{"cells":[{"cell_type":"code","execution_count":1,"metadata":{"executionInfo":{"elapsed":476,"status":"ok","timestamp":1720679526275,"user":{"displayName":"HUANG DONGHAO _","userId":"00977795705617022768"},"user_tz":-480},"id":"uWKRSV6eZsCn"},"outputs":[],"source":["%load_ext autoreload\n","%autoreload 2"]},{"cell_type":"code","execution_count":3,"metadata":{"application/vnd.databricks.v1+cell":{"cellMetadata":{"byteLimit":2048000,"rowLimit":10000},"inputWidgets":{},"nuid":"eb33b19f-1206-41ee-84e2-e6258a12eef7","showTitle":false,"title":""},"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":2534,"status":"ok","timestamp":1720679529344,"user":{"displayName":"HUANG DONGHAO _","userId":"00977795705617022768"},"user_tz":-480},"id":"xwFh14uiZBrI","outputId":"d767799c-34c2-46a5-f052-378146a55321"},"outputs":[],"source":["from pathlib import Path\n","\n","try:\n"," from google.colab import drive\n","\n"," drive.mount(\"/content/drive\")\n"," workding_dir = \"/content/drive/MyDrive/logical-reasoning/\"\n","except ModuleNotFoundError:\n"," workding_dir = str(Path.cwd().parent)"]},{"cell_type":"code","execution_count":4,"metadata":{"application/vnd.databricks.v1+cell":{"cellMetadata":{"byteLimit":2048000,"rowLimit":10000},"inputWidgets":{},"nuid":"6d394937-6c99-4a7c-9d32-7600a280032f","showTitle":false,"title":""},"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":5,"status":"ok","timestamp":1720679529345,"user":{"displayName":"HUANG DONGHAO _","userId":"00977795705617022768"},"user_tz":-480},"id":"G5pNu3zgZBrL","outputId":"160a554f-fb08-4aa0-bc00-0422fb7c1fac"},"outputs":[{"name":"stdout","output_type":"stream","text":["workding dir: /home/inflaton/code/logical-reasoning\n"]}],"source":["import os\n","import sys\n","from pathlib import Path\n","\n","os.chdir(workding_dir)\n","sys.path.append(workding_dir)\n","print(\"workding dir:\", workding_dir)"]},{"cell_type":"code","execution_count":5,"metadata":{"application/vnd.databricks.v1+cell":{"cellMetadata":{"byteLimit":2048000,"rowLimit":10000},"inputWidgets":{},"nuid":"9f67ec60-2f24-411c-84eb-0dd664b44775","showTitle":false,"title":""},"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":3,"status":"ok","timestamp":1720679529345,"user":{"displayName":"HUANG DONGHAO _","userId":"00977795705617022768"},"user_tz":-480},"id":"hPCC-6m7ZBrM","outputId":"c7aa2c96-5e99-440a-c148-201d79465ff9"},"outputs":[{"name":"stdout","output_type":"stream","text":["loading env vars from: /home/inflaton/code/logical-reasoning/.env\n"]},{"data":{"text/plain":["True"]},"execution_count":5,"metadata":{},"output_type":"execute_result"}],"source":["from dotenv import find_dotenv, load_dotenv\n","\n","found_dotenv = find_dotenv(\".env\")\n","\n","if len(found_dotenv) == 0:\n"," found_dotenv = find_dotenv(\".env.example\")\n","print(f\"loading env vars from: {found_dotenv}\")\n","load_dotenv(found_dotenv, override=True)"]},{"cell_type":"code","execution_count":6,"metadata":{"application/vnd.databricks.v1+cell":{"cellMetadata":{"byteLimit":2048000,"rowLimit":10000},"inputWidgets":{},"nuid":"f1597656-8042-4878-9d3b-9ebfb8dd86dc","showTitle":false,"title":""},"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":3,"status":"ok","timestamp":1720679529345,"user":{"displayName":"HUANG DONGHAO _","userId":"00977795705617022768"},"user_tz":-480},"id":"1M3IraVtZBrM","outputId":"29ab35f6-2970-4ade-d85d-3174acf8cda0"},"outputs":[{"name":"stdout","output_type":"stream","text":["internlm/internlm2_5-7b-chat-1m inflaton-ai/InternLM_2_5-7b_LoRA-Adapter False datasets/mgtv results/mgtv-results_nv4090.csv\n"]}],"source":["import os\n","\n","model_name = os.getenv(\"MODEL_NAME\")\n","adapter_name_or_path = os.getenv(\"ADAPTER_NAME_OR_PATH\")\n","load_in_4bit = os.getenv(\"LOAD_IN_4BIT\") == \"true\"\n","data_path = os.getenv(\"LOGICAL_REASONING_DATA_PATH\")\n","results_path = os.getenv(\"LOGICAL_REASONING_RESULTS_PATH\")\n","use_english_datasets = os.getenv(\"USE_ENGLISH_DATASETS\") == \"true\"\n","\n","print(model_name, adapter_name_or_path, load_in_4bit, data_path, results_path)"]},{"cell_type":"code","execution_count":7,"metadata":{"application/vnd.databricks.v1+cell":{"cellMetadata":{"byteLimit":2048000,"rowLimit":10000},"inputWidgets":{},"nuid":"b2a43943-9324-4839-9a47-cfa72de2244b","showTitle":false,"title":""},"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":564,"status":"ok","timestamp":1720679529907,"user":{"displayName":"HUANG DONGHAO _","userId":"00977795705617022768"},"user_tz":-480},"id":"UgMvt6dIZBrM","outputId":"ce37581c-fd26-46c2-ad87-d933d99f68f7"},"outputs":[{"name":"stdout","output_type":"stream","text":["Python 3.11.9\n","\u001b[33mWARNING: Package(s) not found: torchaudio, torchvision\u001b[0m\u001b[33m\n","\u001b[0mName: transformers\n","Version: 4.41.2\n","Summary: State-of-the-art Machine Learning for JAX, PyTorch and TensorFlow\n","Home-page: https://github.com/huggingface/transformers\n","Author: The Hugging Face team (past and future) with the help of all our contributors (https://github.com/huggingface/transformers/graphs/contributors)\n","Author-email: transformers@huggingface.co\n","License: Apache 2.0 License\n","Location: /home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/site-packages\n","Requires: filelock, huggingface-hub, numpy, packaging, pyyaml, regex, requests, safetensors, tokenizers, tqdm\n","Required-by: llamafactory, peft, trl\n","---\n","Name: torch\n","Version: 2.3.1\n","Summary: Tensors and Dynamic neural networks in Python with strong GPU acceleration\n","Home-page: https://pytorch.org/\n","Author: PyTorch Team\n","Author-email: packages@pytorch.org\n","License: BSD-3\n","Location: /home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/site-packages\n","Requires: filelock, fsspec, jinja2, networkx, nvidia-cublas-cu12, nvidia-cuda-cupti-cu12, nvidia-cuda-nvrtc-cu12, nvidia-cuda-runtime-cu12, nvidia-cudnn-cu12, nvidia-cufft-cu12, nvidia-curand-cu12, nvidia-cusolver-cu12, nvidia-cusparse-cu12, nvidia-nccl-cu12, nvidia-nvtx-cu12, sympy, triton, typing-extensions\n","Required-by: accelerate, peft, trl\n","CPU times: user 16.9 ms, sys: 0 ns, total: 16.9 ms\n","Wall time: 1.86 s\n"]}],"source":["%%time\n","!python --version\n","!pip show transformers torch torchvision torchaudio"]},{"cell_type":"code","execution_count":10,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":1685,"status":"ok","timestamp":1720679531591,"user":{"displayName":"HUANG DONGHAO _","userId":"00977795705617022768"},"user_tz":-480},"id":"ZuS_FsLyZBrN","outputId":"2cba0105-c505-4395-afbd-2f2fee6581d0"},"outputs":[{"name":"stdout","output_type":"stream","text":["CUDA is available, we have found 1 GPU(s)\n","NVIDIA GeForce RTX 4090\n","CUDA version: 12.1\n"]}],"source":["from llm_toolkit.llm_utils import *\n","from llm_toolkit.logical_reasoning_utils import *\n","\n","device = check_gpu()"]},{"cell_type":"code","execution_count":15,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["Current Directory:\n","/home/inflaton/code/logical-reasoning\n","Mon Jul 22 21:10:43 2024 \n","+---------------------------------------------------------------------------------------+\n","| NVIDIA-SMI 535.54.04 Driver Version: 536.25 CUDA Version: 12.2 |\n","|-----------------------------------------+----------------------+----------------------+\n","| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |\n","| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |\n","| | | MIG M. |\n","|=========================================+======================+======================|\n","| 0 NVIDIA GeForce RTX 4090 On | 00000000:01:00.0 Off | Off |\n","| 0% 46C P8 13W / 450W | 1450MiB / 24564MiB | 1% Default |\n","| | | N/A |\n","+-----------------------------------------+----------------------+----------------------+\n"," \n","+---------------------------------------------------------------------------------------+\n","| Processes: |\n","| GPU GI CI PID Type Process name GPU Memory |\n","| ID ID Usage |\n","|=======================================================================================|\n","| No running processes found |\n","+---------------------------------------------------------------------------------------+\n","Linux Gen-AI 5.15.133.1-microsoft-standard-WSL2 #1 SMP Thu Oct 5 21:02:42 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux\n","PRETTY_NAME=\"Ubuntu 22.04.2 LTS\"\n","NAME=\"Ubuntu\"\n","VERSION_ID=\"22.04\"\n","VERSION=\"22.04.2 LTS (Jammy Jellyfish)\"\n","VERSION_CODENAME=jammy\n","ID=ubuntu\n","ID_LIKE=debian\n","HOME_URL=\"https://www.ubuntu.com/\"\n","SUPPORT_URL=\"https://help.ubuntu.com/\"\n","BUG_REPORT_URL=\"https://bugs.launchpad.net/ubuntu/\"\n","PRIVACY_POLICY_URL=\"https://www.ubuntu.com/legal/terms-and-policies/privacy-policy\"\n","UBUNTU_CODENAME=jammy\n","Architecture: x86_64\n"," CPU op-mode(s): 32-bit, 64-bit\n"," Address sizes: 39 bits physical, 48 bits virtual\n"," Byte Order: Little Endian\n","CPU(s): 32\n"," On-line CPU(s) list: 0-31\n","Vendor ID: GenuineIntel\n"," Model name: 13th Gen Intel(R) Core(TM) i9-13900KF\n"," CPU family: 6\n"," Model: 183\n"," Thread(s) per core: 2\n"," Core(s) per socket: 16\n"," Socket(s): 1\n"," Stepping: 1\n"," BogoMIPS: 5990.40\n"," Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mc\n"," a cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscal\n"," l nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopo\n"," logy tsc_reliable nonstop_tsc cpuid pni pclmulqdq vmx s\n"," sse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt tsc_dea\n"," dline_timer aes xsave avx f16c rdrand hypervisor lahf_l\n"," m abm 3dnowprefetch ssbd ibrs ibpb stibp ibrs_enhanced \n"," tpr_shadow vnmi ept vpid ept_ad fsgsbase tsc_adjust bmi\n"," 1 avx2 smep bmi2 erms invpcid rdseed adx smap clflushop\n"," t clwb sha_ni xsaveopt xsavec xgetbv1 xsaves avx_vnni u\n"," mip waitpkg gfni vaes vpclmulqdq rdpid movdiri movdir64\n"," b fsrm md_clear serialize flush_l1d arch_capabilities\n","Virtualization features: \n"," Virtualization: VT-x\n"," Hypervisor vendor: Microsoft\n"," Virtualization type: full\n","Caches (sum of all): \n"," L1d: 768 KiB (16 instances)\n"," L1i: 512 KiB (16 instances)\n"," L2: 32 MiB (16 instances)\n"," L3: 36 MiB (1 instance)\n","Vulnerabilities: \n"," Gather data sampling: Not affected\n"," Itlb multihit: Not affected\n"," L1tf: Not affected\n"," Mds: Not affected\n"," Meltdown: Not affected\n"," Mmio stale data: Not affected\n"," Retbleed: Mitigation; Enhanced IBRS\n"," Spec rstack overflow: Not affected\n"," Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl\n"," and seccomp\n"," Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer\n"," sanitization\n"," Spectre v2: Mitigation; Enhanced IBRS, IBPB conditional, RSB fillin\n"," g, PBRSB-eIBRS SW sequence\n"," Srbds: Not affected\n"," Tsx async abort: Not affected\n","MemTotal: 49330020 kB\n","Eval internlm/internlm2_5-7b-chat-1m with inflaton-ai/InternLM_2_5-7b_LoRA-Adapter\n","loading env vars from: /home/inflaton/code/logical-reasoning/.env\n","Adding /home/inflaton/code/logical-reasoning to sys.path\n","loading /home/inflaton/code/logical-reasoning/llm_toolkit/logical_reasoning_utils.py\n","internlm/internlm2_5-7b-chat-1m inflaton-ai/InternLM_2_5-7b_LoRA-Adapter False datasets/mgtv results/mgtv-results_internlm_best.csv\n","(1) GPU = NVIDIA GeForce RTX 4090. Max memory = 23.988 GB.\n","0.0 GB of memory reserved.\n","loading model: internlm/internlm2_5-7b-chat-1m with adapter: inflaton-ai/InternLM_2_5-7b_LoRA-Adapter\n","tokenizer_config.json: 100%|███████████████| 2.51k/2.51k [00:00<00:00, 38.0MB/s]\n","tokenization_internlm2_fast.py: 100%|██████| 7.80k/7.80k [00:00<00:00, 77.2MB/s]\n","tokenization_internlm2.py: 100%|███████████| 8.81k/8.81k [00:00<00:00, 89.2MB/s]\n","tokenizer.model: 100%|█████████████████████| 1.48M/1.48M [00:01<00:00, 1.02MB/s]\n","special_tokens_map.json: 100%|█████████████████| 713/713 [00:00<00:00, 12.8MB/s]\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:10:55,053 >> loading file ./tokenizer.model from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/./tokenizer.model\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:10:55,053 >> loading file added_tokens.json from cache at None\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:10:55,053 >> loading file special_tokens_map.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/special_tokens_map.json\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:10:55,053 >> loading file tokenizer_config.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/tokenizer_config.json\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:10:55,053 >> loading file tokenizer.json from cache at None\n","07/22/2024 21:10:55 - INFO - llamafactory.data.template - Add <|im_end|> to stop words.\n","/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.\n"," warnings.warn(\n","config.json: 100%|█████████████████████████████| 895/895 [00:00<00:00, 15.1MB/s]\n","[INFO|configuration_utils.py:733] 2024-07-22 21:10:55,959 >> loading configuration file config.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/config.json\n","configuration_internlm2.py: 100%|███████████| 8.84k/8.84k [00:00<00:00, 104MB/s]\n","[INFO|configuration_utils.py:733] 2024-07-22 21:10:56,833 >> loading configuration file config.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/config.json\n","[INFO|configuration_utils.py:796] 2024-07-22 21:10:56,833 >> Model config InternLM2Config {\n"," \"_name_or_path\": \"internlm/internlm2_5-7b-chat-1m\",\n"," \"architectures\": [\n"," \"InternLM2ForCausalLM\"\n"," ],\n"," \"attn_implementation\": \"eager\",\n"," \"auto_map\": {\n"," \"AutoConfig\": \"internlm/internlm2_5-7b-chat-1m--configuration_internlm2.InternLM2Config\",\n"," \"AutoModel\": \"internlm/internlm2_5-7b-chat-1m--modeling_internlm2.InternLM2ForCausalLM\",\n"," \"AutoModelForCausalLM\": \"internlm/internlm2_5-7b-chat-1m--modeling_internlm2.InternLM2ForCausalLM\"\n"," },\n"," \"bias\": false,\n"," \"bos_token_id\": 1,\n"," \"eos_token_id\": 2,\n"," \"hidden_act\": \"silu\",\n"," \"hidden_size\": 4096,\n"," \"initializer_range\": 0.02,\n"," \"intermediate_size\": 14336,\n"," \"max_position_embeddings\": 262144,\n"," \"model_type\": \"internlm2\",\n"," \"num_attention_heads\": 32,\n"," \"num_hidden_layers\": 32,\n"," \"num_key_value_heads\": 8,\n"," \"pad_token_id\": 2,\n"," \"pretraining_tp\": 1,\n"," \"rms_norm_eps\": 1e-05,\n"," \"rope_scaling\": {\n"," \"factor\": 2.5,\n"," \"type\": \"dynamic\"\n"," },\n"," \"rope_theta\": 50000000,\n"," \"tie_word_embeddings\": false,\n"," \"torch_dtype\": \"bfloat16\",\n"," \"transformers_version\": \"4.41.2\",\n"," \"use_cache\": true,\n"," \"vocab_size\": 92544\n","}\n","\n","07/22/2024 21:10:56 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.\n","modeling_internlm2.py: 100%|████████████████| 80.7k/80.7k [00:00<00:00, 263kB/s]\n","model.safetensors.index.json: 100%|████████| 18.2k/18.2k [00:00<00:00, 3.49MB/s]\n","[INFO|modeling_utils.py:3474] 2024-07-22 21:10:59,134 >> loading weights file model.safetensors from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/model.safetensors.index.json\n","Downloading shards: 0%| | 0/8 [00:00> Instantiating InternLM2ForCausalLM model under default dtype torch.bfloat16.\n","[INFO|configuration_utils.py:962] 2024-07-22 21:34:35,272 >> Generate config GenerationConfig {\n"," \"bos_token_id\": 1,\n"," \"eos_token_id\": 2,\n"," \"pad_token_id\": 2\n","}\n","\n","Loading checkpoint shards: 100%|██████████████████| 8/8 [00:02<00:00, 3.11it/s]\n","[INFO|modeling_utils.py:4280] 2024-07-22 21:34:41,963 >> All model checkpoint weights were used when initializing InternLM2ForCausalLM.\n","\n","[INFO|modeling_utils.py:4288] 2024-07-22 21:34:41,963 >> All the weights of InternLM2ForCausalLM were initialized from the model checkpoint at internlm/internlm2_5-7b-chat-1m.\n","If your task is similar to the task the model of the checkpoint was trained on, you can already use InternLM2ForCausalLM for predictions without further training.\n","generation_config.json: 100%|██████████████████| 123/123 [00:00<00:00, 1.97MB/s]\n","[INFO|configuration_utils.py:917] 2024-07-22 21:34:42,702 >> loading configuration file generation_config.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/generation_config.json\n","[INFO|configuration_utils.py:962] 2024-07-22 21:34:42,702 >> Generate config GenerationConfig {\n"," \"bos_token_id\": 1,\n"," \"eos_token_id\": [\n"," 2,\n"," 92542\n"," ],\n"," \"pad_token_id\": 2\n","}\n","\n","07/22/2024 21:34:42 - INFO - llamafactory.model.model_utils.attention - Using vanilla attention implementation.\n","adapter_config.json: 100%|█████████████████████| 679/679 [00:00<00:00, 9.82MB/s]\n","adapter_model.safetensors: 100%|███████████| 37.8M/37.8M [00:03<00:00, 11.8MB/s]\n","07/22/2024 21:34:48 - INFO - llamafactory.model.adapter - Merged 1 adapter(s).\n","07/22/2024 21:34:48 - INFO - llamafactory.model.adapter - Loaded adapter(s): inflaton-ai/InternLM_2_5-7b_LoRA-Adapter\n","07/22/2024 21:34:48 - INFO - llamafactory.model.loader - all params: 7,737,708,544\n","(2) GPU = NVIDIA GeForce RTX 4090. Max memory = 23.988 GB.\n","14.818 GB of memory reserved.\n","loading train/test data files\n","Generating train split: 25000 examples [00:00, 110406.75 examples/s]\n","Generating test split: 3000 examples [00:00, 201720.35 examples/s]\n","Map: 100%|██████████████████████| 25000/25000 [00:00<00:00, 48814.27 examples/s]\n","Map: 100%|████████████████████████| 3000/3000 [00:00<00:00, 40855.20 examples/s]\n","DatasetDict({\n"," train: Dataset({\n"," features: ['text', 'label', 'answer', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n"," num_rows: 25000\n"," })\n"," test: Dataset({\n"," features: ['text', 'label', 'answer', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n"," num_rows: 3000\n"," })\n","})\n","--------------------------------------------------\n","text: 甄加索是自杀吗\n","--------------------------------------------------\n","label: 不是\n","--------------------------------------------------\n","answer: nan\n","--------------------------------------------------\n","title: 海岸之谜\n","--------------------------------------------------\n","puzzle: 在远离城市喧嚣的海边小屋，一天清晨，邻居发现甄加索僵卧在沙滩上，已无生命迹象。现场没有发现任何打斗的迹象。请问甄加索的死因是什么？\n","--------------------------------------------------\n","truth: 甄加索是一位热爱自然的画家，他每年都会来到这个海边小屋寻找灵感。在他生命的最后几天，他一直在创作一幅描绘海洋生物的画作。在画即将完成的前一天晚上，他骑着自行车外出，打算在海边观赏夜景。然而，他在沙滩上意外发现了一只搁浅的海豚，为了救助这只海豚，他耗费了极大的体力，最终成功将其送回海中。筋疲力尽的甄加索在沙滩上睡着了，由于他患有严重的心脏病，却未告知旁人，在寒冷的海风中，他的心脏停止了跳动。因此，警方在现场只发现了车轮痕迹和未完成的画作，而没有发现任何他杀的迹象。\n","--------------------------------------------------\n","train_text: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在远离城市喧嚣的海边小屋，一天清晨，邻居发现甄加索僵卧在沙滩上，已无生命迹象。现场没有发现任何打斗的迹象。请问甄加索的死因是什么？\n","\n","**谜底:** 甄加索是一位热爱自然的画家，他每年都会来到这个海边小屋寻找灵感。在他生命的最后几天，他一直在创作一幅描绘海洋生物的画作。在画即将完成的前一天晚上，他骑着自行车外出，打算在海边观赏夜景。然而，他在沙滩上意外发现了一只搁浅的海豚，为了救助这只海豚，他耗费了极大的体力，最终成功将其送回海中。筋疲力尽的甄加索在沙滩上睡着了，由于他患有严重的心脏病，却未告知旁人，在寒冷的海风中，他的心脏停止了跳动。因此，警方在现场只发现了车轮痕迹和未完成的画作，而没有发现任何他杀的迹象。\n","\n","**参与者提出的问题:** 甄加索是自杀吗\n","<|im_end|>\n","<|im_start|>assistant\n","不是\n","--------------------------------------------------\n","prompt: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在远离城市喧嚣的海边小屋，一天清晨，邻居发现甄加索僵卧在沙滩上，已无生命迹象。现场没有发现任何打斗的迹象。请问甄加索的死因是什么？\n","\n","**谜底:** 甄加索是一位热爱自然的画家，他每年都会来到这个海边小屋寻找灵感。在他生命的最后几天，他一直在创作一幅描绘海洋生物的画作。在画即将完成的前一天晚上，他骑着自行车外出，打算在海边观赏夜景。然而，他在沙滩上意外发现了一只搁浅的海豚，为了救助这只海豚，他耗费了极大的体力，最终成功将其送回海中。筋疲力尽的甄加索在沙滩上睡着了，由于他患有严重的心脏病，却未告知旁人，在寒冷的海风中，他的心脏停止了跳动。因此，警方在现场只发现了车轮痕迹和未完成的画作，而没有发现任何他杀的迹象。\n","\n","**参与者提出的问题:** 甄加索是自杀吗\n","<|im_end|>\n","<|im_start|>assistant\n","\n","--------------------------------------------------\n","text: 死者受伤了吗\n","--------------------------------------------------\n","label: 不是\n","--------------------------------------------------\n","answer: nan\n","--------------------------------------------------\n","title: 甄庄哭声\n","--------------------------------------------------\n","puzzle: 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","--------------------------------------------------\n","truth: 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","--------------------------------------------------\n","train_text: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","\n","**谜底:** 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","\n","**参与者提出的问题:** 死者受伤了吗\n","<|im_end|>\n","<|im_start|>assistant\n","不是\n","--------------------------------------------------\n","prompt: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","\n","**谜底:** 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","\n","**参与者提出的问题:** 死者受伤了吗\n","<|im_end|>\n","<|im_start|>assistant\n","\n","Evaluating model: internlm/internlm2_5-7b-chat-1m\n"," 0%| | 0/3000 [00:00> loading file ./tokenizer.model from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/./tokenizer.model\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:42:34,236 >> loading file added_tokens.json from cache at None\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:42:34,236 >> loading file special_tokens_map.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/special_tokens_map.json\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:42:34,236 >> loading file tokenizer_config.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/tokenizer_config.json\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:42:34,236 >> loading file tokenizer.json from cache at None\n","07/22/2024 21:42:34 - INFO - llamafactory.data.template - Add <|im_end|> to stop words.\n","/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.\n"," warnings.warn(\n","[INFO|configuration_utils.py:733] 2024-07-22 21:42:34,806 >> loading configuration file config.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/config.json\n","[INFO|configuration_utils.py:733] 2024-07-22 21:42:35,328 >> loading configuration file config.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/config.json\n","[INFO|configuration_utils.py:796] 2024-07-22 21:42:35,328 >> Model config InternLM2Config {\n"," \"_name_or_path\": \"internlm/internlm2_5-7b-chat-1m\",\n"," \"architectures\": [\n"," \"InternLM2ForCausalLM\"\n"," ],\n"," \"attn_implementation\": \"eager\",\n"," \"auto_map\": {\n"," \"AutoConfig\": \"internlm/internlm2_5-7b-chat-1m--configuration_internlm2.InternLM2Config\",\n"," \"AutoModel\": \"internlm/internlm2_5-7b-chat-1m--modeling_internlm2.InternLM2ForCausalLM\",\n"," \"AutoModelForCausalLM\": \"internlm/internlm2_5-7b-chat-1m--modeling_internlm2.InternLM2ForCausalLM\"\n"," },\n"," \"bias\": false,\n"," \"bos_token_id\": 1,\n"," \"eos_token_id\": 2,\n"," \"hidden_act\": \"silu\",\n"," \"hidden_size\": 4096,\n"," \"initializer_range\": 0.02,\n"," \"intermediate_size\": 14336,\n"," \"max_position_embeddings\": 262144,\n"," \"model_type\": \"internlm2\",\n"," \"num_attention_heads\": 32,\n"," \"num_hidden_layers\": 32,\n"," \"num_key_value_heads\": 8,\n"," \"pad_token_id\": 2,\n"," \"pretraining_tp\": 1,\n"," \"rms_norm_eps\": 1e-05,\n"," \"rope_scaling\": {\n"," \"factor\": 2.5,\n"," \"type\": \"dynamic\"\n"," },\n"," \"rope_theta\": 50000000,\n"," \"tie_word_embeddings\": false,\n"," \"torch_dtype\": \"bfloat16\",\n"," \"transformers_version\": \"4.41.2\",\n"," \"use_cache\": true,\n"," \"vocab_size\": 92544\n","}\n","\n","07/22/2024 21:42:35 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.\n","[INFO|modeling_utils.py:3474] 2024-07-22 21:42:35,599 >> loading weights file model.safetensors from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/model.safetensors.index.json\n","[INFO|modeling_utils.py:1519] 2024-07-22 21:42:35,600 >> Instantiating InternLM2ForCausalLM model under default dtype torch.bfloat16.\n","[INFO|configuration_utils.py:962] 2024-07-22 21:42:35,600 >> Generate config GenerationConfig {\n"," \"bos_token_id\": 1,\n"," \"eos_token_id\": 2,\n"," \"pad_token_id\": 2\n","}\n","\n","Loading checkpoint shards: 100%|██████████████████| 8/8 [00:02<00:00, 2.69it/s]\n","[INFO|modeling_utils.py:4280] 2024-07-22 21:42:38,783 >> All model checkpoint weights were used when initializing InternLM2ForCausalLM.\n","\n","[INFO|modeling_utils.py:4288] 2024-07-22 21:42:38,783 >> All the weights of InternLM2ForCausalLM were initialized from the model checkpoint at internlm/internlm2_5-7b-chat-1m.\n","If your task is similar to the task the model of the checkpoint was trained on, you can already use InternLM2ForCausalLM for predictions without further training.\n","[INFO|configuration_utils.py:917] 2024-07-22 21:42:39,069 >> loading configuration file generation_config.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/generation_config.json\n","[INFO|configuration_utils.py:962] 2024-07-22 21:42:39,069 >> Generate config GenerationConfig {\n"," \"bos_token_id\": 1,\n"," \"eos_token_id\": [\n"," 2,\n"," 92542\n"," ],\n"," \"pad_token_id\": 2\n","}\n","\n","07/22/2024 21:42:39 - INFO - llamafactory.model.model_utils.attention - Using vanilla attention implementation.\n","07/22/2024 21:42:40 - INFO - llamafactory.model.adapter - Merged 1 adapter(s).\n","07/22/2024 21:42:40 - INFO - llamafactory.model.adapter - Loaded adapter(s): inflaton-ai/InternLM_2_5-7b_LoRA-Adapter\n","07/22/2024 21:42:40 - INFO - llamafactory.model.loader - all params: 7,737,708,544\n","(2) GPU = NVIDIA GeForce RTX 4090. Max memory = 23.988 GB.\n","14.818 GB of memory reserved.\n","loading train/test data files\n","DatasetDict({\n"," train: Dataset({\n"," features: ['text', 'label', 'answer', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n"," num_rows: 25000\n"," })\n"," test: Dataset({\n"," features: ['text', 'label', 'answer', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n"," num_rows: 3000\n"," })\n","})\n","--------------------------------------------------\n","text: 甄加索是自杀吗\n","--------------------------------------------------\n","label: 不是\n","--------------------------------------------------\n","answer: nan\n","--------------------------------------------------\n","title: 海岸之谜\n","--------------------------------------------------\n","puzzle: 在远离城市喧嚣的海边小屋，一天清晨，邻居发现甄加索僵卧在沙滩上，已无生命迹象。现场没有发现任何打斗的迹象。请问甄加索的死因是什么？\n","--------------------------------------------------\n","truth: 甄加索是一位热爱自然的画家，他每年都会来到这个海边小屋寻找灵感。在他生命的最后几天，他一直在创作一幅描绘海洋生物的画作。在画即将完成的前一天晚上，他骑着自行车外出，打算在海边观赏夜景。然而，他在沙滩上意外发现了一只搁浅的海豚，为了救助这只海豚，他耗费了极大的体力，最终成功将其送回海中。筋疲力尽的甄加索在沙滩上睡着了，由于他患有严重的心脏病，却未告知旁人，在寒冷的海风中，他的心脏停止了跳动。因此，警方在现场只发现了车轮痕迹和未完成的画作，而没有发现任何他杀的迹象。\n","--------------------------------------------------\n","train_text: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在远离城市喧嚣的海边小屋，一天清晨，邻居发现甄加索僵卧在沙滩上，已无生命迹象。现场没有发现任何打斗的迹象。请问甄加索的死因是什么？\n","\n","**谜底:** 甄加索是一位热爱自然的画家，他每年都会来到这个海边小屋寻找灵感。在他生命的最后几天，他一直在创作一幅描绘海洋生物的画作。在画即将完成的前一天晚上，他骑着自行车外出，打算在海边观赏夜景。然而，他在沙滩上意外发现了一只搁浅的海豚，为了救助这只海豚，他耗费了极大的体力，最终成功将其送回海中。筋疲力尽的甄加索在沙滩上睡着了，由于他患有严重的心脏病，却未告知旁人，在寒冷的海风中，他的心脏停止了跳动。因此，警方在现场只发现了车轮痕迹和未完成的画作，而没有发现任何他杀的迹象。\n","\n","**参与者提出的问题:** 甄加索是自杀吗\n","<|im_end|>\n","<|im_start|>assistant\n","不是\n","--------------------------------------------------\n","prompt: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在远离城市喧嚣的海边小屋，一天清晨，邻居发现甄加索僵卧在沙滩上，已无生命迹象。现场没有发现任何打斗的迹象。请问甄加索的死因是什么？\n","\n","**谜底:** 甄加索是一位热爱自然的画家，他每年都会来到这个海边小屋寻找灵感。在他生命的最后几天，他一直在创作一幅描绘海洋生物的画作。在画即将完成的前一天晚上，他骑着自行车外出，打算在海边观赏夜景。然而，他在沙滩上意外发现了一只搁浅的海豚，为了救助这只海豚，他耗费了极大的体力，最终成功将其送回海中。筋疲力尽的甄加索在沙滩上睡着了，由于他患有严重的心脏病，却未告知旁人，在寒冷的海风中，他的心脏停止了跳动。因此，警方在现场只发现了车轮痕迹和未完成的画作，而没有发现任何他杀的迹象。\n","\n","**参与者提出的问题:** 甄加索是自杀吗\n","<|im_end|>\n","<|im_start|>assistant\n","\n","--------------------------------------------------\n","text: 死者受伤了吗\n","--------------------------------------------------\n","label: 不是\n","--------------------------------------------------\n","answer: nan\n","--------------------------------------------------\n","title: 甄庄哭声\n","--------------------------------------------------\n","puzzle: 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","--------------------------------------------------\n","truth: 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","--------------------------------------------------\n","train_text: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","\n","**谜底:** 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","\n","**参与者提出的问题:** 死者受伤了吗\n","<|im_end|>\n","<|im_start|>assistant\n","不是\n","--------------------------------------------------\n","prompt: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","\n","**谜底:** 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","\n","**参与者提出的问题:** 死者受伤了吗\n","<|im_end|>\n","<|im_start|>assistant\n","\n","Evaluating model: internlm/internlm2_5-7b-chat-1m\n"," 0%| | 0/3000 [00:00> loading file ./tokenizer.model from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/./tokenizer.model\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:50:25,126 >> loading file added_tokens.json from cache at None\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:50:25,126 >> loading file special_tokens_map.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/special_tokens_map.json\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:50:25,126 >> loading file tokenizer_config.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/tokenizer_config.json\n","[INFO|tokenization_utils_base.py:2108] 2024-07-22 21:50:25,126 >> loading file tokenizer.json from cache at None\n","07/22/2024 21:50:25 - INFO - llamafactory.data.template - Add <|im_end|> to stop words.\n","/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.\n"," warnings.warn(\n","[INFO|configuration_utils.py:733] 2024-07-22 21:50:25,733 >> loading configuration file config.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/config.json\n","[INFO|configuration_utils.py:733] 2024-07-22 21:50:26,241 >> loading configuration file config.json from cache at /home/inflaton/.cache/huggingface/hub/models--internlm--internlm2_5-7b-chat-1m/snapshots/8d1a709a04d71440ef3df6ebbe204672f411c8b6/config.json\n","[INFO|configuration_utils.py:796] 2024-07-22 21:50:26,242 >> Model config InternLM2Config {\n"," \"_name_or_path\": \"internlm/internlm2_5-7b-chat-1m\",\n"," \"architectures\": [\n"," \"InternLM2ForCausalLM\"\n"," ],\n"," \"attn_implementation\": \"eager\",\n"," \"auto_map\": {\n"," \"AutoConfig\": \"internlm/internlm2_5-7b-chat-1m--configuration_internlm2.InternLM2Config\",\n"," \"AutoModel\": \"internlm/internlm2_5-7b-chat-1m--modeling_internlm2.InternLM2ForCausalLM\",\n"," \"AutoModelForCausalLM\": \"internlm/internlm2_5-7b-chat-1m--modeling_internlm2.InternLM2ForCausalLM\"\n"," },\n"," \"bias\": false,\n"," \"bos_token_id\": 1,\n"," \"eos_token_id\": 2,\n"," \"hidden_act\": \"silu\",\n"," \"hidden_size\": 4096,\n"," \"initializer_range\": 0.02,\n"," \"intermediate_size\": 14336,\n"," \"max_position_embeddings\": 262144,\n"," \"model_type\": \"internlm2\",\n"," \"num_attention_heads\": 32,\n"," \"num_hidden_layers\": 32,\n"," \"num_key_value_heads\": 8,\n"," \"pad_token_id\": 2,\n"," \"pretraining_tp\": 1,\n"," \"rms_norm_eps\": 1e-05,\n"," \"rope_scaling\": {\n"," \"factor\": 2.5,\n"," \"type\": \"dynamic\"\n"," },\n"," \"rope_theta\": 50000000,\n"," \"tie_word_embeddings\": false,\n"," \"torch_dtype\": \"bfloat16\",\n"," \"transformers_version\": \"4.41.2\",\n"," \"use_cache\": true,\n"," \"vocab_size\": 92544\n","}\n","\n","Traceback (most recent call last):\n"," File \"/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/importlib/metadata/__init__.py\", line 563, in from_name\n"," return next(cls.discover(name=name))\n"," ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n","StopIteration\n","\n","During handling of the above exception, another exception occurred:\n","\n","Traceback (most recent call last):\n"," File \"/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/site-packages/transformers/utils/versions.py\", line 102, in require_version\n"," got_ver = importlib.metadata.version(pkg)\n"," ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n"," File \"/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/importlib/metadata/__init__.py\", line 1009, in version\n"," return distribution(distribution_name).version\n"," ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n"," File \"/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/importlib/metadata/__init__.py\", line 982, in distribution\n"," return Distribution.from_name(distribution_name)\n"," ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n"," File \"/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/importlib/metadata/__init__.py\", line 565, in from_name\n"," raise PackageNotFoundError(name)\n","importlib.metadata.PackageNotFoundError: No package metadata was found for bitsandbytes\n","\n","During handling of the above exception, another exception occurred:\n","\n","Traceback (most recent call last):\n"," File \"/home/inflaton/code/logical-reasoning/llm_toolkit/eval_logical_reasoning.py\", line 49, in \n"," model, tokenizer = load_model(\n"," ^^^^^^^^^^^\n"," File \"/home/inflaton/code/logical-reasoning/llm_toolkit/llm_utils.py\", line 45, in load_model\n"," chat_model = ChatModel(args)\n"," ^^^^^^^^^^^^^^^\n"," File \"/home/inflaton/code/LLaMA-Factory/src/llamafactory/chat/chat_model.py\", line 44, in __init__\n"," self.engine: \"BaseEngine\" = HuggingfaceEngine(model_args, data_args, finetuning_args, generating_args)\n"," ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n"," File \"/home/inflaton/code/LLaMA-Factory/src/llamafactory/chat/hf_engine.py\", line 58, in __init__\n"," self.model = load_model(\n"," ^^^^^^^^^^^\n"," File \"/home/inflaton/code/LLaMA-Factory/src/llamafactory/model/loader.py\", line 132, in load_model\n"," patch_config(config, tokenizer, model_args, init_kwargs, is_trainable)\n"," File \"/home/inflaton/code/LLaMA-Factory/src/llamafactory/model/patcher.py\", line 75, in patch_config\n"," configure_quantization(config, tokenizer, model_args, init_kwargs)\n"," File \"/home/inflaton/code/LLaMA-Factory/src/llamafactory/model/model_utils/quantization.py\", line 160, in configure_quantization\n"," require_version(\"bitsandbytes>=0.39.0\", \"To fix: pip install bitsandbytes>=0.39.0\")\n"," File \"/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/site-packages/transformers/utils/versions.py\", line 104, in require_version\n"," raise importlib.metadata.PackageNotFoundError(\n","importlib.metadata.PackageNotFoundError: No package metadata was found for The 'bitsandbytes>=0.39.0' distribution was not found and is required by this application. \n","To fix: pip install bitsandbytes>=0.39.0\n","CPU times: user 15.4 s, sys: 5.13 s, total: 20.5 s\n","Wall time: 39min 43s\n"]}],"source":["%%time\n","\n","os.environ[\"USING_LLAMA_FACTORY\"] = \"true\"\n","!./scripts/eval-mgtv-nv4090.sh"]},{"cell_type":"code","execution_count":16,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["Current Directory:\n","/home/inflaton/code/logical-reasoning\n","Mon Jul 22 21:50:27 2024 \n","+---------------------------------------------------------------------------------------+\n","| NVIDIA-SMI 535.54.04 Driver Version: 536.25 CUDA Version: 12.2 |\n","|-----------------------------------------+----------------------+----------------------+\n","| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |\n","| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |\n","| | | MIG M. |\n","|=========================================+======================+======================|\n","| 0 NVIDIA GeForce RTX 4090 On | 00000000:01:00.0 Off | Off |\n","| 48% 55C P8 19W / 450W | 1460MiB / 24564MiB | 3% Default |\n","| | | N/A |\n","+-----------------------------------------+----------------------+----------------------+\n"," \n","+---------------------------------------------------------------------------------------+\n","| Processes: |\n","| GPU GI CI PID Type Process name GPU Memory |\n","| ID ID Usage |\n","|=======================================================================================|\n","| No running processes found |\n","+---------------------------------------------------------------------------------------+\n","Linux Gen-AI 5.15.133.1-microsoft-standard-WSL2 #1 SMP Thu Oct 5 21:02:42 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux\n","PRETTY_NAME=\"Ubuntu 22.04.2 LTS\"\n","NAME=\"Ubuntu\"\n","VERSION_ID=\"22.04\"\n","VERSION=\"22.04.2 LTS (Jammy Jellyfish)\"\n","VERSION_CODENAME=jammy\n","ID=ubuntu\n","ID_LIKE=debian\n","HOME_URL=\"https://www.ubuntu.com/\"\n","SUPPORT_URL=\"https://help.ubuntu.com/\"\n","BUG_REPORT_URL=\"https://bugs.launchpad.net/ubuntu/\"\n","PRIVACY_POLICY_URL=\"https://www.ubuntu.com/legal/terms-and-policies/privacy-policy\"\n","UBUNTU_CODENAME=jammy\n","Architecture: x86_64\n"," CPU op-mode(s): 32-bit, 64-bit\n"," Address sizes: 39 bits physical, 48 bits virtual\n"," Byte Order: Little Endian\n","CPU(s): 32\n"," On-line CPU(s) list: 0-31\n","Vendor ID: GenuineIntel\n"," Model name: 13th Gen Intel(R) Core(TM) i9-13900KF\n"," CPU family: 6\n"," Model: 183\n"," Thread(s) per core: 2\n"," Core(s) per socket: 16\n"," Socket(s): 1\n"," Stepping: 1\n"," BogoMIPS: 5990.40\n"," Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mc\n"," a cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscal\n"," l nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopo\n"," logy tsc_reliable nonstop_tsc cpuid pni pclmulqdq vmx s\n"," sse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt tsc_dea\n"," dline_timer aes xsave avx f16c rdrand hypervisor lahf_l\n"," m abm 3dnowprefetch ssbd ibrs ibpb stibp ibrs_enhanced \n"," tpr_shadow vnmi ept vpid ept_ad fsgsbase tsc_adjust bmi\n"," 1 avx2 smep bmi2 erms invpcid rdseed adx smap clflushop\n"," t clwb sha_ni xsaveopt xsavec xgetbv1 xsaves avx_vnni u\n"," mip waitpkg gfni vaes vpclmulqdq rdpid movdiri movdir64\n"," b fsrm md_clear serialize flush_l1d arch_capabilities\n","Virtualization features: \n"," Virtualization: VT-x\n"," Hypervisor vendor: Microsoft\n"," Virtualization type: full\n","Caches (sum of all): \n"," L1d: 768 KiB (16 instances)\n"," L1i: 512 KiB (16 instances)\n"," L2: 32 MiB (16 instances)\n"," L3: 36 MiB (1 instance)\n","Vulnerabilities: \n"," Gather data sampling: Not affected\n"," Itlb multihit: Not affected\n"," L1tf: Not affected\n"," Mds: Not affected\n"," Meltdown: Not affected\n"," Mmio stale data: Not affected\n"," Retbleed: Mitigation; Enhanced IBRS\n"," Spec rstack overflow: Not affected\n"," Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl\n"," and seccomp\n"," Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer\n"," sanitization\n"," Spectre v2: Mitigation; Enhanced IBRS, IBPB conditional, RSB fillin\n"," g, PBRSB-eIBRS SW sequence\n"," Srbds: Not affected\n"," Tsx async abort: Not affected\n","MemTotal: 49330020 kB\n","Eval internlm/internlm2_5-7b-chat-1m with inflaton-ai/InternLM_2_5-7b_LoRA-Adapter\n","loading env vars from: /home/inflaton/code/logical-reasoning/.env\n","Adding /home/inflaton/code/logical-reasoning to sys.path\n","loading /home/inflaton/code/logical-reasoning/llm_toolkit/logical_reasoning_utils.py\n","internlm/internlm2_5-7b-chat-1m inflaton-ai/InternLM_2_5-7b_LoRA-Adapter False datasets/mgtv results/mgtv-results_internlm_best.csv\n","(1) GPU = NVIDIA GeForce RTX 4090. Max memory = 23.988 GB.\n","0.0 GB of memory reserved.\n","loading model: internlm/internlm2_5-7b-chat-1m with adapter: inflaton-ai/InternLM_2_5-7b_LoRA-Adapter\n","Loading checkpoint shards: 100%|██████████████████| 8/8 [00:03<00:00, 2.56it/s]\n","(2) GPU = NVIDIA GeForce RTX 4090. Max memory = 23.988 GB.\n","14.486 GB of memory reserved.\n","loading train/test data files\n","Map: 100%|██████████████████████| 25000/25000 [00:00<00:00, 64711.02 examples/s]\n","Map: 100%|████████████████████████| 3000/3000 [00:00<00:00, 62306.45 examples/s]\n","DatasetDict({\n"," train: Dataset({\n"," features: ['text', 'label', 'answer', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n"," num_rows: 25000\n"," })\n"," test: Dataset({\n"," features: ['text', 'label', 'answer', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n"," num_rows: 3000\n"," })\n","})\n","--------------------------------------------------\n","text: 甄加索是自杀吗\n","--------------------------------------------------\n","label: 不是\n","--------------------------------------------------\n","answer: nan\n","--------------------------------------------------\n","title: 海岸之谜\n","--------------------------------------------------\n","puzzle: 在远离城市喧嚣的海边小屋，一天清晨，邻居发现甄加索僵卧在沙滩上，已无生命迹象。现场没有发现任何打斗的迹象。请问甄加索的死因是什么？\n","--------------------------------------------------\n","truth: 甄加索是一位热爱自然的画家，他每年都会来到这个海边小屋寻找灵感。在他生命的最后几天，他一直在创作一幅描绘海洋生物的画作。在画即将完成的前一天晚上，他骑着自行车外出，打算在海边观赏夜景。然而，他在沙滩上意外发现了一只搁浅的海豚，为了救助这只海豚，他耗费了极大的体力，最终成功将其送回海中。筋疲力尽的甄加索在沙滩上睡着了，由于他患有严重的心脏病，却未告知旁人，在寒冷的海风中，他的心脏停止了跳动。因此，警方在现场只发现了车轮痕迹和未完成的画作，而没有发现任何他杀的迹象。\n","--------------------------------------------------\n","train_text: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在远离城市喧嚣的海边小屋，一天清晨，邻居发现甄加索僵卧在沙滩上，已无生命迹象。现场没有发现任何打斗的迹象。请问甄加索的死因是什么？\n","\n","**谜底:** 甄加索是一位热爱自然的画家，他每年都会来到这个海边小屋寻找灵感。在他生命的最后几天，他一直在创作一幅描绘海洋生物的画作。在画即将完成的前一天晚上，他骑着自行车外出，打算在海边观赏夜景。然而，他在沙滩上意外发现了一只搁浅的海豚，为了救助这只海豚，他耗费了极大的体力，最终成功将其送回海中。筋疲力尽的甄加索在沙滩上睡着了，由于他患有严重的心脏病，却未告知旁人，在寒冷的海风中，他的心脏停止了跳动。因此，警方在现场只发现了车轮痕迹和未完成的画作，而没有发现任何他杀的迹象。\n","\n","**参与者提出的问题:** 甄加索是自杀吗\n","<|im_end|>\n","<|im_start|>assistant\n","不是\n","--------------------------------------------------\n","prompt: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在远离城市喧嚣的海边小屋，一天清晨，邻居发现甄加索僵卧在沙滩上，已无生命迹象。现场没有发现任何打斗的迹象。请问甄加索的死因是什么？\n","\n","**谜底:** 甄加索是一位热爱自然的画家，他每年都会来到这个海边小屋寻找灵感。在他生命的最后几天，他一直在创作一幅描绘海洋生物的画作。在画即将完成的前一天晚上，他骑着自行车外出，打算在海边观赏夜景。然而，他在沙滩上意外发现了一只搁浅的海豚，为了救助这只海豚，他耗费了极大的体力，最终成功将其送回海中。筋疲力尽的甄加索在沙滩上睡着了，由于他患有严重的心脏病，却未告知旁人，在寒冷的海风中，他的心脏停止了跳动。因此，警方在现场只发现了车轮痕迹和未完成的画作，而没有发现任何他杀的迹象。\n","\n","**参与者提出的问题:** 甄加索是自杀吗\n","<|im_end|>\n","<|im_start|>assistant\n","\n","--------------------------------------------------\n","text: 死者受伤了吗\n","--------------------------------------------------\n","label: 不是\n","--------------------------------------------------\n","answer: nan\n","--------------------------------------------------\n","title: 甄庄哭声\n","--------------------------------------------------\n","puzzle: 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","--------------------------------------------------\n","truth: 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","--------------------------------------------------\n","train_text: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","\n","**谜底:** 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","\n","**参与者提出的问题:** 死者受伤了吗\n","<|im_end|>\n","<|im_start|>assistant\n","不是\n","--------------------------------------------------\n","prompt: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","\n","**谜底:** 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","\n","**参与者提出的问题:** 死者受伤了吗\n","<|im_end|>\n","<|im_start|>assistant\n","\n","Evaluating model: internlm/internlm2_5-7b-chat-1m\n"," 0%| | 0/3000 [00:00<|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在远离城市喧嚣的海边小屋，一天清晨，邻居发现甄加索僵卧在沙滩上，已无生命迹象。现场没有发现任何打斗的迹象。请问甄加索的死因是什么？\n","\n","**谜底:** 甄加索是一位热爱自然的画家，他每年都会来到这个海边小屋寻找灵感。在他生命的最后几天，他一直在创作一幅描绘海洋生物的画作。在画即将完成的前一天晚上，他骑着自行车外出，打算在海边观赏夜景。然而，他在沙滩上意外发现了一只搁浅的海豚，为了救助这只海豚，他耗费了极大的体力，最终成功将其送回海中。筋疲力尽的甄加索在沙滩上睡着了，由于他患有严重的心脏病，却未告知旁人，在寒冷的海风中，他的心脏停止了跳动。因此，警方在现场只发现了车轮痕迹和未完成的画作，而没有发现任何他杀的迹象。\n","\n","**参与者提出的问题:** 甄加索是自杀吗\n","<|im_end|>\n","<|im_start|>assistant\n","不是\n","--------------------------------------------------\n","prompt: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在远离城市喧嚣的海边小屋，一天清晨，邻居发现甄加索僵卧在沙滩上，已无生命迹象。现场没有发现任何打斗的迹象。请问甄加索的死因是什么？\n","\n","**谜底:** 甄加索是一位热爱自然的画家，他每年都会来到这个海边小屋寻找灵感。在他生命的最后几天，他一直在创作一幅描绘海洋生物的画作。在画即将完成的前一天晚上，他骑着自行车外出，打算在海边观赏夜景。然而，他在沙滩上意外发现了一只搁浅的海豚，为了救助这只海豚，他耗费了极大的体力，最终成功将其送回海中。筋疲力尽的甄加索在沙滩上睡着了，由于他患有严重的心脏病，却未告知旁人，在寒冷的海风中，他的心脏停止了跳动。因此，警方在现场只发现了车轮痕迹和未完成的画作，而没有发现任何他杀的迹象。\n","\n","**参与者提出的问题:** 甄加索是自杀吗\n","<|im_end|>\n","<|im_start|>assistant\n","\n","--------------------------------------------------\n","text: 死者受伤了吗\n","--------------------------------------------------\n","label: 不是\n","--------------------------------------------------\n","answer: nan\n","--------------------------------------------------\n","title: 甄庄哭声\n","--------------------------------------------------\n","puzzle: 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","--------------------------------------------------\n","truth: 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","--------------------------------------------------\n","train_text: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","\n","**谜底:** 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","\n","**参与者提出的问题:** 死者受伤了吗\n","<|im_end|>\n","<|im_start|>assistant\n","不是\n","--------------------------------------------------\n","prompt: <|im_start|>system\n","You are an expert in logical reasoning.<|im_end|>\n","<|im_start|>user\n","你是一个情景猜谜游戏的主持人。游戏规则如下：\n","\n","1. 参与者会得到一个谜面，谜面会描述一个简单又难以理解的事件。\n","2. 主持人知道谜底，谜底是谜面的答案。\n","3. 参与者可以询问任何封闭式问题来找寻事件的真相。\n","4. 对于每个问题，主持人将根据实际情况回答以下五个选项之一：是、不是、不重要、回答正确、问法错误。各回答的判断标准如下：\n"," - 若谜面和谜底能找到问题的答案，回答：是或者不是\n"," - 若谜面和谜底不能直接或者间接推断出问题的答案，回答：不重要\n"," - 若参与者提问不是一个封闭式问题或者问题难以理解，回答：问法错误\n"," - 若参与者提问基本还原了谜底真相，回答：回答正确\n","5. 回答中不能添加任何其它信息，也不能省略选项中的任何一个字。例如，不可以把“不是”省略成“不”。\n","\n","请严格按照这些规则回答参与者提出的问题。\n","\n","**谜面:** 在一个安静的夜晚，小村庄的湖边突然传来了阵阵哭泣声。第二天早晨，村长甄锐发现湖边的石头上放着一顶破旧的帽子，但没有人知道这顶帽子是从哪里来的，哭泣声又是为何。请还原故事真相。\n","\n","**谜底:** 原来，这顶破旧的帽子属于一个小男孩，他小时候与爷爷在湖边生活。爷爷教他钓鱼、游泳，还告诉他湖中的海龟是他们的朋友。后来，小男孩随父母去了城市生活，但每年夏天都会回到村子探望爷爷。然而，去年夏天，爷爷因病去世，小男孩伤心欲绝。今年夏天，他回到村子，来到湖边，想起和爷爷的美好回忆，忍不住哭泣。他将爷爷的帽子放在湖边的石头上，希望能让爷爷的在天之灵得到安慰。那晚的哭泣声正是小男孩在祭莫他亲爱的爷爷。\n","\n","**参与者提出的问题:** 死者受伤了吗\n","<|im_end|>\n","<|im_start|>assistant\n","\n","Evaluating model: internlm/internlm2_5-7b-chat-1m\n"," 0%| | 0/3000 [00:00\n"," model, tokenizer = load_model(\n"," ^^^^^^^^^^^\n"," File \"/home/inflaton/code/logical-reasoning/llm_toolkit/llm_utils.py\", line 53, in load_model\n"," bnb_config = BitsAndBytesConfig(\n"," ^^^^^^^^^^^^^^^^^^^\n"," File \"/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/site-packages/transformers/utils/quantization_config.py\", line 395, in __init__\n"," self.post_init()\n"," File \"/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/site-packages/transformers/utils/quantization_config.py\", line 453, in post_init\n"," if self.load_in_4bit and not version.parse(importlib.metadata.version(\"bitsandbytes\")) >= version.parse(\n"," ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n"," File \"/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/importlib/metadata/__init__.py\", line 1009, in version\n"," return distribution(distribution_name).version\n"," ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n"," File \"/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/importlib/metadata/__init__.py\", line 982, in distribution\n"," return Distribution.from_name(distribution_name)\n"," ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n"," File \"/home/inflaton/miniconda3/envs/llm-finetuning/lib/python3.11/importlib/metadata/__init__.py\", line 565, in from_name\n"," raise PackageNotFoundError(name)\n","importlib.metadata.PackageNotFoundError: No package metadata was found for bitsandbytes\n","CPU times: user 7.47 s, sys: 2.41 s, total: 9.88 s\n","Wall time: 17min 13s\n"]}],"source":["%%time\n","\n","os.environ[\"USING_LLAMA_FACTORY\"] = \"false\"\n","!./scripts/eval-mgtv-nv4090.sh"]}],"metadata":{"accelerator":"GPU","application/vnd.databricks.v1+notebook":{"dashboards":[],"environmentMetadata":null,"language":"python","notebookMetadata":{"mostRecentlyExecutedCommandWithImplicitDF":{"commandId":-1,"dataframes":["_sqldf"]},"pythonIndentUnit":4},"notebookName":"10_eval-lf-medium-py3.11","widgets":{}},"colab":{"gpuType":"L4","provenance":[]},"kernelspec":{"display_name":"Python 3","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.11.9"}},"nbformat":4,"nbformat_minor":0}