762\u5c3d\u53ef\u80fd\u7b80\u5316\uff0c\u628a\u793e\u4ea4\u6d3b\u52a8\u5c55\u73b0\u573a\u666f\u8bbe\u8ba1\u7684\u66f4\u52a0\u4eba\u6027\u5316\uff0c\u7279\u522b\u662f\u975e\u5e38\u6ce8\u91cd\u5bf9\u827a\u672f\u5bb6\u54c1\u724c\u7684\u5ba3\u4f20\uff0c\u8fd9\u6837\u80fd\u591f\u8ba9\u6d3b\u52a8\u53c2\u4e0e\u8005\u66f4\u76f4\u89c2\u66f4\u5feb\u901f\u7684\u5bf9\u6d3b\u52a8\u6027\u8d28\u6709\u6240\u4e86\u89e3\u3002

\n

\u56e0\u4e3a\u662f\u5e73\u53f0\u5c5e\u6027\uff0cPosh\u7684\u670d\u52a1\u662f\u9488\u5bf9\u6d3b\u52a8\u7ec4\u7ec7\u8005\u548c\u6d3b\u52a8\u53c2\u4e0e\u8005\u53cc\u65b9\u7684\u3002

\n
\n

\u9664\u4e86\u7ed9\u6d3b\u52a8\u7ec4\u7ec7\u8005\u63d0\u4f9b\u4fe1\u606f\u5c55\u73b0\u4e4b\u5916\uff0cPosh\u8fd8\u5e2e\u52a9\u4ed6\u4eec\u5bfb\u627e\u573a\u5730\u4ee5\u53ca\u6444\u5f71\u5e08\u7b49\u7b49\uff0c\u76f8\u5f53\u4e8e\u5c06\u4e1a\u52a1\u8303\u56f4\u4ecb\u5165\u5230\u6d3b\u52a8\u7ec4\u7ec7\u73af\u8282\u4e2d\u53bb\uff0c\u8fd9\u80cc\u540e\u7684\u5e02\u573a\u673a\u4f1a\u4e5f\u662f\u5927\u5927\u589e\u52a0\u3002

\n

\u548cTikTok\u4e00\u6837\uff0cPosh\u5728\u7528\u6237\u7aef\u975e\u5e38\u6ce8\u91cd\u504f\u597d\u63a8\u8350\uff0c\u6b64\u4e3e\u65e8\u5728\u4fc3\u8fdb\u6d3b\u52a8\u7684\u591a\u6837\u6027\uff0c\u5e76\u4e14\u5e2e\u52a9\u548c\u5f15\u5bfc\u6d3b\u52a8\u53c2\u4e0e\u8005\u4e4b\u95f4\u5efa\u7acb\u66f4\u7262\u56fa\u7684\u793e\u4ea4\u5173\u7cfb\u3002

\n

\u4e0e\u6b64\u540c\u65f6\uff0c\u5bf9\u4e8e\u4e00\u4e9b\u4e0d\u9002\u5408\u81ea\u5df1\u6216\u8005\u4e0d\u611f\u5174\u8da3\u7684\u6d3b\u52a8\uff0c\u7528\u6237\u53ef\u4ee5\u9009\u62e9\u6027\u7684\u8fdb\u884c\u8fc7\u6ee4\uff0c\u8fd9\u4e5f\u6709\u52a9\u4e8e\u4fdd\u62a4\u81ea\u5df1\u7684\u9690\u79c1\u548c\u6570\u636e\u5b89\u5168\uff0c\u63d0\u9ad8\u4f7f\u7528Posh\u7684\u8212\u9002\u5ea6\u3002

\n

\"\u793e\u4ea4\u6d3b\u52a8\u5e94\u7528Posh\u83b72200\u4e07\u7f8e\u5143\u878d\u8d44\uff0c\u7f8e\u7248\u201c\u5927\u9ea6\u201d\u9760\u8d5a\u95e8\u7968\u4f63\u91d1\u62a2\u5360\u56db\u5206\u4e4b\u4e00\u5e02\u573a\"

\n

\u6253\u5f00Posh\u53ef\u4ee5\u53d1\u73b0\uff0c\u5176\u9875\u9762\u975e\u5e38\u7b80\u6d01\uff0c\u7528\u6237\u53ef\u4ee5\u9009\u62e9\u201c\u4eca\u65e5\u201d\u201c\u672c\u5468\u201d\u4ee5\u53ca\u201c\u672c\u6708\u201d\u7684\u6d3b\u52a8\uff0c\u8fd9\u4e9b\u6d3b\u52a8\u8986\u76d6\u9886\u57df\u975e\u5e38\u5e7f\u6cdb\uff0c\u65e0\u8bba\u662f\u793e\u4ea4\u9152\u4f1a\uff0c\u6447\u6eda\u6d3e\u5bf9\u8fd8\u662f\u6237\u5916\u6d3b\u52a8\u7b49\u90fd\u53ef\u4ee5\u5728\u4e0a\u9762\u627e\u5230\u5fd7\u540c\u9053\u5408\u7684\u201c\u642d\u5b50\u201d\uff0c\u8fd9\u79cd\u5f62\u5f0f\uff0c\u770b\u8d77\u6765\u5f88\u50cf\u6211\u4eec\u56fd\u5185\u7b80\u5316\u7248\u7684\u201c\u5927\u9ea6\u7f51\u201d\u3002

\n

\u6570\u636e\u8bc1\u660e\uff0cPosh\u8fd9\u7c7b\u5e73\u53f0\u786e\u5b9e\u6709\u5f88\u5927\u7684\u5e02\u573a\u9700\u6c42\u3002

\n

\u4ece\u5f00\u59cb\u8fd0\u8425\u793e\u4ea4\u5e73\u53f0\u81f3\u4eca\uff0cPosh\u7684\u7528\u6237\u6570\u5df2\u7ecf\u7a81\u7834200\u4e07\uff0c\u5728\u5339\u914d\u6d3b\u52a8\u53c2\u4e0e\u8005\u65b9\u9762\u7684\u6570\u636e\u589e\u957f\u4e861400%\u3002

\n

Posh\u7684\u4e3b\u8981\u8425\u6536\u6e20\u9053\u6765\u81ea\u9500\u552e\u95e8\u7968\u83b7\u5f97\u7684\u4f63\u91d1\uff0c\u76ee\u524d\uff0c\u5176\u9500\u552e\u7684\u6bcf\u5f20\u95e8\u7968\u90fd\u4f1a\u6536\u53d610%\u7684\u4f63\u91d1+ 0.99\u7f8e\u5143\u3002

\n

\u5728Posh\u4e0a\uff0c\u6bcf\u6708\u4e3e\u529e\u7ea65000\u573a\u6d3b\u52a8\uff0c\u4e3b\u8981\u96c6\u4e2d\u5728\u591c\u751f\u6d3b\u548c\u793e\u4ea4\u6d3b\u52a8\uff0c\u76ee\u524d\uff0c\u5176\u901a\u8fc7\u9500\u552e\u95e8\u7968\u83b7\u5f97\u7684\u6536\u5165\u8d859500\u4e07\u7f8e\u5143\u3002

\n

\u4e0d\u8fc7\uff0cPosh\u81f3\u4eca\u5c1a\u672a\u76c8\u5229\uff0c\u800c\u5176\u56e2\u961f\u73b0\u9636\u6bb5\u7684\u7ecf\u8425\u4fa7\u91cd\u70b9\u4e5f\u5728\u4e8e\u7528\u6237\u589e\u957f\u65b9\u9762\uff0c\u800c\u4e0d\u662f\u6536\u5165\u548c\u76c8\u5229\u3002

\n

\u5728\u516d\u4e2a\u6708\u7684\u65f6\u95f4\u5185\uff0cPosh\u7684\u5e02\u573a\u5360\u6709\u4efd\u989d\u5df2\u7ecf\u4ece6%\u8dc3\u5347\u81f312%\uff0c\u4e5f\u6070\u6070\u662f\u56e0\u4e3a\u6b64\uff0cPosh\u7684\u5e02\u573a\u65b9\u5411\u5f97\u5230\u4e86\u9a8c\u8bc1\uff0c\u5176\u56e2\u961f\u8bbe\u5b9a\u7684\u76ee\u6807\u662f\uff0c\u52302024\u5e74\u5e95\uff0c\u62ff\u4e0b25%\u7684\u5e02\u573a\u4efd\u989d\u3002

\n

\u5546\u4e1a\u6a21\u5f0f\u8fd8\u662f\u597d\u7684\uff0c\u76ee\u524d\uff0cPosh\u7684\u590d\u8d2d\u7528\u6237\u5927\u6982\u670930\u4e07\uff0c\u4e3a\u4e86\u63d0\u9ad8\u590d\u8d2d\uff0cPosh\u65b9\u9762\u8ba4\u4e3a\uff0c\u4e2a\u6027\u5316\u63a8\u8350\u662f\u4e00\u4e2a\u5fc5\u8981\u7684\u73af\u8282\u3002

\n

\u5230\u4eca\u5e74\u7b2c\u4e09\u5b63\u5ea6\uff0cPosh\u4f1a\u63a8\u51fa\u4e13\u95e8\u7684\u201cFor You\u201d\u9009\u9879\uff0c\u5728\u8fd9\u4e2a\u9009\u9879\u4e2d\uff0cPosh\u53ef\u4ee5\u9488\u5bf9\u7528\u6237\u559c\u597d\u63d0\u4f9b\u66f4\u4e2a\u6027\u5316\u7684\u6d3b\u52a8\u5efa\u8bae\uff0c\u5f53\u7136\uff0c\u8fd9\u4f9d\u8d56\u4e8e\u66f4\u9ad8\u7ea7\u7684\u7b97\u6cd5\u3002

\n

\u56f4\u7ed5\u793e\u4ea4\u5c5e\u6027\uff0cPosh\u8fd8\u8ba1\u5212\u5c06\u5e73\u53f0\u7684\u6d3b\u52a8\u9879\u76ee\u6269\u5c55\u5230\u5065\u8eab\u3001\u827a\u672f\u548c\u7f8e\u98df\u4f53\u9a8c\u7b49\u65b0\u7c7b\u522b\u3002

\n

\u800c\u4f5c\u4e3a\u5e73\u53f0\uff0c\u867d\u7136Posh\u63d0\u4f9b\u7684\u662f\u7ebf\u4e0a\u6e20\u9053\u63a8\u5e7f\u7ebf\u4e0b\u6d3b\u52a8\u4ea4\u6d41\u673a\u4f1a\uff0c\u4f46\u62e5\u6709\u6570\u767e\u4e07\u7ebf\u4e0a\u7528\u6237\u8fd9\u6837\u7684\u8d44\u6e90\u4e5f\u4e0d\u80fd\u6d6a\u8d39\uff0cPosh\u8ba1\u5212\u5c06\u7ebf\u4e0a\u5e73\u53f0\u5f80\u64ad\u5ba2\u65b9\u5411\u8fdb\u884c\u8f6c\u5316\uff0c\u52a0\u5f3a\u4e0e\u7528\u6237\u7684\u4e92\u52a8\u6765\u589e\u5f3a\u7c98\u6027\uff0c\u63d0\u9ad8\u590d\u8d2d\u3002

\n

\"\u793e\u4ea4\u6d3b\u52a8\u5e94\u7528Posh\u83b72200\u4e07\u7f8e\u5143\u878d\u8d44\uff0c\u7f8e\u7248\u201c\u5927\u9ea6\u201d\u9760\u8d5a\u95e8\u7968\u4f63\u91d1\u62a2\u5360\u56db\u5206\u4e4b\u4e00\u5e02\u573a\"

\n

Posh\u7684CEO Avante Price\u66fe\u662f\u7ebd\u7ea6\u6d41\u884c\u4ff1\u4e50\u90e8\u7684\u8457\u540dDJ\uff0c\u53e6\u4e00\u4f4d\u521b\u59cb\u4eba\u5219\u7ecf\u8425\u7740\u4e00\u5bb6\u4e3a\u65f6\u5c1a\u754c\u548c\u97f3\u4e50\u754c\u540d\u4eba\u63d0\u4f9b\u6444\u5f71\u6444\u50cf\u670d\u52a1\u7684\u673a\u6784\u3002

\n

\u4e24\u4f4d\u521b\u59cb\u4eba\u5728\u6f14\u51fa\u548c\u7968\u52a1\u65b9\u9762\u90fd\u6bd4\u8f83\u6709\u7ecf\u9a8c\uff0c\u662f\u540d\u526f\u5176\u5b9e\u7684\u5185\u884c\u4eba\uff0c\u5305\u62ec\u76ee\u524dPosh\u6d89\u53ca\u5230\u7684\u4e3a\u970d\u603b\u7ec4\u7ec7\u8005\u5bfb\u627e\u6444\u50cf\u5e08\u8fd9\u6837\u7684\u670d\u52a1\uff0c\u5176\u521b\u59cb\u4eba\u4e5f\u6709\u7ecf\u9a8c\u548c\u8d44\u6e90\u652f\u6301\u3002

\n

\u4e30\u5bcc\u7684\u884c\u4e1a\u7ecf\u9a8c\u662fPosh\u80fd\u591f\u83b7\u5f97\u5e02\u573a\u548c\u8d44\u672c\u8ba4\u53ef\u7684\u91cd\u8981\u539f\u56e0\uff0cAvante Price \u8868\u793a\uff1a\u201c\u5728\u5f53\u4eca\u6570\u5b57\u5316\u65f6\u4ee3\uff0c\u4eba\u4e0e\u4eba\u4e4b\u95f4\u7684\u8054\u7cfb\u6b63\u5728\u51cf\u5f31\uff0c\u5728Posh\uff0c\u6211\u4eec\u76f8\u4fe1\u901a\u8fc7\u521b\u5efa\u4e00\u4e2a\u652f\u6301\u73b0\u5b9e\u4e16\u754c\u8054\u7cfb\u548c\u6709\u610f\u4e49\u4e92\u52a8\u7684\u5e73\u53f0\uff0c\u53ef\u4ee5\u626d\u8f6c\u8fd9\u4e00\u8d8b\u52bf\u3002\u201d

\n

\u4f5c\u8005\u4e28\u5468\u53ef \u516c\u4f17\u53f7\uff1a\u626c\u5e06\u51fa\u6d77

\n

\u672c\u6587\u7531 @\u626c\u5e06\u51fa\u6d77 \u539f\u521b\u53d1\u5e03\u4e8e\u4eba\u4eba\u90fd\u662f\u4ea7\u54c1\u7ecf\u7406\u3002\u672a\u7ecf\u4f5c\u8005\u8bb8\u53ef\uff0c\u7981\u6b62\u8f6c\u8f7d

\n

\u9898\u56fe\u6765\u81ea Unsplash\uff0c\u57fa\u4e8eCC0\u534f\u8bae

\n

\u8be5\u6587\u89c2\u70b9\u4ec5\u4ee3\u8868\u4f5c\u8005\u672c\u4eba\uff0c\u4eba\u4eba\u90fd\u662f\u4ea7\u54c1\u7ecf\u7406\u5e73\u53f0\u4ec5\u63d0\u4f9b\u4fe1\u606f\u5b58\u50a8\u7a7a\u95f4\u670d\u52a1

\n
"}],"shard":{"index":26,"name":"turns-00026.parquet","rows":75832}}