{ "cells": [ { "cell_type": "code", "execution_count": null, "metadata": { "id": "Bj6c3-Kn2fm0" }, "outputs": [], "source": [ "%%capture\n", "import os, re\n", "if \"COLAB_\" not in \"\".join(os.environ.keys()):\n", " !pip install unsloth\n", "else:\n", " # Do this only in Colab notebooks! Otherwise use pip install unsloth\n", " import torch; v = re.match(r\"[0-9\\.]{3,}\", str(torch.__version__)).group(0)\n", " xformers = \"xformers==\" + (\"0.0.32.post2\" if v == \"2.8.0\" else \"0.0.29.post3\")\n", " !pip install --no-deps bitsandbytes accelerate {xformers} peft trl triton cut_cross_entropy unsloth_zoo\n", " !pip install sentencepiece protobuf \"datasets>=3.4.1,<4.0.0\" \"huggingface_hub>=0.34.0\" hf_transfer\n", " !pip install --no-deps unsloth\n", "!pip install transformers==4.55.4\n", "import torch; torch._dynamo.config.recompile_limit = 64;\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "k1HtbA3f2vCo", "outputId": "cd33e5b7-f56e-472c-9298-5889bf9090b3" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "πŸ¦₯ Unsloth: Will patch your computer to enable 2x faster free finetuning.\n", "πŸ¦₯ Unsloth Zoo will now patch everything to make training faster!\n" ] } ], "source": [ "import unsloth" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "IF6VdNcw9z-F", "outputId": "f899588a-cd2a-4f81-a623-31b6ee94d9b2" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\u001b[33m⚠️ Warning: 'huggingface-cli login' is deprecated. Use 'hf auth login' instead.\u001b[0m\n", "The token has not been saved to the git credentials helper. Pass `add_to_git_credential=True` in this function directly or `--add-to-git-credential` if using via `hf`CLI if you want to set the git credential as well.\n", "Token is valid (permission: fineGrained).\n", "The token `Toadlivesmatter1981` has been saved to /root/.cache/huggingface/stored_tokens\n", "Your token has been saved to /root/.cache/huggingface/token\n", "Login successful.\n", "The current active token is: `Toadlivesmatter1981`\n" ] } ], "source": [ "import torch\n", "!huggingface-cli login --token #your token here" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "9A90MibIanc1" }, "outputs": [], "source": [ "import torch\n", "from datasets import load_dataset, DatasetDict, Dataset\n", "from peft import get_peft_model, LoraConfig, TaskType, PeftModel\n", "from peft import prepare_model_for_kbit_training\n", "from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig\n", "import bitsandbytes" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "9sG7xqPT2byV", "outputId": "34070417-72ca-4520-ce3b-9a5e996f566b" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: bitsandbytes in /usr/local/lib/python3.12/dist-packages (0.48.1)\n", "Requirement already satisfied: torch<3,>=2.3 in /usr/local/lib/python3.12/dist-packages (from bitsandbytes) (2.8.0+cu126)\n", "Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.12/dist-packages (from bitsandbytes) (2.0.2)\n", "Requirement already satisfied: packaging>=20.9 in /usr/local/lib/python3.12/dist-packages (from bitsandbytes) (25.0)\n", "Requirement already satisfied: filelock in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (3.20.0)\n", "Requirement already satisfied: typing-extensions>=4.10.0 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (4.15.0)\n", "Requirement already satisfied: setuptools in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (75.2.0)\n", "Requirement already satisfied: sympy>=1.13.3 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (1.13.3)\n", "Requirement already satisfied: networkx in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (3.5)\n", "Requirement already satisfied: jinja2 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (3.1.6)\n", "Requirement already satisfied: fsspec in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (2025.3.0)\n", "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.6.77 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (12.6.77)\n", "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.6.77 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (12.6.77)\n", "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.6.80 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (12.6.80)\n", "Requirement already satisfied: nvidia-cudnn-cu12==9.10.2.21 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (9.10.2.21)\n", "Requirement already satisfied: nvidia-cublas-cu12==12.6.4.1 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (12.6.4.1)\n", "Requirement already satisfied: nvidia-cufft-cu12==11.3.0.4 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (11.3.0.4)\n", "Requirement already satisfied: nvidia-curand-cu12==10.3.7.77 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (10.3.7.77)\n", "Requirement already satisfied: nvidia-cusolver-cu12==11.7.1.2 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (11.7.1.2)\n", "Requirement already satisfied: nvidia-cusparse-cu12==12.5.4.2 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (12.5.4.2)\n", "Requirement already satisfied: nvidia-cusparselt-cu12==0.7.1 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (0.7.1)\n", "Requirement already satisfied: nvidia-nccl-cu12==2.27.3 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (2.27.3)\n", "Requirement already satisfied: nvidia-nvtx-cu12==12.6.77 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (12.6.77)\n", "Requirement already satisfied: nvidia-nvjitlink-cu12==12.6.85 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (12.6.85)\n", "Requirement already satisfied: nvidia-cufile-cu12==1.11.1.6 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (1.11.1.6)\n", "Requirement already satisfied: triton==3.4.0 in /usr/local/lib/python3.12/dist-packages (from torch<3,>=2.3->bitsandbytes) (3.4.0)\n", "Requirement already satisfied: mpmath<1.4,>=1.1.0 in /usr/local/lib/python3.12/dist-packages (from sympy>=1.13.3->torch<3,>=2.3->bitsandbytes) (1.3.0)\n", "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.12/dist-packages (from jinja2->torch<3,>=2.3->bitsandbytes) (3.0.3)\n" ] } ], "source": [ "!pip install -U bitsandbytes" ] }, { "cell_type": "markdown", "metadata": { "id": "FHvF4etyyQ92" }, "source": [ "### Installation" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "kvDtwWESDlQ7" }, "outputs": [], "source": [ "import os\n", "import torch\n", "\n", "# Disable ALL compilation paths\n", "os.environ[\"UNSLOTH_COMPILE\"] = \"0\"\n", "os.environ[\"TORCH_COMPILE\"] = \"0\" # Add this\n", "os.environ[\"TORCHINDUCTOR_DISABLE\"] = \"1\" # Add this\n", "torch._dynamo.config.disable = True\n", "torch._dynamo.config.suppress_errors = True # Add this\n", "torch.backends.cuda.enable_flash_sdp(False)" ] }, { "cell_type": "markdown", "metadata": { "id": "ZkH_y8UC9lvv" }, "source": [ "### Unsloth" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 533, "referenced_widgets": [ "c46bf3fd6a6842a99760a5db5490eee5", "0ede2a31d9324af0a7ef7353bdfd47d5", "ae3ebabe00974ff890e291949498a4f1", "2bcc23d9a99f48d99384f13c5d03b4a2", "5c621ad7d7df4fb5bfaa06557b53bb97", "1b3a3c391b8a40ca80432e25e035bdc1", "3b1e7e21535c49c1adc35fc973528608", "8b2a37e18365467193f7c80daa5994c2", "a52dae64336940d9b8265b8ef75e0d8e", "15ee8d6996fc4500a98b848c71788d0f", "1b903601b3ef4b19a62139545cbfa9f8", "c8d15adeebdc4412ac8faf41c20f6d09", "54b03abf49e34dd3bfba662d43587a67", "1da134804b954afaa627e66b54d94067", "fe9166cf7cb746e3829782204acb3c58", "e751ddbe6d9b47e685751cb38c936df3", "5597babb85514691a45878ca1027a2c7", "462acf69e5af4b9b9568828f012826c5", "fe78bb625fc54493997c7c40907b1332", "88575dddf91e4ab6af83276187131515", "936cdac923aa478191b6b42b4932e0bb", "cfb6fb245a7142e19e92abf10d70a7ba", "da4dfbcd0c824a3e85b8d7d9a0a792c7", "a6f415b33af3481fa69d772d8fd8c8b5", "22e1f83f93fb4f5cb6b936c52333d566", "5c0b39b905b04fc5ae233add3b255010", "a6c4047a11634f57bfd9af475dc45160", "95c27932de1a4b6896fa21787ad02a8b", "4dcfbbe5bc174c489f7768b992a21a06", "c10e74dd628c4695a3ea1fc8984ec6be", "af535dc786bc4e99aef024df9abb4cf9", "04c1a1a9c5f0496eba006b031aa97c5a", "b641e61c402443e6b706670612b08c72", "f219e5dd218f406f917925d4cffb8ba0", "8cbab7ef53d74a1796a59e43c29b20d4", "8b031178f746458a9af55fce18597091", "3703dbbf7eaf4e75bdeb348cf420f0a3", "5349abff12324bda8420d94e74634435", "855c42250b2240298bf884aea29c7335", "c273dd0506544fa4966d75c4621c9ab5", "9cb4983df1fa4cbf8f3d6914d2863437", "32ec2ff99acf4c049bce8eed53eb6857", "29313cac7f9c4092b3951225c6c5f549", "4f00a68928144c92a22955d0ca9587eb", "dcdf8548e5944c2e813c84f91e17bb39", "477262afa0124458b4c61507ae0bf66a", "9a1bea9b23a9441ca49c00ec5704ea38", "ea3b91d795824ab98304171fd4f7f1f5", "759812d5fb8049c6b32fbabe26ccb02c", "269f215e74b744fa907ac0259c0b83bd", "6b3ec11d107d4aa1969417c63b8148c9", "51912864e5a84c24aab93103f4943500", "e99d5492711840448fb67fe20079ae82", "86010db241c84cf6a2442eeaee84541b", "24caaad308d54b6c80ce3d1a994b8cda", "943821f3327c47d5af152d00aceb787f", "4611411bc934456689091f602cbfa24d", "3fb6aa95a7ba4a3f848738a729f4a80c", "597bbc35d05d4578a994d0e4c7abe4e7", "ef5e27b7dc9d4df987f7eaa89cabad24", "491852f7d0254ae3865c72cd8a5e0069", "040b1ef2563c420ea7ae80e7f05f9a30", "8010ba192dad432594ba8993ae424cf5", "f2d74b5746ba4240b1366456a6a99bd3", "b1f7850026ba462f949b10acb7ec8365", "88c38e3712df4835946cff5ae1cddbca", "046e575d4d23423288d4937ab512b723", "137d2ad435ae490abb9783fbcc4278c6", "e63dbdf961894e72b0f95daaf90ad4f1", "393c804920ae443ba1012e5bc9f7b06d", "27c51c59bf7a425799cbca39c06ac561", "baac44df202c499db3887f41401a4f85", "77ae25868f2342a49a370791d58dfef7", "bbce7e45ab764cd1996c5d4c38de22be", "af742823a63d43cfa617d108586b5c9f", "dae7e90e42054612b7e4aea3fc0d1200", "34f5e9fa26744b83b64f97c39c0dd027", "f3ef233e29394cc29013141e57ec9e9e", "9b47691ad9bd4014bae9f13902adc3c1", "00d3ec106184440e83ea25e9712c4902", "0f92d72f0f43450cbb06dd8db4747a94", "6c843db3502840208e5d887f41df8b66", "a87a7523660f43d9b948c2362fb38881", "0aeb643a4725471d9e609db822bd746b", "abda05bf0bcb4443ac75f98a9f199280", "4c06c837e65c402b8a35ca816e031d94", "8b5f195cbae942c199b2907897ba2df6", "c4c9d0595c0f4c1ca847f2ea30a3a945", "c09138002c4b4403a92b627d1e9105f0", "011ff3d1e294438684903401f36295b1", "e55ba41a840149f6b2d5ebdbdf5f4d50", "47e792f493c84f008767b04975e0badb", "e94366d8726d4fef98cc8b5a1aef683a", "f64f0280114a4a4b8985a36aa600cf9d", "262a0f5b409b478d8e1df257ce2df754", "38ea1fd58a114549bb2d537d3f85ed01", "2ab3d6f6b35e4d8aa4fb4964affbe31f", "93e8b9c1f7774a9893c98269e88243fc", "1710b6dc762c45698eef5c612caa94fd" ] }, "id": "DkIvEkIIkEyB", "outputId": "fbc9099f-383f-4489-e15a-776f7f2cd952" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "Loading model with compilation disabled...\n" ] }, { "output_type": "stream", "name": "stderr", "text": [ "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n" ] }, { "output_type": "stream", "name": "stdout", "text": [ "==((====))== Unsloth 2025.10.7: Fast Cohere2 patching. Transformers: 4.55.4.\n", " \\\\ /| NVIDIA L4. Num GPUs = 1. Max memory: 22.161 GB. Platform: Linux.\n", "O^O/ \\_/ \\ Torch: 2.8.0+cu126. CUDA: 8.9. CUDA Toolkit: 12.6. Triton: 3.4.0\n", "\\ / Bfloat16 = TRUE. FA [Xformers = 0.0.32.post2. FA2 = False]\n", " \"-____-\" Free license: http://github.com/unslothai/unsloth\n", "Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "model.safetensors.index.json: 0%| | 0.00/58.3k [00:00 and .\\nThen, provide your solution between '" ], "application/vnd.google.colaboratory.intrinsic+json": { "type": "string" } }, "metadata": {}, "execution_count": 15 } ], "source": [ "reasoning_start = \"\" # Acts as \n", "reasoning_end = \"\" # Acts as \n", "solution_start = \"\"\n", "solution_end = \"\"\n", "\n", "system_prompt = \\\n", "f\"\"\"You are given a problem.\n", "Think about the problem and provide your working out.\n", "Place it between {reasoning_start} and {reasoning_end}.\n", "Then, provide your solution between {solution_start}{solution_end}\"\"\"\n", "system_prompt" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 35 }, "id": "8wkaD6-gZP5K", "outputId": "870eb299-cf4b-4ae0-9a9a-578520a42c02" }, "outputs": [ { "data": { "application/vnd.google.colaboratory.intrinsic+json": { "type": "string" }, "text/plain": [ "'You are given a information and a problem.\\nThink about the problem and provide code to visualize the data .\\nPlace your solution between '" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "solution_start = \"\"\n", "solution_end = \"\"\n", "\n", "system_prompt = \\\n", "f\"\"\"You are given a information and a problem.\n", "Think about the problem and provide code to visualize the data .\n", "Place your solution between {solution_start}{solution_end}\"\"\"\n", "system_prompt" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 35 }, "id": "WyHQFueXaARE", "outputId": "fcdc7593-9363-483a-c4fe-705262dce7a9" }, "outputs": [ { "data": { "application/vnd.google.colaboratory.intrinsic+json": { "type": "string" }, "text/plain": [ "'You are given an article.\\nSummarize it.\\nPlace your solution between '" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "solution_start = \"\"\n", "solution_end = \"\"\n", "\n", "system_prompt = \\\n", "f\"\"\"You are given an article.\n", "Summarize it.\n", "Place your solution between {solution_start}{solution_end}\"\"\"\n", "system_prompt" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 35 }, "id": "Hp50RZ7saZtB", "outputId": "ccb6e8db-5067-48ff-a978-45fc760e0f00" }, "outputs": [ { "data": { "application/vnd.google.colaboratory.intrinsic+json": { "type": "string" }, "text/plain": [ "'Sort this item into one of the listed categories.\\nPlace your solution between '" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "solution_start = \"\"\n", "solution_end = \"\"\n", "\n", "system_prompt = \\\n", "f\"\"\"Sort this item into one of the listed categories.\n", "Place your solution between {solution_start}{solution_end}\"\"\"\n", "system_prompt" ] }, { "cell_type": "markdown", "metadata": { "id": "zGgs0MJkDkYL" }, "source": [ "We create a simple chat template below. Notice `add_generation_prompt` includes prepending `` to guide the model to start its reasoning process." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 618, "referenced_widgets": [ "9e569beef90a4298b2108e8e53cf472c", "fd4d9887c5fa42399f2827ffde895d91", "1f7dd514ef12489bb5f65ade53dfee03", "0c9fb3f828b74a14b662b645fc0d7ea6", "5030863392aa4d2fb7093b8cdd022d2b", "345ec5703e8441bebcc9677fd13ceddd", "67dd49c5407344db94af34221fc17e3e", "c8b9eef9d28d489ca39e3708425eb5e8", "a2c059e71dc64a2c89a6c064f9f1b728", "4ecd1e548ca24a229425068fee64616e", "d17d5633d26842dd8312a3a9d452205f", "d82e142c54e745f995d45721e0e8a1a1", "357d36426a1f406ca52f107f965e10f9", "43e374663fac4ebab35ca28c741bd5d2", "2218ff86bf1341ca85f047828cd1c506", "1494da7a0063486aa65d62f72e6c87d9", "3bb5b09407d24c7b873f2a2e19d9e0ca", "6bc7bb6e65d64c298db28b8384832ad0", "7913c54c35ed4b1bbf739529918b668d", "4bc054d517f14fe1b188009d594a2ed6", "3524774fe0ca47ff87b66e8cb7355a02", "ff571a65f5e141c6a6deb77c8995568e", "ba3fdeb2e19c415b9fe50bcd466b8405", "e6d4b4e3b11d45e6bdb9aaefd1774966", "59d8edcb469f42feb04358ceab22c605", "be2964124ea5408b95683ac4858fd07f", "fb3a3907209d4b90bf3623a609c5408c", "9077bb85471842778694440efe0cb0c9", "efd002288b4340a7ae73504a0df2e366", "af278e789e294164918c92d41221b110", "eab17266cbb74865aff538d099847926", "1dc3a5f54ee94b91a5813a339c2bbbe1", "10ab127a7fee44ddb970eabf8264ce7b", "96dd3f73751e476a95cda8e0d69af91d", "085a755dc1f24019b6ee99008981269f", "90cd232f14a64bccb0170122c208eefd", "52c5254f37e64d53ba4d0c919b773f31", "0c160ce8be974b15be462f5e9093a86b", "cead13807e844f7d87f6c0ac25f5b082", "0ac6b38773194754b81f64b83d4d1c1b", "7e4d58b88a74480e971998557f34a5b1", "8197c560780f4dc0a8210c2184f43d86", "e41c7149a8b145548b202321bdc53d84", "f939a61147164a8a8674e0d30ca8bfee", "e4879191b081487884e92e1ce86bf5e9", "2f2b60d3139f46c1b8bfd037ac414240", "e6207d500e6e431fa8b01ce8c6c5de90", "5f067338524a41cb8d25ebdca4ed29c6", "75272c680aed4d0895b7104d965da302", "5ebe5a338b484ababfb518be46ee2125", "f686d4acd4834ca1a3b35487a60984fe", "ae9a37d6ca464b4c92b75d73734f09fd", "1dbcf04db5ef464c9688d30f37308ba3", "f02e04a7233840c88f802bbfd0f5ac00", "57fc2fd1191e489c97b001d8dee8b801", "fb8679c34be940f89a19f29e6c88455a", "9e7e22118d5a49159bd13f9aba05e165", "d990794f3cb4405aafa110f858d9a477", "9c0ba9a31649487da8f8b3339ee92593", "099d3583d154467699e0d799b3b38715", "4feb457dd71943d5b3481b7f56741f90", "b79b6e03feb5435cbfb3e62028b60e82", "ee5a559d4a3a49fab15b0eaea0ea110b", "16c1b46faaaf4b41b9fc5add1892109d", "3a4f41b16d26449e986b6a9d6016627e", "e907a2b04fbe4270afd01684a80e51f7" ] }, "id": "AXxM2lStVIkd", "outputId": "37764f0e-1e69-4612-fbdd-7961176717ab" }, "outputs": [ { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "9e569beef90a4298b2108e8e53cf472c", "version_major": 2, "version_minor": 0 }, "text/plain": [ "README.md: 0.00B [00:00, ?B/s]" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "d82e142c54e745f995d45721e0e8a1a1", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Resolving data files: 0%| | 0/20 [00:00 629\u001b[0;31m download_files(\n\u001b[0m\u001b[1;32m 630\u001b[0m \u001b[0mxet_download_info\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;31mKeyboardInterrupt\u001b[0m: ", "\nDuring handling of the above exception, another exception occurred:\n", "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)", "\u001b[0;32m/tmp/ipython-input-3259331980.py\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mnumpy\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m \u001b[0mdataset\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mload_dataset\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"open-r1/OpenR1-Math-220k\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msplit\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m\"train\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 6\u001b[0m \u001b[0;31m#dataset = dataset.to_pandas()[\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 7\u001b[0m \u001b[0;31m# [\"expected_answer\", \"problem\", \"generated_solution\"]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/load.py\u001b[0m in \u001b[0;36mload_dataset\u001b[0;34m(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, keep_in_memory, save_infos, revision, token, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)\u001b[0m\n\u001b[1;32m 2082\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2083\u001b[0m \u001b[0;31m# Download and prepare data\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2084\u001b[0;31m builder_instance.download_and_prepare(\n\u001b[0m\u001b[1;32m 2085\u001b[0m \u001b[0mdownload_config\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdownload_config\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2086\u001b[0m \u001b[0mdownload_mode\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdownload_mode\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/builder.py\u001b[0m in \u001b[0;36mdownload_and_prepare\u001b[0;34m(self, output_dir, download_config, download_mode, verification_mode, dl_manager, base_path, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)\u001b[0m\n\u001b[1;32m 923\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mnum_proc\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 924\u001b[0m \u001b[0mprepare_split_kwargs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"num_proc\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnum_proc\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 925\u001b[0;31m self._download_and_prepare(\n\u001b[0m\u001b[1;32m 926\u001b[0m \u001b[0mdl_manager\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdl_manager\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 927\u001b[0m \u001b[0mverification_mode\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mverification_mode\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/builder.py\u001b[0m in \u001b[0;36m_download_and_prepare\u001b[0;34m(self, dl_manager, verification_mode, **prepare_split_kwargs)\u001b[0m\n\u001b[1;32m 977\u001b[0m \u001b[0msplit_dict\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mSplitDict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataset_name\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdataset_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 978\u001b[0m \u001b[0msplit_generators_kwargs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_make_split_generators_kwargs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mprepare_split_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 979\u001b[0;31m \u001b[0msplit_generators\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_split_generators\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdl_manager\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0msplit_generators_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 980\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 981\u001b[0m \u001b[0;31m# Checksums verification\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/packaged_modules/parquet/parquet.py\u001b[0m in \u001b[0;36m_split_generators\u001b[0;34m(self, dl_manager)\u001b[0m\n\u001b[1;32m 47\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"At least one data file must be specified, but got data_files={self.config.data_files}\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 48\u001b[0m \u001b[0mdl_manager\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdownload_config\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mextract_on_the_fly\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 49\u001b[0;31m \u001b[0mdata_files\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdl_manager\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdownload_and_extract\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata_files\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 50\u001b[0m \u001b[0msplits\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 51\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0msplit_name\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfiles\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mdata_files\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mitems\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/download/download_manager.py\u001b[0m in \u001b[0;36mdownload_and_extract\u001b[0;34m(self, url_or_urls)\u001b[0m\n\u001b[1;32m 324\u001b[0m \u001b[0mextracted_path\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;31m \u001b[0m\u001b[0;31m`\u001b[0m\u001b[0mstr\u001b[0m\u001b[0;31m`\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mextracted\u001b[0m \u001b[0mpaths\u001b[0m \u001b[0mof\u001b[0m \u001b[0mgiven\u001b[0m \u001b[0mURL\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 325\u001b[0m \"\"\"\n\u001b[0;32m--> 326\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mextract\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdownload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0murl_or_urls\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 327\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 328\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mget_recorded_sizes_checksums\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/download/download_manager.py\u001b[0m in \u001b[0;36mdownload\u001b[0;34m(self, url_or_urls)\u001b[0m\n\u001b[1;32m 157\u001b[0m \u001b[0mstart_time\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdatetime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 158\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mstack_multiprocessing_download_progress_bars\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 159\u001b[0;31m downloaded_path_or_paths = map_nested(\n\u001b[0m\u001b[1;32m 160\u001b[0m \u001b[0mdownload_func\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 161\u001b[0m \u001b[0murl_or_urls\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/utils/py_utils.py\u001b[0m in \u001b[0;36mmap_nested\u001b[0;34m(function, data_struct, dict_only, map_list, map_tuple, map_numpy, num_proc, parallel_min_length, batched, batch_size, types, disable_tqdm, desc)\u001b[0m\n\u001b[1;32m 502\u001b[0m \u001b[0mnum_proc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 503\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0many\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtypes\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterable\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mv\u001b[0m \u001b[0;32min\u001b[0m \u001b[0miterable\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 504\u001b[0;31m mapped = [\n\u001b[0m\u001b[1;32m 505\u001b[0m map_nested(\n\u001b[1;32m 506\u001b[0m \u001b[0mfunction\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfunction\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/utils/py_utils.py\u001b[0m in \u001b[0;36m\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 503\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0many\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtypes\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterable\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mv\u001b[0m \u001b[0;32min\u001b[0m \u001b[0miterable\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 504\u001b[0m mapped = [\n\u001b[0;32m--> 505\u001b[0;31m map_nested(\n\u001b[0m\u001b[1;32m 506\u001b[0m \u001b[0mfunction\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfunction\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 507\u001b[0m \u001b[0mdata_struct\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/utils/py_utils.py\u001b[0m in \u001b[0;36mmap_nested\u001b[0;34m(function, data_struct, dict_only, map_list, map_tuple, map_numpy, num_proc, parallel_min_length, batched, batch_size, types, disable_tqdm, desc)\u001b[0m\n\u001b[1;32m 519\u001b[0m \u001b[0mbatch_size\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmax\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterable\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m//\u001b[0m \u001b[0mnum_proc\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterable\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0mnum_proc\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 520\u001b[0m \u001b[0miterable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miter_batched\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterable\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 521\u001b[0;31m mapped = [\n\u001b[0m\u001b[1;32m 522\u001b[0m \u001b[0m_single_map_nested\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunction\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatched\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtypes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 523\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mobj\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mhf_tqdm\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterable\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdisable\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdisable_tqdm\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdesc\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdesc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/utils/py_utils.py\u001b[0m in \u001b[0;36m\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 520\u001b[0m \u001b[0miterable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miter_batched\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterable\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 521\u001b[0m mapped = [\n\u001b[0;32m--> 522\u001b[0;31m \u001b[0m_single_map_nested\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunction\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatched\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtypes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 523\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mobj\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mhf_tqdm\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterable\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdisable\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdisable_tqdm\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdesc\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdesc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 524\u001b[0m ]\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/utils/py_utils.py\u001b[0m in \u001b[0;36m_single_map_nested\u001b[0;34m(args)\u001b[0m\n\u001b[1;32m 388\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mdict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtypes\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mv\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mdata_struct\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 389\u001b[0m ):\n\u001b[0;32m--> 390\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mmapped_item\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mbatch\u001b[0m \u001b[0;32min\u001b[0m \u001b[0miter_batched\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata_struct\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mmapped_item\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mfunction\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 391\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 392\u001b[0m \u001b[0;31m# Reduce logging to keep things readable in multiprocessing with tqdm\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/utils/py_utils.py\u001b[0m in \u001b[0;36m\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 388\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mdict\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtypes\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mv\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mdata_struct\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 389\u001b[0m ):\n\u001b[0;32m--> 390\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mmapped_item\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mbatch\u001b[0m \u001b[0;32min\u001b[0m \u001b[0miter_batched\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata_struct\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mmapped_item\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mfunction\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 391\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 392\u001b[0m \u001b[0;31m# Reduce logging to keep things readable in multiprocessing with tqdm\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/download/download_manager.py\u001b[0m in \u001b[0;36m_download_batched\u001b[0;34m(self, url_or_filenames, download_config)\u001b[0m\n\u001b[1;32m 217\u001b[0m )\n\u001b[1;32m 218\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 219\u001b[0;31m return [\n\u001b[0m\u001b[1;32m 220\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_download_single\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0murl_or_filename\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdownload_config\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdownload_config\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 221\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0murl_or_filename\u001b[0m \u001b[0;32min\u001b[0m \u001b[0murl_or_filenames\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/download/download_manager.py\u001b[0m in \u001b[0;36m\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 218\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 219\u001b[0m return [\n\u001b[0;32m--> 220\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_download_single\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0murl_or_filename\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdownload_config\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdownload_config\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 221\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0murl_or_filename\u001b[0m \u001b[0;32min\u001b[0m \u001b[0murl_or_filenames\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 222\u001b[0m ]\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/download/download_manager.py\u001b[0m in \u001b[0;36m_download_single\u001b[0;34m(self, url_or_filename, download_config)\u001b[0m\n\u001b[1;32m 227\u001b[0m \u001b[0;31m# append the relative path to the base_path\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 228\u001b[0m \u001b[0murl_or_filename\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0murl_or_path_join\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_base_path\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0murl_or_filename\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 229\u001b[0;31m \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcached_path\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0murl_or_filename\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdownload_config\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdownload_config\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 230\u001b[0m \u001b[0mout\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtracked_str\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 231\u001b[0m \u001b[0mout\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_origin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0murl_or_filename\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/datasets/utils/file_utils.py\u001b[0m in \u001b[0;36mcached_path\u001b[0;34m(url_or_filename, download_config, **download_kwargs)\u001b[0m\n\u001b[1;32m 195\u001b[0m \u001b[0mlibrary_version\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0m__version__\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 196\u001b[0m \u001b[0muser_agent\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mget_datasets_user_agent\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdownload_config\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0muser_agent\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 197\u001b[0;31m \u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhf_hub_download\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 198\u001b[0m \u001b[0mrepo_id\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mresolved_path\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrepo_id\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 199\u001b[0m \u001b[0mrepo_type\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mresolved_path\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrepo_type\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/huggingface_hub/utils/_validators.py\u001b[0m in \u001b[0;36m_inner_fn\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m 112\u001b[0m \u001b[0mkwargs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msmoothly_deprecate_use_auth_token\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfn_name\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__name__\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhas_token\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mhas_token\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 113\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 114\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 115\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 116\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0m_inner_fn\u001b[0m \u001b[0;31m# type: ignore\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/huggingface_hub/hf_api.py\u001b[0m in \u001b[0;36mhf_hub_download\u001b[0;34m(self, repo_id, filename, subfolder, repo_type, revision, cache_dir, local_dir, force_download, proxies, etag_timeout, token, local_files_only, resume_download, force_filename, local_dir_use_symlinks)\u001b[0m\n\u001b[1;32m 5526\u001b[0m \u001b[0mtoken\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtoken\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5527\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 5528\u001b[0;31m return hf_hub_download(\n\u001b[0m\u001b[1;32m 5529\u001b[0m \u001b[0mrepo_id\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mrepo_id\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5530\u001b[0m \u001b[0mfilename\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfilename\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/huggingface_hub/utils/_validators.py\u001b[0m in \u001b[0;36m_inner_fn\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m 112\u001b[0m \u001b[0mkwargs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msmoothly_deprecate_use_auth_token\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfn_name\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__name__\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhas_token\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mhas_token\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 113\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 114\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 115\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 116\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0m_inner_fn\u001b[0m \u001b[0;31m# type: ignore\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/huggingface_hub/file_download.py\u001b[0m in \u001b[0;36mhf_hub_download\u001b[0;34m(repo_id, filename, subfolder, repo_type, revision, library_name, library_version, cache_dir, local_dir, user_agent, force_download, proxies, etag_timeout, token, local_files_only, headers, endpoint, resume_download, force_filename, local_dir_use_symlinks)\u001b[0m\n\u001b[1;32m 1008\u001b[0m )\n\u001b[1;32m 1009\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1010\u001b[0;31m return _hf_hub_download_to_cache_dir(\n\u001b[0m\u001b[1;32m 1011\u001b[0m \u001b[0;31m# Destination\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1012\u001b[0m \u001b[0mcache_dir\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcache_dir\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/huggingface_hub/file_download.py\u001b[0m in \u001b[0;36m_hf_hub_download_to_cache_dir\u001b[0;34m(cache_dir, repo_id, filename, repo_type, revision, endpoint, etag_timeout, headers, proxies, token, local_files_only, force_download)\u001b[0m\n\u001b[1;32m 1169\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1170\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mWeakFileLock\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlock_path\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1171\u001b[0;31m _download_to_tmp_and_move(\n\u001b[0m\u001b[1;32m 1172\u001b[0m \u001b[0mincomplete_path\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mPath\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mblob_path\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m\".incomplete\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1173\u001b[0m \u001b[0mdestination_path\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mPath\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mblob_path\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/huggingface_hub/file_download.py\u001b[0m in \u001b[0;36m_download_to_tmp_and_move\u001b[0;34m(incomplete_path, destination_path, url_to_download, proxies, headers, expected_size, filename, force_download, etag, xet_file_data)\u001b[0m\n\u001b[1;32m 1721\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mxet_file_data\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mis_xet_available\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1722\u001b[0m \u001b[0mlogger\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdebug\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Xet Storage is enabled for this repo. Downloading file from Xet Storage..\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1723\u001b[0;31m xet_get(\n\u001b[0m\u001b[1;32m 1724\u001b[0m \u001b[0mincomplete_path\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mincomplete_path\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1725\u001b[0m \u001b[0mxet_file_data\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mxet_file_data\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/huggingface_hub/file_download.py\u001b[0m in \u001b[0;36mxet_get\u001b[0;34m(incomplete_path, xet_file_data, headers, expected_size, displayed_filename, _tqdm_bar)\u001b[0m\n\u001b[1;32m 622\u001b[0m )\n\u001b[1;32m 623\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 624\u001b[0;31m \u001b[0;32mwith\u001b[0m \u001b[0mprogress_cm\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mprogress\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 625\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 626\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mprogress_updater\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mprogress_bytes\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;32m/usr/local/lib/python3.11/dist-packages/tqdm/std.py\u001b[0m in \u001b[0;36m__exit__\u001b[0;34m(self, exc_type, exc_value, traceback)\u001b[0m\n\u001b[1;32m 1136\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1137\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1138\u001b[0;31m \u001b[0;32mdef\u001b[0m \u001b[0m__exit__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mexc_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mexc_value\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtraceback\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1139\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1140\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;31mKeyboardInterrupt\u001b[0m: " ] } ], "source": [ "from datasets import load_dataset\n", "import pandas as pd\n", "import numpy as np\n", "\n", "dataset = load_dataset(\"open-r1/OpenR1-Math-220k\", split = \"train\")\n", "#dataset = dataset.to_pandas()[\n", " # [\"expected_answer\", \"problem\", \"generated_solution\"]\n", "#]\n", "\n", "dataset = dataset.to_pandas()[\n", " [\"answer\", \"problem\", \"generations\"]\n", "]\n", "# Try converting to number - if not, replace with NaN\n", "#is_number = pd.to_numeric(pd.Series(dataset[\"expected_answer\"]), errors = \"coerce\").notnull()\n", "# Select only numbers\n", "#dataset = dataset.iloc[np.where(is_number)[0]]\n", "\n", "dataset" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "W6PXOpQLDdtO", "outputId": "fe2e6685-be72-4c71-a882-4e6035a06a53" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "Extracted 3000 responses\n" ] } ], "source": [ "from datasets import load_dataset\n", "\n", "dataset = load_dataset(\"microsoft/ms_marco\", \"v1.1\", split=\"train\")\n", "dataset = dataset.shuffle(seed = 60)\n", "data = []\n", "\n", "for i in dataset:\n", " a, b, c = i[\"passages\"], i[\"query\"], i[\"answers\"]\n", " if(len(a)+len(b)+len(c) < 4096):\n", " if len(data)<3000:\n", " data.append(i)\n", "\n", "dataset = data\n", "\n", "convos = []\n", "problems = []\n", "tools = []\n", "\n", "for example in dataset:\n", " tools.append(example[\"passages\"])\n", " problems.append(example[\"query\"])\n", " convos.append(example[\"answers\"])\n", "\n", "\n", "searches = []\n", "for i in tools:\n", " string = \"\"\n", " x = i.get(\"passage_text\")\n", " for y in x:\n", " string += (\"/\"+y+\"/ , \")\n", " searches.append(string)\n", "\n", "tools = searches\n", "\n", "print(f\"Extracted {len(convos)} responses\")" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "mFFFtSa5WS1B", "outputId": "f1a7caf3-02a4-47ec-d25a-c06b5da3f7f3" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "4000\n", "Extracted 4000 GPT responses\n" ] } ], "source": [ "from datasets import load_dataset\n", "import random\n", "\n", "ds = load_dataset(\"interstellarninja/hermes_reasoning_tool_use\", split=\"train\")\n", "ds = ds.shuffle(seed = 70)\n", "dataS = []\n", "num = 0\n", "\n", "for i in ds:\n", "\n", " if i[\"scenario_category\"] == \"single\":\n", " if num < 4000:\n", " num += 1\n", " dataS.append(i)\n", "\n", "ds = dataS\n", "print(len(ds))\n", "\n", "convos2 = []\n", "problems2 = []\n", "tools2 = []\n", "\n", "for example in ds:\n", " conversations = example['conversations']\n", "\n", " for turn in conversations:\n", " if turn['from'] == 'gpt':\n", " convos2.append(turn['value'])\n", " problems2.append(example['task'])\n", " tools2.append(example['tools'])\n", " break\n", "\n", "print(f\"Extracted {len(convos2)} GPT responses\")" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "v6v4WHgrASqu", "outputId": "47c6f7da-6c0a-4b0c-b716-b93fc39a19a5" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\u001b[1;30;43mStreaming output truncated to the last 5000 lines.\u001b[0m\n", "33223\n", "24002\n", "6098\n", "35600\n", "8938\n", "73604\n", "65875\n", "34877\n", "25478\n", "16984\n", "11497\n", "62263\n", "31096\n", "10037\n", "12229\n", "86914\n", "51023\n", "56554\n", "18879\n", "34018\n", "23838\n", "43753\n", "31858\n", "47089\n", "18119\n", "29201\n", "11839\n", "36689\n", "51933\n", "63554\n", "82514\n", "15293\n", "9633\n", "31357\n", "13067\n", "13701\n", "80124\n", "23854\n", "48954\n", "67685\n", "17700\n", "17570\n", "3803\n", "30945\n", "46041\n", "40780\n", "78508\n", "19361\n", "64665\n", "61115\n", "17060\n", "39824\n", "22351\n", "27797\n", "23294\n", "56580\n", "34354\n", "38265\n", "34198\n", "13765\n", "24811\n", "23027\n", "6859\n", "56544\n", "132117\n", "27949\n", "56105\n", "41205\n", "61559\n", "21775\n", "21090\n", "69684\n", "31976\n", "38118\n", "22069\n", "50655\n", "21934\n", "25372\n", "35986\n", "89758\n", "22505\n", "50832\n", "54811\n", "29458\n", "24472\n", "7504\n", "37306\n", "20588\n", "36581\n", "93273\n", "73619\n", "9252\n", "7219\n", "21981\n", "33857\n", "34579\n", "24936\n", "101430\n", "102491\n", "31462\n", "11812\n", "5505\n", "23084\n", "84700\n", "43379\n", "6038\n", "62877\n", "21442\n", "32731\n", "66787\n", "84931\n", "22988\n", "26442\n", "15238\n", "56135\n", "14777\n", "29141\n", "6897\n", "64877\n", "38659\n", "49935\n", "21258\n", "46479\n", "15758\n", "16482\n", "20433\n", "48305\n", "36635\n", "84102\n", "30348\n", "36314\n", "25745\n", "26342\n", "37236\n", "21718\n", "37066\n", "31344\n", "46135\n", "15279\n", "423787\n", "42709\n", "64721\n", "49388\n", "9596\n", "19356\n", "23397\n", "34755\n", "25316\n", "44025\n", "47321\n", "45606\n", "34683\n", "46719\n", "30307\n", "33669\n", "18327\n", "17683\n", "33272\n", "36663\n", "31928\n", "15661\n", "31821\n", "26512\n", "26344\n", "14108\n", "65044\n", "29107\n", "50899\n", "78543\n", "62178\n", "30648\n", "16175\n", "56644\n", "15685\n", "59666\n", "72583\n", "54388\n", "56900\n", "16342\n", "42241\n", "35615\n", "12404\n", "165426\n", "20536\n", "14773\n", "31686\n", "27107\n", "37911\n", "28035\n", "20021\n", "9756\n", "6950\n", "20676\n", "35707\n", "12250\n", "82311\n", "47308\n", "20643\n", "20697\n", "56034\n", "19958\n", "38113\n", "76983\n", "54717\n", "22611\n", "18297\n", "11463\n", "51868\n", "27576\n", "47096\n", "18994\n", "16759\n", "26466\n", "77610\n", "31582\n", "28485\n", "28303\n", "53966\n", "28752\n", "36479\n", "30041\n", "31029\n", "95050\n", "28316\n", "17202\n", "47634\n", "21323\n", "17528\n", "8381\n", "50124\n", "22088\n", "25059\n", "21465\n", "46384\n", "33031\n", "31392\n", "15672\n", "23790\n", "12673\n", "78385\n", "24901\n", "15110\n", "109278\n", "22143\n", "16647\n", "22605\n", "23344\n", "42346\n", "55178\n", "52447\n", "8785\n", "53731\n", "18146\n", "113190\n", "66925\n", "11503\n", "38299\n", "19564\n", "56367\n", "10311\n", "104109\n", "9220\n", "26691\n", "37863\n", "10818\n", "18821\n", "13976\n", "64009\n", "16155\n", "30427\n", "55497\n", "83160\n", "59049\n", "128255\n", "45400\n", "31570\n", "46067\n", "14789\n", "21383\n", "39223\n", "18323\n", "36350\n", "117343\n", "61111\n", "31139\n", "113939\n", "36233\n", "57908\n", "49553\n", "24277\n", "45004\n", "52746\n", "85469\n", "29354\n", "35359\n", "3380\n", "25461\n", "49315\n", "10307\n", "24867\n", "29574\n", "28601\n", "30477\n", "22760\n", "12335\n", "93807\n", "59679\n", "15076\n", "31379\n", "31771\n", "28616\n", "25506\n", "69813\n", "17817\n", "52816\n", "22484\n", "21122\n", "162673\n", "41751\n", "57151\n", "33624\n", "21958\n", "122880\n", "45374\n", "15940\n", "24139\n", "27564\n", "9295\n", "31397\n", "46980\n", "34365\n", "24433\n", "96878\n", "58682\n", "54365\n", "15350\n", "27901\n", "96536\n", "18930\n", "27678\n", "27698\n", "23351\n", "15804\n", "10117\n", "32276\n", "80518\n", "4477\n", "26320\n", "70949\n", "42065\n", "13162\n", "27365\n", "29035\n", "69710\n", "37728\n", "10590\n", "17248\n", "26985\n", "19512\n", "20284\n", "14811\n", "13782\n", "18074\n", "22161\n", "12615\n", "47819\n", "17294\n", "24422\n", "23797\n", "28721\n", "23989\n", "42120\n", "22815\n", "27166\n", "59954\n", "12780\n", "19642\n", "64833\n", "56993\n", "70863\n", "50812\n", "44196\n", "14688\n", "35829\n", "53078\n", "5489\n", "19216\n", "17499\n", "83848\n", "45169\n", "15343\n", "23261\n", "71769\n", "78322\n", "24929\n", "30624\n", "79107\n", "31824\n", "17872\n", "82353\n", "24515\n", "67320\n", "104938\n", "9011\n", "64192\n", "16695\n", "55139\n", "35207\n", "71039\n", "7217\n", "32801\n", "24594\n", "18270\n", "8374\n", "24982\n", "17124\n", "41456\n", "22290\n", "19231\n", "10927\n", "40655\n", "18769\n", "34608\n", "51722\n", "69365\n", "60978\n", "27078\n", "16818\n", "58025\n", "16003\n", "19502\n", "34497\n", "54817\n", "24655\n", "65190\n", "29132\n", "69578\n", "21521\n", "22271\n", "29637\n", "22638\n", "46383\n", "9809\n", "17962\n", "83676\n", "25270\n", "82433\n", "21198\n", "50624\n", "48694\n", "24866\n", "27260\n", "10921\n", "135075\n", "32122\n", "31586\n", "31825\n", "13281\n", "16741\n", "180920\n", "43802\n", "27262\n", "87193\n", "18445\n", "54897\n", "20768\n", "25035\n", "25044\n", "121385\n", "37162\n", "19300\n", "37240\n", "18518\n", "46249\n", "28300\n", "48817\n", "46839\n", "17382\n", "31838\n", "30472\n", "15767\n", "32586\n", "51460\n", "20892\n", "16032\n", "13604\n", "50609\n", "35263\n", "28228\n", "48448\n", "34013\n", "19539\n", "74938\n", "17473\n", "123900\n", "34219\n", "44971\n", "15178\n", "71320\n", "80340\n", "22411\n", "42140\n", "7362\n", "17416\n", "13308\n", "144893\n", "43486\n", "8550\n", "33537\n", "15815\n", "18371\n", "30842\n", "14060\n", "65965\n", "15427\n", "52793\n", "51890\n", "29983\n", "16529\n", "54071\n", "6444\n", "10643\n", "33990\n", "18074\n", "19701\n", "59472\n", "28612\n", "68255\n", "12035\n", "219425\n", "12702\n", "33635\n", "28074\n", "49610\n", "17897\n", "32007\n", "41531\n", "10041\n", "24948\n", "48975\n", "23301\n", "10058\n", "22894\n", "53890\n", "14910\n", "40664\n", "93389\n", "7230\n", "18020\n", "35725\n", "12780\n", "39978\n", "11919\n", "11787\n", "20899\n", "33484\n", "19278\n", "24534\n", "16676\n", "58532\n", "84031\n", "29464\n", "33542\n", "10635\n", "72473\n", "66623\n", "18791\n", "46469\n", "8579\n", "24457\n", "48894\n", "37947\n", "27477\n", "16666\n", "83966\n", "26278\n", "39237\n", "5538\n", "15542\n", "8277\n", "58022\n", "24997\n", "46381\n", "103522\n", "23509\n", "21928\n", "19572\n", "11269\n", "60019\n", "124343\n", "41329\n", "135828\n", "16742\n", "9464\n", "27627\n", "84283\n", "13805\n", "65274\n", "43300\n", "49722\n", "39573\n", "37577\n", "27687\n", "60445\n", "10316\n", "10972\n", "103570\n", "72052\n", "18750\n", "56231\n", "3866\n", "18242\n", "55728\n", "22183\n", "37385\n", "26186\n", "67728\n", "32996\n", "20448\n", "62228\n", "49422\n", "11030\n", "23579\n", "22332\n", "30285\n", "53143\n", "30448\n", "26836\n", "61413\n", "35274\n", "69149\n", "22374\n", "14957\n", "33254\n", "80143\n", "88779\n", "36451\n", "18372\n", "16301\n", "16437\n", "25315\n", "26816\n", "80829\n", "77444\n", "25083\n", "37414\n", "42947\n", "38782\n", "60303\n", "70477\n", "169702\n", "17347\n", "13332\n", "27739\n", "11145\n", "45740\n", "15393\n", "31687\n", "32297\n", "19942\n", "17291\n", "45207\n", "49271\n", "23935\n", "62477\n", "15407\n", "22921\n", "19452\n", "26040\n", "88915\n", "36843\n", "58095\n", "15566\n", "59527\n", "35285\n", "13568\n", "10728\n", "29988\n", "5486\n", "45196\n", "42072\n", "32139\n", "37918\n", "46400\n", "29369\n", "21415\n", "21180\n", "31109\n", "41076\n", "19850\n", "43834\n", "32226\n", "25566\n", "36240\n", "43563\n", "14408\n", "66144\n", "76284\n", "42291\n", "35281\n", "14017\n", "25945\n", "78985\n", "15164\n", "25414\n", "73700\n", "50688\n", "31690\n", "18048\n", "20992\n", "3847\n", "11512\n", "145128\n", "47713\n", "59041\n", "27454\n", "28564\n", "38936\n", "14285\n", "11138\n", "10879\n", "25156\n", "47758\n", "67465\n", "26922\n", "29347\n", "48875\n", "47292\n", "7817\n", "9096\n", "33150\n", "38753\n", "59938\n", "31620\n", "44519\n", "28789\n", "8012\n", "2914\n", "29347\n", "18644\n", "21012\n", "32729\n", "16579\n", "24475\n", "39546\n", "19109\n", "168312\n", "8355\n", "30874\n", "88901\n", "107939\n", "33192\n", "85561\n", "6868\n", "22705\n", "23612\n", "39992\n", "13899\n", "17038\n", "33675\n", "13878\n", "11927\n", "36780\n", "39061\n", "11499\n", "12522\n", "5618\n", "36599\n", "23142\n", "24790\n", "23534\n", "21358\n", "139688\n", "10713\n", "10618\n", "22900\n", "58281\n", "45551\n", "64721\n", "27257\n", "10424\n", "33931\n", "9745\n", "34459\n", "4690\n", "12381\n", "70952\n", "50971\n", "19033\n", "57026\n", "31043\n", "42565\n", "15157\n", "16238\n", "51465\n", "14173\n", "28733\n", "175762\n", "51155\n", "27519\n", "65682\n", "94442\n", "62625\n", "18541\n", "30660\n", "71468\n", "13958\n", "5107\n", "10511\n", "26385\n", "21056\n", "42365\n", "12005\n", "14890\n", "19947\n", "54613\n", "26342\n", "19648\n", "28137\n", "50298\n", "18525\n", "28752\n", "8158\n", "28095\n", "87607\n", "12411\n", "15008\n", "33077\n", "28009\n", "31694\n", "16853\n", "46606\n", "80638\n", "33840\n", "42239\n", "11204\n", "53895\n", "20702\n", "14100\n", "37034\n", "16072\n", "38262\n", "52216\n", "24283\n", "141467\n", "18077\n", "40148\n", "47534\n", "14118\n", "24163\n", "52643\n", "22527\n", "10271\n", "37539\n", "39367\n", "135594\n", "84187\n", "14480\n", "6657\n", "124963\n", "34432\n", "8615\n", "12367\n", "109524\n", "103894\n", "81126\n", "24151\n", "47848\n", "26843\n", "62939\n", "24100\n", "14663\n", "19598\n", "56039\n", "30786\n", "32239\n", "22233\n", "15330\n", "34777\n", "15899\n", "16132\n", "30147\n", "26446\n", "31985\n", "16623\n", "12484\n", "17871\n", "16707\n", "6267\n", "12817\n", "20428\n", "27247\n", "22287\n", "17331\n", "10132\n", "32055\n", "14280\n", "9091\n", "17083\n", "24760\n", "31327\n", "30428\n", "18490\n", "23616\n", "60273\n", "41968\n", "42562\n", "19414\n", "5870\n", "19569\n", "65837\n", "28689\n", "123916\n", "43441\n", "72033\n", "34584\n", "29163\n", "16410\n", "11188\n", "70825\n", "45282\n", "96041\n", "51426\n", "56799\n", "168771\n", "20059\n", "119617\n", "18898\n", "12540\n", "7879\n", "40657\n", "28158\n", "41412\n", "7257\n", "28940\n", "25517\n", "65350\n", "18547\n", "21556\n", "38745\n", "25133\n", "24208\n", "23533\n", "19334\n", "59289\n", "19753\n", "18125\n", "54252\n", "18069\n", "6922\n", "49525\n", "15518\n", "36962\n", "81187\n", "80735\n", "179332\n", "9505\n", "50162\n", "9824\n", "34301\n", "11931\n", "4161\n", "50148\n", "29686\n", "18419\n", "21114\n", "28212\n", "25900\n", "17171\n", "31847\n", "28903\n", "34223\n", "26584\n", "25300\n", "62736\n", "75734\n", "60045\n", "27321\n", "28218\n", "15021\n", "82297\n", "40325\n", "33428\n", "35387\n", "33946\n", "74648\n", "54479\n", "11787\n", "28894\n", "38238\n", "87407\n", "10549\n", "48200\n", "27708\n", "11822\n", "24537\n", "22698\n", "71363\n", "33585\n", "52325\n", "36203\n", "17585\n", "15724\n", "43966\n", "46294\n", "33180\n", "52421\n", "47361\n", "31521\n", "62884\n", "30187\n", "36342\n", "93090\n", "55068\n", "22137\n", "57538\n", "69332\n", "21619\n", "22787\n", "22860\n", "79868\n", "41884\n", "46682\n", "22891\n", "39171\n", "29466\n", "11866\n", "10374\n", "23550\n", "16212\n", "28797\n", "19516\n", "32459\n", "33732\n", "28809\n", "32353\n", "22963\n", "33563\n", "55679\n", "17485\n", "16449\n", "47655\n", "12422\n", "40657\n", "16991\n", "34529\n", "81582\n", "74008\n", "23961\n", "53238\n", "25791\n", "18575\n", "24869\n", "30412\n", "14363\n", "36548\n", "36306\n", "29052\n", "37274\n", "51727\n", "125250\n", "18992\n", "57516\n", "25855\n", "31261\n", "20624\n", "21300\n", "30949\n", "40062\n", "11626\n", "105386\n", "80277\n", "42816\n", "24056\n", "9057\n", "21162\n", "12056\n", "52343\n", "56460\n", "12674\n", "31005\n", "32084\n", "14963\n", "50260\n", "26263\n", "32277\n", "10991\n", "25743\n", "17857\n", "31087\n", "29544\n", "26581\n", "24669\n", "26721\n", "68594\n", "8896\n", "44514\n", "44253\n", "12799\n", "5136\n", "11785\n", "29510\n", "79226\n", "9768\n", "91239\n", "70062\n", "48487\n", "39383\n", "37380\n", "13451\n", "20043\n", "25124\n", "53466\n", "25834\n", "18030\n", "44133\n", "61382\n", "29400\n", "38575\n", "27861\n", "91459\n", "67971\n", "51417\n", "34615\n", "13949\n", "14729\n", "54433\n", "35667\n", "32307\n", "23132\n", "24806\n", "19200\n", "25085\n", "24583\n", "18191\n", "24762\n", "16444\n", "19888\n", "8319\n", "61228\n", "22193\n", "13611\n", "17717\n", "18253\n", "8487\n", "15681\n", "17781\n", "45004\n", "72714\n", "27487\n", "68210\n", "30798\n", "48422\n", "38680\n", "24231\n", "45368\n", "39531\n", "31450\n", "40761\n", "9822\n", "9213\n", "6211\n", "53454\n", "19207\n", "53651\n", "17260\n", "118355\n", "44180\n", "8762\n", "70906\n", "100422\n", "16504\n", "38585\n", "9858\n", "22754\n", "12239\n", "13210\n", "75492\n", "11286\n", "40761\n", "6460\n", "27727\n", "27674\n", "51395\n", "23850\n", "19660\n", "18337\n", "37542\n", "7728\n", "18080\n", "61435\n", "26512\n", "27567\n", "10908\n", "43445\n", "14567\n", "144083\n", "30860\n", "27964\n", "9109\n", "41778\n", "29940\n", "14721\n", "10281\n", "79046\n", "21035\n", "30369\n", "34459\n", "21191\n", "9850\n", "59432\n", "17491\n", "37424\n", "6131\n", "31090\n", "28268\n", "11379\n", "61157\n", "86914\n", "247507\n", "13702\n", "18328\n", "15738\n", "57099\n", "14642\n", "28605\n", "28593\n", "11371\n", "2371\n", "68639\n", "43825\n", "40739\n", "70693\n", "11645\n", "15877\n", "107084\n", "139328\n", "21867\n", "16804\n", "56978\n", "100340\n", "21337\n", "13232\n", "74223\n", "16220\n", "28861\n", "22772\n", "18632\n", "31188\n", "20542\n", "35828\n", "79213\n", "31150\n", "16163\n", "21445\n", "26822\n", "59565\n", "62972\n", "74929\n", "18800\n", "11315\n", "53592\n", "33865\n", "10625\n", "88401\n", "13102\n", "24020\n", "61866\n", "14782\n", "18329\n", "66736\n", "20510\n", "35693\n", "75350\n", "17665\n", "48961\n", "36881\n", "19237\n", "27516\n", "24575\n", "24459\n", "55571\n", "38042\n", "12664\n", "19894\n", "23775\n", "22836\n", "19945\n", "49632\n", "33857\n", "21196\n", "42949\n", "29173\n", "16386\n", "30451\n", "28238\n", "17797\n", "77573\n", "103609\n", "92994\n", "17514\n", "25143\n", "33914\n", "20584\n", "22723\n", "71710\n", "13423\n", "23471\n", "48367\n", "69082\n", "35539\n", "52184\n", "48656\n", "57309\n", "11223\n", "17122\n", "42695\n", "15681\n", "30846\n", "19685\n", "17072\n", "10557\n", "48409\n", "60825\n", "15344\n", "21714\n", "14881\n", "27897\n", "34949\n", "27555\n", "39226\n", "2935\n", "19192\n", "47647\n", "5426\n", "136359\n", "15369\n", "19783\n", "85035\n", "23803\n", "39207\n", "22520\n", "50140\n", "12395\n", "9033\n", "15781\n", "24968\n", "57157\n", "79498\n", "22004\n", "12590\n", "9532\n", "8552\n", "87292\n", "59537\n", "15504\n", "89136\n", "23703\n", "81608\n", "11805\n", "14222\n", "23023\n", "36608\n", "74433\n", "18042\n", "17885\n", "8417\n", "6398\n", "59046\n", "67933\n", "37532\n", "52304\n", "13650\n", "22581\n", "47833\n", "7310\n", "16646\n", "25820\n", "70625\n", "49006\n", "28220\n", "27073\n", "42417\n", "16241\n", "21987\n", "13482\n", "18481\n", "26498\n", "92341\n", "16514\n", "10543\n", "20529\n", "9533\n", "14518\n", "8061\n", "11614\n", "29646\n", "19789\n", "46846\n", "54768\n", "23237\n", "15760\n", "17290\n", "47132\n", "40139\n", "13825\n", "21258\n", "87047\n", "64131\n", "18348\n", "34551\n", "15995\n", "8506\n", "42492\n", "48899\n", "22571\n", "29723\n", "23071\n", "18120\n", "3305\n", "47193\n", "51458\n", "17908\n", "47448\n", "95979\n", "32188\n", "17612\n", "26782\n", "14944\n", "43135\n", "34814\n", "38463\n", "22903\n", "60636\n", "45036\n", "12154\n", "29541\n", "19465\n", "73915\n", "93158\n", "26768\n", "16635\n", "26677\n", "11372\n", "57991\n", "25259\n", "39576\n", "44291\n", "47735\n", "25016\n", "49824\n", "22920\n", "105084\n", "66183\n", "16645\n", "8676\n", "10960\n", "25974\n", "146136\n", "32946\n", "19208\n", "10495\n", "39978\n", "12220\n", "46404\n", "11203\n", "40059\n", "41790\n", "38193\n", "18414\n", "54307\n", "10898\n", "31460\n", "10580\n", "17411\n", "31139\n", "40341\n", "29915\n", "64710\n", "16502\n", "53709\n", "12498\n", "15121\n", "14636\n", "21160\n", "31792\n", "37708\n", "78651\n", "26188\n", "7805\n", "39914\n", "57082\n", "20273\n", "37875\n", "44806\n", "49945\n", "30062\n", "18640\n", "10587\n", "51940\n", "12889\n", "17729\n", "124957\n", "30296\n", "60722\n", "22411\n", "53390\n", "9184\n", "72358\n", "10403\n", "15102\n", "16855\n", "44088\n", "35863\n", "52642\n", "23467\n", "70843\n", "21005\n", "9079\n", "22520\n", "45913\n", "18806\n", "52325\n", "30489\n", "23874\n", "13330\n", "58043\n", "112668\n", "27938\n", "11576\n", "50765\n", "23190\n", "37292\n", "21531\n", "30492\n", "18468\n", "21016\n", "58699\n", "60776\n", "44362\n", "56856\n", "32375\n", "16079\n", "29871\n", "66281\n", "30816\n", "41325\n", "25249\n", "6103\n", "28695\n", "54046\n", "113108\n", "46501\n", "26163\n", "45163\n", "22737\n", "27485\n", "26429\n", "15363\n", "65388\n", "28793\n", "42878\n", "45182\n", "28507\n", "64649\n", "20646\n", "16244\n", "51431\n", "8273\n", "20203\n", "44166\n", "30804\n", "10411\n", "19978\n", "71607\n", "26042\n", "49746\n", "121608\n", "23039\n", "25245\n", "28454\n", "40455\n", "93866\n", "21695\n", "9388\n", "40512\n", "36141\n", "78754\n", "12789\n", "38890\n", "5915\n", "17949\n", "21007\n", "11456\n", "22368\n", "13435\n", "62228\n", "104452\n", "21941\n", "13755\n", "8672\n", "16954\n", "40173\n", "20258\n", "24511\n", "20913\n", "61219\n", "17467\n", "7529\n", "7934\n", "5265\n", "53509\n", "42295\n", "16763\n", "15901\n", "26548\n", "14040\n", "19448\n", "26702\n", "59256\n", "66164\n", "30619\n", "37931\n", "38918\n", "27472\n", "86269\n", "58272\n", "7453\n", "23608\n", "33243\n", "4133\n", "18848\n", "7723\n", "44129\n", "121434\n", "26932\n", "24039\n", "35271\n", "64717\n", "32498\n", "20968\n", "62952\n", "63880\n", "109194\n", "33733\n", "138007\n", "40264\n", "85724\n", "32164\n", "47534\n", "3314\n", "29000\n", "15884\n", "18270\n", "19429\n", "96210\n", "9469\n", "25902\n", "35395\n", "18038\n", "27643\n", "11930\n", "39070\n", "15546\n", "152861\n", "73800\n", "46931\n", "27024\n", "23816\n", "6982\n", "70508\n", "19530\n", "18415\n", "26252\n", "39005\n", "42754\n", "35517\n", "109275\n", "30147\n", "39237\n", "51848\n", "19598\n", "138496\n", "50044\n", "18145\n", "21709\n", "51271\n", "83828\n", "41735\n", "38017\n", "29603\n", "18336\n", "16898\n", "12486\n", "35256\n", "27390\n", "23120\n", "33721\n", "18968\n", "42198\n", "18409\n", "38341\n", "11185\n", "102839\n", "20197\n", "40540\n", "60082\n", "72891\n", "16164\n", "34240\n", "104188\n", "59102\n", "103406\n", "24793\n", "7979\n", "30991\n", "45986\n", "35965\n", "24080\n", "38872\n", "13013\n", "17280\n", "19792\n", "8439\n", "9305\n", "17976\n", "37072\n", "35363\n", "30929\n", "14079\n", "67767\n", "22127\n", "37005\n", "10609\n", "23688\n", "49289\n", "28179\n", "16890\n", "94673\n", "79226\n", "23717\n", "29211\n", "21916\n", "21299\n", "26530\n", "40492\n", "16760\n", "46777\n", "45595\n", "56858\n", "17527\n", "55574\n", "257428\n", "11770\n", "11190\n", "20935\n", "23332\n", "6431\n", "72355\n", "53272\n", "31360\n", "43837\n", "94543\n", "24975\n", "18516\n", "82126\n", "33399\n", "20697\n", "17078\n", "22549\n", "35194\n", "12550\n", "37096\n", "31221\n", "47152\n", "53169\n", "35134\n", "13195\n", "21686\n", "21142\n", "82039\n", "48610\n", "39425\n", "22328\n", "38224\n", "15587\n", "74649\n", "21968\n", "21762\n", "31819\n", "13538\n", "36985\n", "14960\n", "17421\n", "53175\n", "16041\n", "41564\n", "12561\n", "221017\n", "24656\n", "57912\n", "32679\n", "67852\n", "17608\n", "55974\n", "15757\n", "64323\n", "20992\n", "59606\n", "73243\n", "78525\n", "29335\n", "84067\n", "55419\n", "6463\n", "29146\n", "26385\n", "45698\n", "24068\n", "57309\n", "41886\n", "26516\n", "58801\n", "34005\n", "77269\n", "48875\n", "28247\n", "69616\n", "25888\n", "15482\n", "36943\n", "56573\n", "22563\n", "29008\n", "22515\n", "11106\n", "44693\n", "56525\n", "24762\n", "24810\n", "22217\n", "21907\n", "18414\n", "60857\n", "40953\n", "76020\n", "26480\n", "13264\n", "26580\n", "17318\n", "26159\n", "18719\n", "18537\n", "52225\n", "11422\n", "21096\n", "29897\n", "19103\n", "20826\n", "26346\n", "37726\n", "48277\n", "20287\n", "31028\n", "50017\n", "40188\n", "45120\n", "30621\n", "50429\n", "23205\n", "20102\n", "84251\n", "7237\n", "21329\n", "11431\n", "6724\n", "31429\n", "76766\n", "50248\n", "31758\n", "27584\n", "28784\n", "68873\n", "24594\n", "32516\n", "50711\n", "45443\n", "50590\n", "20761\n", "20577\n", "10425\n", "26301\n", "33571\n", "33282\n", "9401\n", "26029\n", "41601\n", "39132\n", "25762\n", "43671\n", "12689\n", "30756\n", "24228\n", "24660\n", "20558\n", "46094\n", "16513\n", "55775\n", "18183\n", "27347\n", "17998\n", "29844\n", "11651\n", "27045\n", "10923\n", "102395\n", "9909\n", "89167\n", "14696\n", "36804\n", "38639\n", "22685\n", "26189\n", "26338\n", "11822\n", "45653\n", "106506\n", "9332\n", "25156\n", "18901\n", "15528\n", "19502\n", "35345\n", "34649\n", "61736\n", "50277\n", "76630\n", "8533\n", "22043\n", "14619\n", "34939\n", "46961\n", "58053\n", "39478\n", "16328\n", "78616\n", "25193\n", "36834\n", "34881\n", "28181\n", "22244\n", "19831\n", "52175\n", "30835\n", "15954\n", "19978\n", "66952\n", "26860\n", "38887\n", "15877\n", "36475\n", "26821\n", "37168\n", "13807\n", "64917\n", "52823\n", "47871\n", "49630\n", "19670\n", "21829\n", "39974\n", "13255\n", "8146\n", "21576\n", "36989\n", "14708\n", "70396\n", "36344\n", "78833\n", "6658\n", "21877\n", "20458\n", "63296\n", "19877\n", "16588\n", "24842\n", "16286\n", "68606\n", "14369\n", "20823\n", "10192\n", "86700\n", "33465\n", "56546\n", "57543\n", "11796\n", "6979\n", "15758\n", "20521\n", "8837\n", "17760\n", "44353\n", "14648\n", "118605\n", "88740\n", "33630\n", "13451\n", "32768\n", "26796\n", "47268\n", "60099\n", "40002\n", "17929\n", "56186\n", "33050\n", "70036\n", "9437\n", "17400\n", "18011\n", "3526\n", "27385\n", "42549\n", "27195\n", "34473\n", "55815\n", "39191\n", "50191\n", "21752\n", "15764\n", "22643\n", "18411\n", "26891\n", "29456\n", "86774\n", "14202\n", "32500\n", "10094\n", "59543\n", "9037\n", "29126\n", "20504\n", "54272\n", "49300\n", "32115\n", "34749\n", "69756\n", "23233\n", "66737\n", "32151\n", "19563\n", "75565\n", "6958\n", "36138\n", "57294\n", "26347\n", "36736\n", "35898\n", "42874\n", "37862\n", "12483\n", "47812\n", "38063\n", "28307\n", "39993\n", "48049\n", "23191\n", "32158\n", "20328\n", "13164\n", "34671\n", "39433\n", "45559\n", "33112\n", "67725\n", "35305\n", "20668\n", "70433\n", "36814\n", "14285\n", "40140\n", "23564\n", "26464\n", "12857\n", "70741\n", "56395\n", "19631\n", "9788\n", "25641\n", "21961\n", "38303\n", "15747\n", "6138\n", "21886\n", "16924\n", "23119\n", "40601\n", "21323\n", "29742\n", "8272\n", "17542\n", "27850\n", "20293\n", "46082\n", "6928\n", "14755\n", "11067\n", "20985\n", "24555\n", "10316\n", "74312\n", "22656\n", "19662\n", "67307\n", "18328\n", "17710\n", "27596\n", "39733\n", "16107\n", "54281\n", "19304\n", "23761\n", "36865\n", "50214\n", "25212\n", "47208\n", "53022\n", "20334\n", "16585\n", "33375\n", "31616\n", "13943\n", "40860\n", "26355\n", "16357\n", "13173\n", "27535\n", "20673\n", "51401\n", "33293\n", "24352\n", "23891\n", "16958\n", "42291\n", "6006\n", "12724\n", "26715\n", "18022\n", "17557\n", "49694\n", "5754\n", "105823\n", "57038\n", "18870\n", "50706\n", "20689\n", "43017\n", "32498\n", "85190\n", "27592\n", "13301\n", "62025\n", "75382\n", "16011\n", "27972\n", "68107\n", "29487\n", "37064\n", "78518\n", "24400\n", "14021\n", "23937\n", "10823\n", "22964\n", "17938\n", "54791\n", "27532\n", "11953\n", "83624\n", "70761\n", "29303\n", "18726\n", "18647\n", "13465\n", "26524\n", "33298\n", "17972\n", "63813\n", "23788\n", "43343\n", "54274\n", "45958\n", "44821\n", "36118\n", "41362\n", "5081\n", "31437\n", "22306\n", "133818\n", "24079\n", "39124\n", "33200\n", "12527\n", "14382\n", "19062\n", "19825\n", "47744\n", "40207\n", "79070\n", "15881\n", "27963\n", "18908\n", "25233\n", "54548\n", "9742\n", "32223\n", "24410\n", "17575\n", "29225\n", "17569\n", "50377\n", "15244\n", "7512\n", "32791\n", "17494\n", "39874\n", "24356\n", "24351\n", "27247\n", "24345\n", "38863\n", "29557\n", "60108\n", "20269\n", "6529\n", "28921\n", "17712\n", "40544\n", "31812\n", "19292\n", "19397\n", "27406\n", "31036\n", "82056\n", "71901\n", "13847\n", "14759\n", "8492\n", "23811\n", "23880\n", "41632\n", "28948\n", "22402\n", "13246\n", "19350\n", "15969\n", "8496\n", "18340\n", "28454\n", "13189\n", "23521\n", "20232\n", "67669\n", "54537\n", "35130\n", "8148\n", "50817\n", "42926\n", "63741\n", "76827\n", "169675\n", "23544\n", "32368\n", "38582\n", "25005\n", "18074\n", "52408\n", "53557\n", "51928\n", "16215\n", "57867\n", "20392\n", "41979\n", "19026\n", "19247\n", "7641\n", "42735\n", "119275\n", "62120\n", "21164\n", "8520\n", "48804\n", "22576\n", "28293\n", "29318\n", "6546\n", "31935\n", "28932\n", "51722\n", "40949\n", "13703\n", "38256\n", "15595\n", "14127\n", "25538\n", "25578\n", "10193\n", "79149\n", "34584\n", "47198\n", "22481\n", "46738\n", "37387\n", "34481\n", "25999\n", "12673\n", "81057\n", "13771\n", "13545\n", "53866\n", "11821\n", "56432\n", "42826\n", "35326\n", "25264\n", "45458\n", "10065\n", "28136\n", "16152\n", "6110\n", "35855\n", "16406\n", "85971\n", "35927\n", "14597\n", "29171\n", "22183\n", "91080\n", "29289\n", "81013\n", "104917\n", "49703\n", "15485\n", "8167\n", "34266\n", "90324\n", "20804\n", "18653\n", "13990\n", "28537\n", "18741\n", "45078\n", "28823\n", "61436\n", "22239\n", "81981\n", "25804\n", "19933\n", "64973\n", "47092\n", "18036\n", "57256\n", "32449\n", "31172\n", "9889\n", "53116\n", "20045\n", "29997\n", "12331\n", "59322\n", "54571\n", "14038\n", "20402\n", "31581\n", "70435\n", "13961\n", "20060\n", "63571\n", "19820\n", "17269\n", "9007\n", "16495\n", "110855\n", "16649\n", "18848\n", "17236\n", "37066\n", "57244\n", "85400\n", "64853\n", "12143\n", "20811\n", "14652\n", "69640\n", "21907\n", "115920\n", "24822\n", "15921\n", "31733\n", "26230\n", "26539\n", "22283\n", "119133\n", "26254\n", "58029\n", "129703\n", "38846\n", "41871\n", "20996\n", "25536\n", "30751\n", "27313\n", "153758\n", "38743\n", "10289\n", "51326\n", "20581\n", "59157\n", "56283\n", "21476\n", "16748\n", "107590\n", "51854\n", "17271\n", "12758\n", "12713\n", "12753\n", "18768\n", "3344\n", "34435\n", "21937\n", "31667\n", "21217\n", "69187\n", "17476\n", "37007\n", "9924\n", "16912\n", "18017\n", "62277\n", "34188\n", "27763\n", "37733\n", "78021\n", "46957\n", "34102\n", "124879\n", "8789\n", "88729\n", "32019\n", "19176\n", "22674\n", "61464\n", "53705\n", "31149\n", "25371\n", "21611\n", "63460\n", "39373\n", "36354\n", "21454\n", "46377\n", "83546\n", "26122\n", "46045\n", "19579\n", "18386\n", "69592\n", "19755\n", "45696\n", "13851\n", "59599\n", "15101\n", "54346\n", "11486\n", "20393\n", "61260\n", "20145\n", "76466\n", "57787\n", "80805\n", "8033\n", "18339\n", "27561\n", "11829\n", "71083\n", "10999\n", "28675\n", "8355\n", "20658\n", "39871\n", "16788\n", "18783\n", "27899\n", "19928\n", "6889\n", "27856\n", "11566\n", "58482\n", "67777\n", "44829\n", "28755\n", "41452\n", "20049\n", "24175\n", "13479\n", "33020\n", "12509\n", "67876\n", "23268\n", "56573\n", "141662\n", "29851\n", "25513\n", "53586\n", "14514\n", "28586\n", "10426\n", "42206\n", "39927\n", "14292\n", "141617\n", "31177\n", "50557\n", "31746\n", "55809\n", "22163\n", "37743\n", "17829\n", "97910\n", "42365\n", "37565\n", "23229\n", "9612\n", "53945\n", "82067\n", "61094\n", "24020\n", "22161\n", "24237\n", "61768\n", "78321\n", "26719\n", "22607\n", "35977\n", "16651\n", "17502\n", "39457\n", "62444\n", "62098\n", "29790\n", "57906\n", "32525\n", "1343\n", "30478\n", "20692\n", "23216\n", "40323\n", "11045\n", "34061\n", "22833\n", "11092\n", "47369\n", "43546\n", "15299\n", "16267\n", "28819\n", "23696\n", "12245\n", "37705\n", "70816\n", "17212\n", "18918\n", "20461\n", "73914\n", "62349\n", "64970\n", "9927\n", "82888\n", "55648\n", "14654\n", "106778\n", "18525\n", "20565\n", "19039\n", "60445\n", "8321\n", "55167\n", "48526\n", "43965\n", "44904\n", "9088\n", "38985\n", "37180\n", "60948\n", "23745\n", "51681\n", "31747\n", "33715\n", "18181\n", "20925\n", "29085\n", "142926\n", "15060\n", "47898\n", "26428\n", "18673\n", "33734\n", "36256\n", "89979\n", "54761\n", "100625\n", "50615\n", "39911\n", "41526\n", "85809\n", "4405\n", "9282\n", "54006\n", "22513\n", "19640\n", "16561\n", "81245\n", "19595\n", "31425\n", "11093\n", "14005\n", "49614\n", "266048\n", "43403\n", "24761\n", "43939\n", "22588\n", "31355\n", "22824\n", "28045\n", "41501\n", "28705\n", "46038\n", "8976\n", "32906\n", "43996\n", "21981\n", "42376\n", "13464\n", "16219\n", "31506\n", "28079\n", "49643\n", "33535\n", "21032\n", "11893\n", "43461\n", "17430\n", "23104\n", "23461\n", "123760\n", "21034\n", "23193\n", "32213\n", "26300\n", "40270\n", "23650\n", "18321\n", "14534\n", "72190\n", "59400\n", "39947\n", "91886\n", "32886\n", "55607\n", "39933\n", "24580\n", "52245\n", "36506\n", "109762\n", "15193\n", "17326\n", "8772\n", "29824\n", "82685\n", "13420\n", "31830\n", "22152\n", "7040\n", "7515\n", "40003\n", "77704\n", "40558\n", "26582\n", "36590\n", "189254\n", "25821\n", "35628\n", "27942\n", "23258\n", "17497\n", "14583\n", "17672\n", "38017\n", "46515\n", "15341\n", "24912\n", "46900\n", "11404\n", "12619\n", "22311\n", "44446\n", "26763\n", "25768\n", "37208\n", "27728\n", "137001\n", "14768\n", "10993\n", "28354\n", "39662\n", "9663\n", "42627\n", "20742\n", "12354\n", "23118\n", "45127\n", "10130\n", "33727\n", "25076\n", "19388\n", "68775\n", "38356\n", "35657\n", "53455\n", "58939\n", "19239\n", "15803\n", "9437\n", "50348\n", "35120\n", "20884\n", "46754\n", "46501\n", "40765\n", "60848\n", "18473\n", "22184\n", "48003\n", "34369\n", "41862\n", "8646\n", "70758\n", "34157\n", "58069\n", "23344\n", "14033\n", "100271\n", "35689\n", "14895\n", "19520\n", "14786\n", "16240\n", "56188\n", "30426\n", "22037\n", "23290\n", "22325\n", "16762\n", "30092\n", "78705\n", "21614\n", "49112\n", "27766\n", "87200\n", "55044\n", "21550\n", "49252\n", "35131\n", "51372\n", "16587\n", "36806\n", "36508\n", "41002\n", "22694\n", "38486\n", "29250\n", "22649\n", "42418\n", "17295\n", "55470\n", "6577\n", "30443\n", "30030\n", "19136\n", "11823\n", "34473\n", "18619\n", "4924\n", "46030\n", "34955\n", "19069\n", "10915\n", "20145\n", "40361\n", "41782\n", "42013\n", "14591\n", "18894\n", "65811\n", "17019\n", "9481\n", "24726\n", "15859\n", "6181\n", "65668\n", "13378\n", "15818\n", "48024\n", "22478\n", "103837\n", "34054\n", "46497\n", "89282\n", "13294\n", "67869\n", "28422\n", "37628\n", "10040\n", "44734\n", "46263\n", "30600\n", "57606\n", "17689\n", "37808\n", "30353\n", "11092\n", "26717\n", "31735\n", "42263\n", "6331\n", "58925\n", "31722\n", "39361\n", "40509\n", "100401\n", "35475\n", "35585\n", "17059\n", "30733\n", "113100\n", "15319\n", "29423\n", "32281\n", "13389\n", "27121\n", "43271\n", "86411\n", "40256\n", "23145\n", "51220\n", "101006\n", "20251\n", "37350\n", "18855\n", "93945\n", "24210\n", "16523\n", "17486\n", "21313\n", "49974\n", "7218\n", "23756\n", "13682\n", "32050\n", "22775\n", "13637\n", "9392\n", "60389\n", "14129\n", "100794\n", "74062\n", "27867\n", "6207\n", "6994\n", "31197\n", "245555\n", "35501\n", "19836\n", "61325\n", "25418\n", "31225\n", "59042\n", "37104\n", "13719\n", "10928\n", "23107\n", "101637\n", "26357\n", "19790\n", "42064\n", "29209\n", "9599\n", "106264\n", "29256\n", "20037\n", "41135\n", "79947\n", "34595\n", "27022\n", "105982\n", "46143\n", "12143\n", "11598\n", "7427\n", "95816\n", "24259\n", "22344\n", "12744\n", "46980\n", "33318\n", "27282\n", "17973\n", "45584\n", "17890\n", "13314\n", "6303\n", "32760\n", "11429\n", "11511\n", "37443\n", "23211\n", "25282\n", "5036\n", "93560\n", "60302\n", "32748\n", "11185\n", "39101\n", "10569\n", "28331\n", "7754\n", "37316\n", "160991\n", "31667\n", "10433\n", "24837\n", "10443\n", "37592\n", "44977\n", "6595\n", "79060\n", "28736\n", "16478\n", "34018\n", "21802\n", "35803\n", "8279\n", "15961\n", "29511\n", "9480\n", "62582\n", "25042\n", "75685\n", "202907\n", "39173\n", "11490\n", "30111\n", "22778\n", "14990\n", "31490\n", "27631\n", "50002\n", "43700\n", "17446\n", "29373\n", "39869\n", "57292\n", "34754\n", "41014\n", "25564\n", "12824\n", "61235\n", "47224\n", "19276\n", "10919\n", "110212\n", "27979\n", "40086\n", "25688\n", "16789\n", "40063\n", "37605\n", "15559\n", "46042\n", "19871\n", "55948\n", "48997\n", "13303\n", "23676\n", "63616\n", "43213\n", "39769\n", "23990\n", "103087\n", "32890\n", "24194\n", "9161\n", "30780\n", "12999\n", "17697\n", "28827\n", "42688\n", "18421\n", "68139\n", "74010\n", "25375\n", "26539\n", "38315\n", "28212\n", "15013\n", "44848\n", "15715\n", "99609\n", "57253\n", "22743\n", "23998\n", "5418\n", "27768\n", "34001\n", "25065\n", "67413\n", "29515\n", "9430\n", "37514\n", "18287\n", "21745\n", "67261\n", "69610\n", "36314\n", "55422\n", "25970\n", "82256\n", "35818\n", "16054\n", "25869\n", "46468\n", "37243\n", "42566\n", "34096\n", "25766\n", "58938\n", "32000\n", "39709\n", "28703\n", "23235\n", "19390\n", "32867\n", "33312\n", "7890\n", "31805\n", "24927\n", "17554\n", "12632\n", "38335\n", "23949\n", "21201\n", "8662\n", "20974\n", "43347\n", "20272\n", "32598\n", "65845\n", "46119\n", "17476\n", "15926\n", "49245\n", "9768\n", "21132\n", "31368\n", "149080\n", "30531\n", "48430\n", "40138\n", "45125\n", "13270\n", "24679\n", "91601\n", "32668\n", "8410\n", "28165\n", "23431\n", "10783\n", "17478\n", "26248\n", "86479\n", "26312\n", "108876\n", "5555\n", "56432\n", "131762\n", "19492\n", "50475\n", "23722\n", "35710\n", "72460\n", "19443\n", "8737\n", "41182\n", "28017\n", "60750\n", "20779\n", "30300\n", "21515\n", "10925\n", "70717\n", "94241\n", "25949\n", "47503\n", "3186\n", "16186\n", "26752\n", "55928\n", "39084\n", "38630\n", "24291\n", "39315\n", "51661\n", "35063\n", "11150\n", "29095\n", "14718\n", "10141\n", "42712\n", "34705\n", "36054\n", "44506\n", "23799\n", "21441\n", "20182\n", "41748\n", "46759\n", "14551\n", "87813\n", "31189\n", "46057\n", "45346\n", "29144\n", "18012\n", "22532\n", "9776\n", "32034\n", "7169\n", "58960\n", "59352\n", "50767\n", "8692\n", "44304\n", "28506\n", "17976\n", "70535\n", "19205\n", "22110\n", "23723\n", "16287\n", "145920\n", "71161\n", "23564\n", "10530\n", "32225\n", "15373\n", "19541\n", "15050\n", "17124\n", "57291\n", "23887\n", "24811\n", "19795\n", "52200\n", "34255\n", "52969\n", "19894\n", "71303\n", "25185\n", "11843\n", "18029\n", "13913\n", "20713\n", "31501\n", "50054\n", "21778\n", "49303\n", "48232\n", "19838\n", "54969\n", "14122\n", "23290\n", "53697\n", "50134\n", "12288\n", "91232\n", "28300\n", "28346\n", "20291\n", "102970\n", "48309\n", "23265\n", "49456\n", "42092\n", "53658\n", "31813\n", "23041\n", "25055\n", "37572\n", "32347\n", "45471\n", "14619\n", "14744\n", "32870\n", "13659\n", "6894\n", "25746\n", "24389\n", "33822\n", "20091\n", "22995\n", "7443\n", "104041\n", "20341\n", "21911\n", "23299\n", "17533\n", "28046\n", "19767\n", "62897\n", "28395\n", "24900\n", "88460\n", "42944\n", "21752\n", "9028\n", "4769\n", "75166\n", "43088\n", "25186\n", "13614\n", "22344\n", "101090\n", "25269\n", "15543\n", "30183\n", "85367\n", "30384\n", "22323\n", "27772\n", "22192\n", "29614\n", "36092\n", "11789\n", "15640\n", "7774\n", "40320\n", "148283\n", "15643\n", "60364\n", "21768\n", "73533\n", "116112\n", "18304\n", "29825\n", "56849\n", "19080\n", "12909\n", "16100\n", "37272\n", "28695\n", "44461\n", "33285\n", "50365\n", "31818\n", "29152\n", "39969\n", "9192\n", "23560\n", "23360\n", "60738\n", "17255\n", "26119\n", "17994\n", "22991\n", "12465\n", "14023\n", "15642\n", "19567\n", "20069\n", "34811\n", "48159\n", "67770\n", "39870\n", "23164\n", "20787\n", "17455\n", "32072\n", "75056\n", "167406\n", "42279\n", "51084\n", "43298\n", "63945\n", "73805\n", "19165\n", "30792\n", "37615\n", "44740\n", "30288\n", "10021\n", "99631\n", "11936\n", "21249\n", "87176\n", "4394\n", "31026\n", "13483\n", "19430\n", "25216\n", "12128\n", "30310\n", "30060\n", "46325\n", "23866\n", "16022\n", "10094\n", "34899\n", "33737\n", "20900\n", "17503\n", "28327\n", "10384\n", "32018\n", "31812\n", "33936\n", "104746\n", "25087\n", "15265\n", "30032\n", "109398\n", "40773\n", "37579\n", "108132\n", "23219\n", "58948\n", "34779\n", "20628\n", "20589\n", "46827\n", "17724\n", "44119\n", "37355\n", "19906\n", "9086\n", "31802\n", "20962\n", "11450\n", "51829\n", "65941\n", "54196\n", "53582\n", "55466\n", "65228\n", "31013\n", "20869\n", "29488\n", "27141\n", "20489\n", "24276\n", "34588\n", "81526\n", "67347\n", "24450\n", "24217\n", "36567\n", "7787\n", "194389\n", "18590\n", "27164\n", "37403\n", "18209\n", "35573\n", "15381\n", "33294\n", "27731\n", "10877\n", "80683\n", "166070\n", "21113\n", "29495\n", "24056\n", "18006\n", "57672\n", "88386\n", "23408\n", "75139\n", "25699\n", "21502\n", "38305\n", "31826\n", "33194\n", "28271\n", "28731\n", "10957\n", "95272\n", "15007\n", "36928\n", "13938\n", "40550\n", "23251\n", "11680\n", "72881\n", "51321\n", "26480\n", "22618\n", "20862\n", "22093\n", "57052\n", "15609\n", "8729\n", "36834\n", "13621\n", "8473\n", "30124\n", "17212\n", "22083\n", "28426\n", "66431\n", "22413\n", "20449\n", "17290\n", "18594\n", "16764\n", "17307\n", "85170\n", "25586\n", "25698\n", "19637\n", "10190\n", "73158\n", "43586\n", "18815\n", "25110\n", "13293\n", "45918\n", "25978\n", "8010\n", "125437\n", "34755\n", "21201\n", "48231\n", "10853\n", "17543\n", "41689\n", "47827\n", "57155\n", "48925\n", "24979\n", "25095\n", "15217\n", "29093\n", "22021\n", "41395\n", "18853\n", "104590\n", "87087\n", "11000\n", "128006\n", "54980\n", "41742\n", "48524\n", "174970\n", "34739\n", "44353\n", "11200\n", "13355\n", "17992\n", "39591\n", "11866\n", "30628\n", "19389\n", "19937\n", "15710\n", "18233\n", "26836\n", "53727\n", "11367\n", "65416\n", "12474\n", "8609\n", "15115\n", "83864\n", "71785\n", "106668\n", "26388\n", "11337\n", "41884\n", "58046\n", "7523\n", "82815\n", "68870\n", "21779\n", "8907\n", "23868\n", "96145\n", "74242\n", "27106\n", "36227\n", "13230\n", "9167\n", "16547\n", "54399\n", "32714\n", "77306\n", "21307\n", "17956\n", "18570\n", "41884\n", "173928\n", "8588\n", "44126\n", "33195\n", "97831\n", "127203\n", "17962\n", "47469\n", "7715\n", "24129\n", "33748\n", "43799\n", "63598\n", "12648\n", "62991\n", "38306\n", "61116\n", "89892\n", "20528\n", "23585\n", "24851\n", "33791\n", "80693\n", "51664\n", "8422\n", "38428\n", "10615\n", "71738\n", "9277\n", "50787\n", "13739\n", "42475\n", "45048\n", "20884\n", "26134\n", "23630\n", "29976\n", "28050\n", "22609\n", "32241\n", "45251\n", "140749\n", "4997\n", "52451\n", "23669\n", "13442\n", "52114\n", "32993\n", "6038\n", "22107\n", "53614\n", "68165\n", "37980\n", "143985\n", "16377\n", "21972\n", "27851\n", "27849\n", "46851\n", "28759\n", "10240\n", "15197\n", "53865\n", "9193\n", "76659\n", "12839\n", "10041\n", "11816\n", "14692\n", "19478\n", "45716\n", "100724\n", "6969\n", "14831\n", "71013\n", "35327\n", "10847\n", "26972\n", "7131\n", "10341\n", "28768\n", "83616\n", "62761\n", "62351\n", "47288\n", "44775\n", "29773\n", "107018\n", "19785\n", "36797\n", "11171\n", "19374\n", "24692\n", "24357\n", "16488\n", "39568\n", "23112\n", "90254\n", "8692\n", "38031\n", "8122\n", "13062\n", "29073\n", "17031\n", "40219\n", "39146\n", "8807\n", "26039\n", "25373\n", "35064\n", "12524\n", "105198\n", "28781\n", "284410\n", "51431\n", "64718\n", "23022\n", "39787\n", "37443\n", "36015\n", "45699\n", "48386\n", "89050\n", "41142\n", "27273\n", "19748\n", "6041\n", "51842\n", "10459\n", "22964\n", "164242\n", "98297\n", "12195\n", "68001\n", "32167\n", "66401\n", "34051\n", "35973\n", "23820\n", "17800\n", "5030\n", "179276\n", "32123\n", "12608\n", "193789\n", "14745\n", "97554\n", "15688\n", "57926\n", "87091\n", "24925\n", "28467\n", "20606\n", "6595\n", "19283\n", "90015\n", "16573\n", "34641\n", "83009\n", "38801\n", "61777\n", "8357\n", "33530\n", "17208\n", "71928\n", "7782\n", "49081\n", "53035\n", "72947\n", "13177\n", "20142\n", "5550\n", "106089\n", "25790\n", "13956\n", "26903\n", "15652\n", "35144\n", "24185\n", "7136\n", "45536\n", "42170\n", "13197\n", "26809\n", "13380\n", "29539\n", "44655\n", "46311\n", "15101\n", "23003\n", "38992\n", "15818\n", "21519\n", "55646\n", "66978\n", "26172\n", "27922\n", "19504\n", "43141\n", "4635\n", "56931\n", "10363\n", "26696\n", "103453\n", "46872\n", "9258\n", "64904\n", "86962\n", "19409\n", "26849\n", "22412\n", "23802\n", "17771\n", "21669\n", "20790\n", "48573\n", "30197\n", "71479\n", "67085\n", "45416\n", "15660\n", "14242\n", "22997\n", "95750\n", "29069\n", "17514\n", "25280\n", "19560\n", "35367\n", "55002\n", "32114\n", "36431\n", "17663\n", "13013\n", "79251\n", "20827\n", "11374\n", "22503\n", "13232\n", "17668\n", "33707\n", "18329\n", "45412\n", "15797\n", "9512\n", "9663\n", "14586\n", "35785\n", "20830\n", "50887\n", "98631\n", "13493\n", "40377\n", "32609\n", "35932\n", "28667\n", "11159\n", "51768\n", "3945\n", "40581\n", "16388\n", "95636\n", "5077\n", "115518\n", "43800\n", "18787\n", "43771\n", "21765\n", "28849\n", "32473\n", "29946\n", "17516\n", "9285\n", "39848\n", "50007\n", "15046\n", "52365\n", "33007\n", "49327\n", "53171\n", "21108\n", "16191\n", "16011\n", "36003\n", "4901\n", "39332\n", "66027\n", "17127\n", "37691\n", "10763\n", "34771\n", "69576\n", "29663\n", "30102\n", "12650\n", "14087\n", "26179\n", "12775\n", "24256\n", "42583\n", "31243\n", "5589\n", "20128\n", "8168\n", "18181\n", "94570\n", "5986\n", "57916\n", "18689\n", "110460\n", "28074\n", "30144\n", "13780\n", "30070\n", "8237\n", "19763\n", "104740\n", "7928\n", "24874\n", "14802\n", "45449\n", "62486\n", "13786\n", "43147\n", "57036\n", "31734\n", "49762\n", "42143\n", "16782\n", "19146\n", "108892\n", "44044\n", "31500\n", "48709\n", "8303\n", "11806\n", "23277\n", "52249\n", "45429\n", "6497\n", "31326\n", "61564\n", "134939\n", "22398\n", "57451\n", "16586\n", "12650\n", "38008\n", "44055\n", "3875\n", "33172\n", "66540\n", "40993\n", "37376\n", "4632\n", "115622\n", "6780\n", "61571\n", "20153\n", "26856\n", "25384\n", "9328\n", "8779\n", "18113\n", "11278\n", "25269\n", "112891\n", "45452\n", "15116\n", "71845\n", "19609\n", "52398\n", "45115\n", "33967\n", "38341\n", "11043\n", "9456\n", "22052\n", "56187\n", "41500\n", "51815\n", "13804\n", "23138\n", "53209\n", "29245\n", "40869\n", "9384\n", "39881\n", "29436\n", "34034\n", "33679\n", "28950\n", "63654\n", "19787\n", "98880\n", "25562\n", "40543\n", "24069\n", "25836\n", "15878\n", "38175\n", "16967\n", "18943\n", "63301\n", "23951\n", "15446\n", "3461\n", "18690\n", "10137\n", "24772\n", "6961\n", "19232\n", "13926\n", "42340\n", "8294\n", "6608\n", "60106\n", "17735\n", "30661\n", "40103\n", "30759\n", "59201\n", "13214\n", "35553\n", "63445\n", "34737\n", "69139\n", "18484\n", "111796\n", "35790\n", "37600\n", "21282\n", "12413\n", "60810\n", "42273\n", "31679\n", "7061\n", "74391\n", "46542\n", "23633\n", "31111\n", "33230\n", "45527\n", "45108\n", "9511\n", "17143\n", "37998\n", "19996\n", "41351\n", "9725\n", "65230\n", "23323\n", "29056\n", "9821\n", "38524\n", "25087\n", "23322\n", "67666\n", "29556\n", "81663\n", "65833\n", "18464\n", "17686\n", "16030\n", "46668\n", "51351\n", "90499\n", "16298\n", "28776\n", "39614\n", "53566\n", "9431\n", "10846\n", "29218\n", "109410\n", "74105\n", "29199\n", "7146\n", "77358\n", "35502\n", "34453\n", "9550\n", "13049\n", "27647\n", "22190\n", "37137\n", "47345\n", "27116\n", "24550\n", "101688\n", "34294\n", "59520\n", "32366\n", "87876\n", "18858\n", "36865\n", "89178\n", "21444\n", "17232\n", "23615\n", "28765\n", "10994\n", "54777\n", "18085\n", "40217\n", "19786\n", "63074\n", "26020\n", "32832\n", "32247\n", "24137\n", "62424\n", "57890\n", "9961\n", "23342\n", "71723\n", "36015\n", "16045\n", "35471\n", "4506\n", "36027\n", "22707\n", "20166\n", "40483\n", "66449\n", "47560\n", "32386\n", "18231\n", "114016\n", "10720\n", "34056\n", "76663\n", "31760\n", "19247\n", "24014\n", "23042\n", "18323\n", "127945\n", "60265\n", "32962\n", "29259\n", "35464\n", "11523\n", "43598\n", "91005\n", "53229\n", "29341\n", "12671\n", "120471\n", "28743\n", "36519\n", "13585\n", "12825\n", "71072\n", "27882\n", "46374\n", "30267\n", "20148\n", "20567\n", "16545\n", "6056\n", "31805\n", "32467\n", "56885\n", "29327\n", "28078\n", "39635\n", "31911\n", "26564\n", "37818\n", "30074\n", "13440\n", "36226\n", "17007\n", "32364\n", "22263\n", "24829\n", "7915\n", "60254\n", "40637\n", "42739\n", "7854\n", "24139\n", "22151\n", "39214\n", "121463\n", "23540\n", "13606\n", "56619\n", "12298\n", "64701\n", "35547\n", "16724\n", "17520\n", "17140\n", "29108\n", "15654\n", "33213\n", "64542\n", "19322\n", "77146\n", "19303\n", "69812\n", "35479\n", "39096\n", "25113\n", "5919\n", "15166\n", "38326\n", "72994\n", "31490\n", "100794\n", "13335\n", "18550\n", "30793\n", "39427\n", "13983\n", "74592\n", "6664\n", "38237\n", "9605\n", "12087\n", "42821\n", "59814\n", "30301\n", "26095\n", "22240\n", "67011\n", "19124\n", "25220\n", "29848\n", "36345\n", "36188\n", "21649\n", "19544\n", "15732\n", "64194\n", "20962\n", "19652\n", "48806\n", "24568\n", "33756\n", "44089\n", "39983\n", "19051\n", "34736\n", "61233\n", "177020\n", "26615\n", "13507\n", "25660\n", "17589\n", "29261\n", "10293\n", "27736\n", "27940\n", "15255\n", "11026\n", "8348\n", "20316\n", "40303\n", "17616\n", "15281\n", "22341\n", "26697\n", "25222\n", "16973\n", "40209\n", "69951\n", "7474\n", "18660\n", "5063\n", "8000\n", "89304\n", "53231\n", "15540\n", "29630\n", "21605\n", "62190\n", "12977\n", "71763\n", "11838\n", "29982\n", "32016\n", "16470\n", "26114\n", "29114\n", "34897\n", "23642\n", "40030\n", "84175\n", "12598\n", "14897\n", "34563\n", "49228\n", "27631\n", "16860\n", "33602\n", "6100\n", "21693\n", "9426\n", "27082\n", "37176\n", "39313\n", "21576\n", "14893\n", "38972\n", "9742\n", "41317\n", "43802\n", "17327\n", "22122\n", "160538\n", "250913\n", "56600\n", "2519\n", "17315\n", "111641\n", "17922\n", "24170\n", "54643\n", "49344\n", "21155\n", "22293\n", "72761\n", "55146\n", "16375\n", "39000\n", "30355\n", "30047\n", "55524\n", "9169\n", "132508\n", "36419\n", "56967\n", "27252\n", "8403\n", "51599\n", "56782\n", "44610\n", "6625\n", "66094\n", "12836\n", "79563\n", "11519\n", "19106\n", "25208\n", "16648\n", "22960\n", "22927\n", "14770\n", "32999\n", "16355\n", "125444\n", "14206\n", "20281\n", "20456\n", "30756\n", "28724\n", "113276\n", "36496\n", "43443\n", "16472\n", "11729\n", "21902\n", "36264\n", "7235\n", "30925\n", "18732\n", "17952\n", "79669\n", "21089\n", "36485\n", "31524\n", "40610\n", "38284\n", "87675\n", "33065\n", "18363\n", "29926\n", "28538\n", "50043\n", "8910\n", "37069\n", "48424\n", "8535\n", "13345\n", "22516\n", "14002\n", "15636\n", "14025\n", "55016\n", "19961\n", "21947\n", "5127\n", "18438\n", "30466\n", "11224\n", "16339\n", "18190\n", "10549\n", "19954\n", "50862\n", "24144\n", "34946\n", "18088\n", "39541\n", "16903\n", "29730\n", "25167\n", "44128\n", "67364\n", "17269\n", "17215\n", "63003\n", "53261\n", "18144\n", "29003\n", "79607\n", "48551\n", "47416\n", "25074\n", "39751\n", "58410\n", "14049\n", "27368\n", "13439\n", "8929\n", "38149\n", "15564\n", "9508\n", "23327\n", "18319\n", "8973\n", "43694\n", "72736\n", "59879\n", "10187\n", "14335\n", "41177\n", "15348\n", "8479\n", "75480\n", "30380\n", "55410\n", "51633\n", "22207\n", "19588\n", "20655\n", "22188\n", "139739\n", "31191\n", "21373\n", "40888\n", "58130\n", "33038\n", "132327\n", "24571\n", "30098\n", "34295\n", "14826\n", "76785\n", "43659\n", "24593\n", "11341\n", "18039\n", "81885\n", "17308\n", "30381\n", "10634\n", "47916\n", "17513\n", "15479\n", "10394\n", "49788\n", "51817\n", "9040\n", "12993\n", "32345\n", "39542\n", "21720\n", "22717\n", "66047\n", "22983\n", "46558\n", "9426\n", "26911\n", "13842\n", "20650\n", "23660\n", "28274\n", "28592\n", "44559\n", "19999\n", "25932\n", "23058\n", "75987\n", "48256\n", "59237\n", "28542\n", "147745\n", "22681\n", "15338\n", "19443\n", "36394\n", "34039\n", "10572\n", "25346\n", "23998\n", "37557\n", "64850\n", "28128\n", "72165\n", "48072\n", "21609\n", "66918\n", "19886\n", "23642\n", "49798\n", "28849\n", "39932\n", "46298\n", "25218\n", "20925\n", "35399\n", "19820\n", "32254\n", "14706\n", "41857\n", "7742\n", "12296\n", "22228\n", "37720\n", "67744\n", "9053\n", "33436\n", "14708\n", "48571\n", "58986\n", "26514\n", "47744\n", "10328\n", "25141\n", "11769\n", "57483\n", "15323\n", "127432\n", "14811\n", "12617\n", "21428\n", "34525\n", "99226\n", "25185\n", "33125\n", "29554\n", "25338\n", "18171\n", "25982\n", "16400\n", "18150\n", "4454\n", "34868\n", "24827\n", "42420\n", "31782\n", "54741\n", "7692\n", "8993\n", "23630\n", "20676\n", "20005\n", "30470\n", "17034\n", "42061\n", "19922\n", "18862\n", "30466\n", "25845\n", "48268\n", "18490\n", "14004\n", "84123\n", "49320\n", "36852\n", "17408\n", "11022\n", "33645\n", "19907\n", "78556\n", "62735\n", "19349\n", "101104\n", "22236\n", "19692\n", "29193\n", "21832\n", "21564\n", "29748\n", "44001\n", "37540\n", "20949\n", "86186\n", "25030\n", "51127\n", "14523\n", "26805\n", "72962\n", "63725\n", "23066\n", "63337\n", "27243\n", "51056\n", "49787\n", "23581\n", "36271\n", "9336\n", "20681\n", "33408\n", "35625\n", "34943\n", "19540\n", "120587\n", "124853\n", "11478\n", "47277\n", "49715\n", "24848\n", "9795\n", "18810\n", "20085\n", "27734\n", "37922\n", "30853\n", "23768\n", "18777\n", "45074\n", "15054\n", "27835\n", "42147\n", "41325\n", "26854\n", "14495\n", "52884\n", "8527\n", "57968\n", "327458\n", "58039\n", "19081\n", "23211\n", "19146\n", "20403\n", "5806\n", "14802\n", "17840\n", "26191\n", "33202\n", "17534\n", "23597\n", "62754\n", "26380\n", "42945\n", "47797\n", "30349\n", "28097\n", "107762\n", "48844\n", "44160\n", "25230\n", "8739\n", "80802\n", "16057\n", "62180\n", "56966\n", "28169\n", "36219\n", "59164\n", "20752\n", "21981\n", "7907\n", "40026\n", "15023\n", "49914\n", "74726\n", "22969\n", "15548\n", "24706\n", "20288\n", "57038\n", "28840\n", "33479\n", "56734\n", "19591\n", "59087\n", "41903\n", "49815\n", "49315\n", "19866\n", "20195\n", "10739\n", "55401\n", "63684\n", "10771\n", "35872\n", "77872\n", "4437\n", "28575\n", "11052\n", "65902\n", "14712\n", "65606\n", "32317\n", "43503\n", "26347\n", "23080\n", "39775\n", "106771\n", "16547\n", "40287\n", "45980\n", "46179\n", "18853\n", "53281\n", "27090\n", "26380\n", "23168\n", "21608\n", "27275\n", "134070\n", "15867\n", "21972\n", "22163\n", "47626\n", "16273\n", "21740\n", "35301\n", "65790\n", "57572\n", "22210\n", "14325\n", "93876\n", "50285\n", "24437\n", "104008\n", "17258\n", "49679\n", "25415\n", "28574\n", "15436\n", "31887\n", "18523\n", "15037\n", "71157\n", "21003\n", "22113\n", "59324\n", "78264\n", "19501\n", "30418\n", "115849\n", "91333\n", "13122\n", "13559\n", "32822\n", "25757\n", "37229\n", "26617\n", "24415\n", "30619\n", "21695\n", "25066\n", "23563\n", "48395\n", "62566\n", "25672\n", "12214\n", "25050\n", "43581\n", "21400\n", "49536\n", "18261\n", "33348\n", "52302\n", "18702\n", "96600\n", "20521\n", "16510\n", "35622\n", "16260\n", "18016\n", "19666\n", "63232\n", "16868\n", "41647\n", "45005\n", "23056\n", "122503\n", "38429\n", "38193\n", "27710\n", "27892\n", "15030\n", "32550\n", "13774\n", "33917\n", "13986\n", "21889\n", "14970\n", "26231\n", "42767\n", "81619\n", "38887\n", "23977\n", "19052\n", "42574\n", "47040\n", "39929\n", "33977\n", "217436\n", "23247\n", "17763\n", "88651\n", "74146\n", "25010\n", "14636\n", "9227\n", "37556\n", "29396\n", "17537\n", "55611\n", "71041\n", "46275\n", "66128\n", "37425\n", "28442\n", "22482\n", "29482\n", "32599\n", "47558\n", "31973\n", "96785\n", "20924\n", "28801\n", "60943\n", "33656\n", "26205\n", "404\n", "29785\n", "21675\n", "36260\n", "17443\n", "46756\n", "36746\n", "42346\n", "33578\n", "17737\n", "78822\n", "31413\n", "27201\n", "27689\n", "28796\n", "82798\n", "14420\n", "14107\n", "49142\n", "138739\n", "58194\n", "54305\n", "11245\n", "69107\n", "45282\n", "17529\n", "22027\n", "73071\n", "13002\n", "45760\n", "31313\n", "16545\n", "9052\n", "22132\n", "21056\n", "26839\n", "17032\n", "29515\n", "35908\n", "20977\n", "45238\n", "33849\n", "15905\n", "71086\n", "35986\n", "42389\n", "31134\n", "30454\n", "130394\n", "19899\n", "21465\n", "59554\n", "25858\n", "36495\n", "34491\n", "17224\n", "20341\n", "23239\n", "19269\n", "27102\n", "33126\n", "10524\n", "9498\n", "12317\n", "19823\n", "23037\n", "23049\n", "20961\n", "10152\n", "41258\n", "75051\n", "18680\n", "26762\n", "222125\n", "47847\n", "11046\n", "68058\n", "38346\n", "31222\n", "36487\n", "13904\n", "35289\n", "27074\n", "72180\n", "106453\n", "34282\n", "13311\n", "23967\n", "61197\n", "16087\n", "13437\n", "26489\n", "16548\n", "5483\n", "56711\n", "89150\n", "21238\n", "79928\n", "76774\n", "16039\n", "11283\n", "28051\n", "39977\n", "15115\n", "35714\n", "35281\n", "52446\n", "44945\n", "20406\n", "27360\n", "103787\n", "32540\n", "118614\n", "59484\n", "63916\n", "17963\n", "18493\n", "34969\n", "14666\n", "7543\n", "24710\n", "19695\n", "83015\n", "16068\n", "40432\n", "29123\n", "36572\n", "46734\n", "23773\n", "22841\n", "14881\n", "43489\n", "27155\n", "21565\n", "12541\n", "26963\n", "21164\n", "9397\n", "24058\n", "16693\n", "21663\n", "25829\n", "59508\n", "37501\n", "119153\n", "40754\n", "11213\n", "22613\n", "17188\n", "16585\n", "24938\n", "23995\n", "94934\n", "33099\n", "30724\n", "21130\n", "22925\n", "27664\n", "14870\n", "10428\n", "23571\n", "83624\n", "34312\n", "12967\n", "19680\n", "32432\n", "42175\n", "52081\n", "55526\n", "20168\n", "17945\n", "22927\n", "65039\n", "32427\n", "29071\n", "15344\n", "90047\n", "105928\n", "90900\n", "13896\n", "53185\n", "96047\n", "11543\n", "32245\n", "97696\n", "19261\n", "115355\n", "62674\n", "8561\n", "85035\n", "48537\n", "44508\n", "54338\n", "76367\n", "21795\n", "14516\n", "47728\n", "53069\n", "115752\n", "42930\n", "15743\n", "50544\n", "42531\n", "33690\n", "20662\n", "29018\n", "20491\n", "15314\n", "36138\n", "12404\n", "20181\n", "17598\n", "23377\n", "23252\n", "17495\n", "15904\n", "23821\n", "33514\n", "107759\n", "59658\n", "25009\n", "28260\n", "15397\n", "20716\n", "18668\n", "64986\n", "15771\n", "121279\n", "42252\n", "39733\n", "19799\n", "34229\n", "23765\n", "28385\n", "24183\n", "Extracted 886 responses\n" ] } ], "source": [ "from datasets import load_dataset\n", "import pandas as pd\n", "import numpy as np\n", "\n", "thing = load_dataset(\"ccdv/arxiv-summarization\", split = \"train\")\n", "other = []\n", "\n", "for i in thing:\n", " a, b = i[\"abstract\"], i[\"article\"]\n", " print(len(a)+len(b))\n", " if len(a)+len(b) < 4096:\n", " if len(other)<1000:\n", " other.append(i)\n", "thing = other\n", "\n", "convos3 = []\n", "problems3 = []\n", "\n", "for example in thing:\n", " convos3.append(example[\"abstract\"])\n", " problems3.append(example[\"article\"])\n", "\n", "print(f\"Extracted {len(convos3)} responses\")\n", "\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "2Em1-A6R3Td0", "outputId": "3f6f5a88-6554-42c2-a81b-71979aa4483a" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "\n", "Processed 10505 examples\n", "{'messages': [{'role': 'system', 'content': 'Summarize the data given'}, {'role': 'user', 'content': 'for fixed integers @xmath0 and @xmath1 , we consider the admissible sequences of @xmath2 lattice paths in a colored @xmath3 square given in @xcite . each admissible sequence of paths can be associated with a partition @xmath10 of @xmath4 . in section \\n [ paths ] , we show that the number of self - conjugate admissible sequences of paths associated with @xmath10 is equal to the number of standard young tableaux of shape @xmath10 , and thus can be calculated using the hook length formula . \\n we extend this result to include the non - self - conjugate admissible sequences of paths and show that the number of all such admissible sequences of paths is equal to the sum of squares of the number of standard young tableaux of partitions of @xmath4 with height less than or equal to @xmath11 . using the rsk correspondence in @xcite , \\n it is shown in ( @xcite , corollary 7.23.12 ) that the sum of squares of the number of standard young tableaux of partitions of @xmath4 with height less than or equal to @xmath11 is equal to the number of @xmath6-avoiding permutations of @xmath7 . in section [ multiplicities ] \\n , we apply our results to the representation theory of the affine kac - moody algebra @xmath8 . \\n let @xmath12 , @xmath13 and @xmath14 denote the simple roots , simple coroots , and fundamental weights respectively . \\n note that @xmath15 \\n . for @xmath16 , set @xmath17 and @xmath18 . as shown in @xcite , @xmath19 \\n are maximal dominant weights of the irreducible @xmath8-module @xmath9 . \\n we show that the multiplicity of the weight @xmath19 in @xmath9 is the number of @xmath6-avoiding permutations of @xmath7 , which proves conjecture 4.13 in @xcite . \\n for fixed integers @xmath0 and @xmath1 , consider the @xmath3 square containing @xmath20 unit boxes in the fourth quadrant so that the top left corner of the square is at the origin . \\n we assign color @xmath21 to a box if its upper left corner has coordinates @xmath22 . \\n this gives the following @xmath3 colored square @xmath23 : a lattice path @xmath25 on @xmath23 is a path joining the lower left corner @xmath26 to the upper right corner @xmath27 moving unit lengths up or right . for two lattice paths @xmath28 on @xmath23 \\n we say that @xmath29 if the boxes above @xmath30 are also above @xmath25 . \\n now , we draw @xmath2 lattice paths , @xmath31 on @xmath23 such that @xmath32 . for integers \\n @xmath33 , where @xmath34 , @xmath35 , we define @xmath36 to be the number of @xmath37-colored boxes between @xmath38 and @xmath39 . \\n we define @xmath40 to be the number of @xmath37-colored boxes below @xmath41 and @xmath42 to be the number of @xmath37-colored boxes above @xmath43 . \\n denote by @xmath49 the set of all admissible sequences of @xmath2 paths . \\n notice that there are @xmath4 0-colored boxes in @xmath23 and hence for any admissible sequence of paths , @xmath50 . \\n in addition , it follows from definition [ pathsdef](2 ) that @xmath51 for any admissible sequence of paths . \\n thus , we can and do associate an admissible sequence of paths @xmath44 on @xmath23 with a partition @xmath52 of @xmath4 . in this case , we say that this admissible sequence of paths is of type @xmath10 and often draw @xmath10 as a young diagram . \\n figure [ adseq](a ) is an element of @xmath53 , where @xmath54 and @xmath55 are shown in figures [ adseq](b ) , [ adseq](c ) , and [ adseq](d ) , respectively . \\n notice that this admissible sequence of paths is of type @xmath56 .'}, {'role': 'assistant', 'content': 'for @xmath0 and @xmath1 , we consider certain admissible sequences of @xmath2 lattice paths in a colored @xmath3 square . \\n we show that the number of such admissible sequences of lattice paths is given by the sum of squares of the number of standard young tableaux of partitions of @xmath4 with height @xmath5 , which is also the number of @xmath6-avoiding permutations of @xmath7 . \\n finally , we apply this result to the representation theory of the affine lie algebra @xmath8 and show that this quantity gives the multiplicity of certain maximal dominant weights in the irreducible module @xmath9 .'}]}\n" ] } ], "source": [ "from datasets import load_dataset\n", "import re\n", "import random\n", "\n", "def parse_chat(chat_string):\n", "\n", " messages = []\n", "\n", "\n", " parts = re.split(r'(USER:|ASSISTANT:|FUNCTION RESPONSE:)', chat_string)\n", "\n", "\n", " current_role = None\n", " for part in parts:\n", " part = part.strip()\n", " if not part:\n", " continue\n", "\n", " if part == 'USER:':\n", " current_role = 'USER'\n", " elif part == 'ASSISTANT:':\n", " current_role = 'ASSISTANT'\n", " elif part == 'FUNCTION RESPONSE:':\n", " current_role = 'ASSISTANT'\n", " else:\n", "\n", " content = part.replace('<|endoftext|>', '').strip()\n", " if content and current_role:\n", " messages.append({\"from\": current_role, \"value\": content})\n", "\n", " return messages\n", "\n", "\n", "thing = load_dataset(\"glaiveai/glaive-function-calling-v2\", split=\"train\")\n", "x = []\n", "num = 0\n", "for i in thing:\n", " if num<3000:\n", " x.append(i)\n", " num+=1\n", "thing = x\n", "#thing = thing.shuffle(seed = 23)\n", "\n", "\n", "for i in thing:\n", " system = i[\"system\"]\n", " content = [{\"role\": \"system\", \"content\": system}]\n", "\n", " string = i[\"chat\"]\n", " if len(string) > 4096:\n", " continue\n", " chat = parse_chat(i[\"chat\"])\n", "\n", " for x in chat:\n", " if x['from'] == 'USER':\n", " content.append({\"role\": \"user\", \"content\": x['value']})\n", " elif x['from'] == 'ASSISTANT':\n", " content.append({\"role\": \"assistant\", \"content\": x['value']})\n", "\n", " example = {\"messages\": content}\n", " data.append(example)\n", "\n", "print(type(data))\n", "print(f\"Processed {len(data)} examples\")\n", "print(data[0])\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "sQyKXNaZ9Rm7", "outputId": "510520c9-1839-45e6-adf9-2a35cb95cb15" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "{'messages': [{'role': 'system', 'content': 'Summarize the data given'}, {'role': 'user', 'content': \"* the anomalous x - ray pulsars ( axps ) * are a group of x - ray pulsars whose spin periods fall in a narrow range ( @xmath0 s ) , whose x - ray spectra are very soft , and which show no evidence that they accrete from a binary companion ( see mereghetti 1999 for a recent review ) . these objects may be isolated neutron stars with extremely strong ( @xmath1 g ) surface magnetic fields , or they may be accreting from a `` fallback '' accretion disk . \\n optical measurements could potentially help discriminate between these models . \\n an optical counterpart to one axp , 4u 0142 + 61 , has recently been identified and shown to have peculiar optical colors ( hulleman et al . \\n * the radio - quiet neutron stars ( rqnss ) * are a group of compact x - ray sources found near the center of young supernova remnants . \\n their x - ray spectra are roughly consistent with young , cooling neutron stars , but they show no evidence for the non - thermal emission associated with `` classical '' young pulsars like the crab ( see brazier & johnston 1999 for a review ) \\n . the x - ray spectral properties of the rqnss and the axps are similar ( see , e.g. , chakrabarty et al . \\n below in table 1 , the general properties of the three rqnss as our targets in the southern sky are listed . \\n clcccc & & @xmath2 & age & @xmath3 & + source & snr & ( kpc ) & ( @xmath4 yr ) & ( kev ) & refs + 1e 08204247 & pup a & 2.0 & 3.7 & 0.28 & 1 - 3 + 1e 16145055 & rcw 103 & 3.3 & 1 - 3 & 0.56 & 4 - 6 + 1e 12075209 & pks 120952 & 1.5 & 7 & 0.25 & 7 - 9 + + + + \\n our observations were made using the magellan instant camera ( magic ) on the magellan-1/walter baade 6.5-meter telescope at las campanas observatory , chile . \\n magic is a ccd filter photometer built by mit and cfa for the @xmath5 focus of the baade telescope . \\n the current detector is a 2048@xmath62048 site ccd with a 69 mas / pixel scale and a 142@xmath6142 arcsec field of view . \\n we used the sloan filter set , which have the following central wavelengths ( fukugita et al . \\n 1996 ) : @xmath7=3540 ; @xmath8=4770 ; @xmath9=6230 ; @xmath10=7620 ; and @xmath11=9130 . \\n brazier , k.t.s . \\n , & johnston , s. 1999 , mnras , 303 , l1 bignami , g.f . , caraveo , g.a . , & mereghetti , s. 1992 , apj , 389 , l67 chakrabarty , d. et al . \\n 2001 , apj , 548 , 800 fukugita , m. et al . \\n 1996 , aj , 111 , 1748 garmire , g.p . , \\n pavlov , g.g . \\n , & garmire , a.b . \\n 2000 , iauc , 7350 , 2 gotthelf , e.v . , \\n petre , r. , & hwang , u. 1997 , 487 , l175 helfand , d.j . , & becker , r.h . \\n 1984 , nature , 307 , 215 hulleman , f. , kerkwijk , m.h . , & kulkarni , s.r . \\n 2000 , nature , 408 , 689 mereghetti , s. 1999 , in the neutron star black hole connection , ed . c. kouveliotou et al . , ( dordrecht : kluwer ) mereghetti , s. , caraveo , p. , & bignami , g.f . \\n 1992 , a & a , 263 , 172 mereghetti , s. , bignami , g.f . \\n , & caraveo , p.a . \\n 1996 , apj , 464 , 842 pavlov , g. g. , zavlin , v.e . , & trmper , j. 1999 , apj , 511 , l45 petre , r. , becker , c.m . , & \\n winkler , p.f . \\n 1996 , apj , 465 , l43 petre , et al . 1982 , apj , 258 , 22 seward , f.d . \\n 1990 , apjss , 73 , 781 tuohy , i. , & garmire , g. 1980 , apj , 239 , 107\"}, {'role': 'assistant', 'content': 'we report on our search for the optical counterparts of the southern hemisphere anomalous x - ray pulsar 1e1048.1 - 5937 and the radio - quiet neutron stars in supernova remnants puppis a , rcw 103 , and pks 1209 - 52 . \\n the observations were carried out with the new mit / cfa magic camera on the magellan - i 6.5 m telescope in chile . \\n we present deep multiband optical images of the x - ray error circles for each of these targets and discuss the resulting candidates and limits . \\n # 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in'}]}\n" ] } ], "source": [ "print(data[1])\n", "data = data[:17000]" ] }, { "cell_type": "markdown", "metadata": { "id": "JVRFqoSdIEVK" }, "source": [ "We have to format the dataset to follow our GRPO style formatting:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 141 }, "id": "-ZFSjGVheJBo", "outputId": "8bf1c6ae-779b-4fcc-8492-d5d993ba77e0" }, "outputs": [ { "ename": "NameError", "evalue": "name 'problems3' is not defined", "output_type": "error", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", "\u001b[0;32m/tmp/ipython-input-3851025860.py\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mproblems3\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", "\u001b[0;31mNameError\u001b[0m: name 'problems3' is not defined" ] } ], "source": [ "problems3[0]" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "Z9ydcV_Abfi6", "outputId": "f37149b6-7373-4d98-8293-5503bbd37056" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "886\n" ] } ], "source": [ "def format_dataset(x, y):\n", " expected_answer = x\n", " problem = y\n", "\n", "\n", " # Add our custom formatting\n", " final_prompt = expected_answer\n", " return {\"messages\":\n", " [\n", " {\"role\" : \"system\", \"content\" : \"Summarize the data given\"},\n", " {\"role\" : \"user\", \"content\" : problem},\n", " {\"role\" : \"assistant\", \"content\" : final_prompt},\n", "\n", " ]\n", " }\n", "data = []\n", "for i in range(len(convos3)):\n", " data.append(format_dataset(convos3[i], problems3[i]))\n", "print(len(data))" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "cuLw2RpBLJ8Y", "outputId": "225618a4-16b2-471d-9775-c98d87a25e34" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "Length of convos: 3000\n", "Length of problems: 3000\n", "Length of tools: 3000\n" ] } ], "source": [ "print(f\"Length of convos: {len(convos)}\")\n", "print(f\"Length of problems: {len(problems)}\")\n", "print(f\"Length of tools: {len(tools)}\")" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "Q7NvWHwrtyX1", "outputId": "06f8682d-d3ec-4567-bcb2-59beb050386a" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "4886\n" ] } ], "source": [ "def format_dataset2(x, y, z):\n", " expected_answer = x\n", " problem = y\n", "\n", "\n", " # Add our custom formatting\n", " final_prompt = expected_answer\n", " return {\"messages\":\n", " [\n", " {\"role\" : \"system\", \"content\" : (system_prompt+\" Here are your tools:\"+z)},\n", " {\"role\" : \"user\", \"content\" : problem},\n", " {\"role\" : \"assistant\", \"content\" : final_prompt},\n", "\n", " ]\n", " }\n", "#data = []\n", "for i in range(len(convos2)):\n", " data.append(format_dataset2(convos2[i], problems2[i], tools2[i]))\n", "print(len(data))" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "kGDRiQ64EVBR", "outputId": "b8f93c15-e07f-48ae-e8cf-4e3027bbe814" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "7886\n" ] } ], "source": [ "def format_dataset3(x, y, z):\n", " if(len(x)>0):\n", " expected_answer = x[0]\n", "\n", " else:\n", " expected_answer = \"No answer found\"\n", "\n", "\n", " problem = y\n", "\n", "\n", " # Add our custom formatting\n", " final_prompt = expected_answer\n", " return {\"messages\":\n", " [\n", " {\"role\" : \"system\", \"content\" : (system_prompt)},\n", " {\"role\" : \"user\", \"content\" : problem+\" Here are your sources:\"+z},\n", " {\"role\" : \"assistant\", \"content\" : final_prompt},\n", "\n", " ]\n", " }\n", "#data = []\n", "for i in range(len(convos)):\n", " data.append(format_dataset3(convos[i], problems[i], tools[i]))\n", "print(len(data))" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "28zhWKDlfxA-" }, "outputs": [], "source": [ "import random\n", "random.shuffle(data)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "GPcuhxPvLXzz", "outputId": "52e25993-1265-41c9-e1c5-5efada04c1a3" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "{'messages': [{'role': 'system', 'content': 'You are given a problem.\\nThink about the problem and provide your working out.\\nPlace it between and .\\nThen, provide your solution between '}, {'role': 'user', 'content': 'how to sew fly front zipper Here are your sources:/4. Sew the fronts together along the center front line. From the bottom of the zipper down, sew with a regular stitch. From the bottom of the zipper and up, use a basting stitch. The point here is to baste your fly shut. A short stitch will make it more difficult to rip out later. / , /Close zipper to be sure that it is positioned correctly and evenly and that the tops of your shorts meet. Step 6 | With the zipper closed position the Fly Shield from the inside of the shorts aligning the folded edge with the topstitching of the Fly Facing. Pin in place and flip garment to the right side. Baste along the zipper through all layers. Remove pins, open zipper and stitch along the zipper. Step 7 | Place a bar tack across the bottom of the fly catching both the Fly Facing and Fly Shield. With the zipper open stitch across each side of the zipper tape. Trim the excess zipper tape even with the top of the shorts. Do not cut the zipper tapes without first stitching across the zipper tape! That’s it, that wasn’t so bad, was it./ , /Spread work flat with right side up. 1 Working from the right side, pin zipper in place along topstitching line. 2 Topstitch, removing pins as you go along, leaving long tails of thread at the bottom. 3 On inside, unpin folded-back facing. 4 Remove basting at fly edge and press. Spread work flat with right side up. 1 Working from the right side, pin zipper in place along topstitching line. 2 Topstitch, removing pins as you go along, leaving long tails of thread at the bottom. 3 On inside, unpin folded-back facing. 4 Bring thread ends to inside and knot./ , /Pin into place. 5 Using your sewing machines zipper foot, sew a line of stitching. 6 Turn your work wrong side up and pin the other side of the zipper tape to the β€˜ right β€˜ extension. Machine sew into place. 7 Working on the right of your fashion fabric, we now need to top stitch to get the fly front finish. / , /So you need to lay the fly shield under the zipper as you top stitch. You want the catch the long edge of the shield as you topstitch. And your topstitching should be between the teeth of the zipper and the center front seam, which is where the red line is on this close-up. Now it should look like this. The fabric to the right of the zipper is the fly shield. Turn the zipper onto the left side of the fly extension, right side down. / , /5. Spread work flat with right side up. 1 Working from the right side, pin zipper in place along topstitching line. 2 Topstitch, removing pins as you go along, leaving long tails of thread at the bottom. 3 On inside, unpin folded-back facing. 4 Bring thread ends to inside and knot. 5 Remove basting at fly edge and press/ , /Step 6: Pin the zipper in place as shown and then stitch it in place close to the zipper teeth using a zipper foot. Turn the zipper over so the front side of the zipper is revealed. Now stitch the zipper again 1/8” from the fold. This stitch will show when the zipper is opened, so it should be carefully done. Mark the position of the zipper base or end with a pin. Next, mark the final stitching line with chalk or a marking pen, so it is 1 ¼” from the center front seam. Mark the bottom curve as well making sure to avoid the zipper base. Following the marked line, sew the final stitching line./ , /Place the unsewn side of the zipper tape to the right side of the fly and pin. 8. Using a zipper foot, stitch the unsewn side of the zipper to the right seam allowance. Again stitch only to the seam allowance and not through to the front of the pants. 9. Switch back to your regular sewing foot. Fold the fly back toward to the right and baste in place to secure/ , '}, {'role': 'assistant', 'content': 'Pin the zipper in place as shown and then stitch it in place close to the zipper teeth using a zipper foot. Turn the zipper over so the front side of the zipper is revealed. Now stitch the zipper again 1/8” from the fold.'}]}\n" ] } ], "source": [ "print(data[3])" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "W0Y4u8ubS1OH" }, "outputs": [], "source": [ "from datasets import Dataset\n", "\n", "\n", "processed_data = [{\"Messages\": example} for example in data]\n", "hf_dataset = Dataset.from_list(processed_data)\n", "dataset = hf_dataset" ] }, { "cell_type": "markdown", "metadata": { "id": "X5NI47rOIRP2" }, "source": [ "Check to see if it worked:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "bTzhsRNN4q47" }, "outputs": [], "source": [ "dataset = hf_dataset" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "6h9TpoXq2xIO", "outputId": "3865ea4c-776f-44f9-bd85-46675239fb98" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "{'Messages': {'messages': [{'content': 'You are given a problem.\\nThink about the problem and provide your working out.\\nPlace it between and .\\nThen, provide your solution between ',\n", " 'role': 'system'},\n", " {'content': \"what is cumin good for Here are your sources:/Cumin is the dried seed of the herb Cuminum cyminum, a member of the parsley family. The cumin plant grows to 30–50 cm (12–20 in) tall and is harvested by hand. It is an annual herbaceous plant, with a slender, glabrous, branched stem which is 20–30 cm (8–12 in) tall and has a diameter of 3–5 cm (1 1⁄4 –2 in). Cumin seed is used as a spice for its distinctive flavour and aroma. It is globally popular and an essential flavouring in many cuisines, particularly South Asian (where it is called jeera), Northern African, and Latin American cuisines./ , /Digestion: Cumin is extremely good for digestion and related problems. The very aroma of cumin, which comes from an aromatic organic compound called Cuminaldehyde, the main component of its essential oil, activates our salivary glands in our mouth, which facilitates the primary digestion of food. Cumin also suppresses the development of coughing in the respiratory system since it dries up the excess mucus. Cumin is rich in iron and has considerable amount of vitamin-C, which are essential for a healthy immune system and keeps infections from forming or becoming worse./ , /People take cumin for digestion problems including diarrhea, colic, bowel spasms, and gas. Cumin is also used to increase urine flow to relieve bloating (as a diuretic); to start menstruation; and to increase sexual desire (as an aphrodisiac). / , /Cumin is commonly used in ayurveda, traditional Indian medicine, as a carminative. A carminative is a substance that helps to prevent the formation of gas in the intestines and also helps to alleviate symptoms of gas. Cumin is a spice used in culinary traditions and traditional medicines from around the world. It is a small seed used often in Indian dishes such as curries and has a pungent, nutty and earthy flavor./ , /Cumin seeds are from a different plant all together not from the cilantro plant. Cumin is very strong and bold tasting, and used in chili, hummus, Indian food etc....its good added to just about anything but has a very distinctive taste so you have to be in the mood for it. Cumin is a hot spice that doesn't actually have a spicy flavor. If you want the warmth of spicy food without having a burning hot tongue, this is for you. I use it in chili, tacos, and enchiladas./ , /Cumin seeds are used as a spice for their distinctive aroma, popular in North African, Middle Eastern, Western Chinese, Indian, Cuban and Mexican cuisine. Cumin's distinctive flavour and strong, warm aroma is due to its essential oil content. Cumin is a hot spice that doesn't actually have a spicy flavor. If you want the warmth of spicy food without having a burning hot tongue, this is for you. I use it in chili, tacos, and enchiladas./ , /Mix with salt and a teaspoon of coriander leaf juice. Cumin is also said to help relieve symptoms of the common cold due to it s antiseptic properties. Again, you ll want to boil the seeds in a tea and then drink a couple of times a day. If you also have a sore throat then try adding some dry ginger to help soothe it. The health benefits of cumin for digestive disorders has been well known throughout history. It can help with flatulence, indigestion, diarrhea, nausea, morning sickness, and atonic dyspepsia. In this case, the seeds are boiled in water to make a tea of sorts - 1 teaspoon seeds to 1 glass water/ , /Ground cumin, or Cuminum cyminum, is a spice native to regions of the Mediterranean and Asia. Now used worldwide, cumin adds distinctive savory flavor to soups, entrees and salads. Cumin contains a heart-healthy antioxidant called curcumin, along with other compounds that provide health benefits. Cumin may support heart health, reduce your risk for anemia and help fight infections. / , /Expanding your spice cabinet to include cumin may help to prevent chronic disease and fight off infection. Cumin is a spice used in culinary traditions and traditional medicines from around the world. It is a small seed used often in Indian dishes such as curries and has a pungent, nutty and earthy flavor. It is rich in antioxidants, calcium, iron and magnesium. In addition to preventing disease, incorporating more cumin into your diet may reduce blood sugar and improve diges/ , \",\n", " 'role': 'user'},\n", " {'content': 'Digestion and related problems.', 'role': 'assistant'}]}}" ] }, "metadata": {}, "execution_count": 59 } ], "source": [ "dataset[6586]" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "LTdXBKcslhRH", "outputId": "f8362c9c-b4d3-4a27-f4f3-7fcbe4dea4f3" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "2454" ] }, "metadata": {}, "execution_count": 60 } ], "source": [ "import torch\n", "import gc\n", "\n", "# Clear cache before starting\n", "torch.cuda.empty_cache()\n", "gc.collect()\n", "\n", "# Then start training\n" ] }, { "cell_type": "markdown", "metadata": { "id": "iHV9BXYiIYaq" }, "source": [ "Let's truncate the pre fine-tuning dataset to `max_seq_length/2` since we don't want too long reasoning traces.\n", "\n", "Note this might take 2 minutes!" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 176 }, "id": "eXqy2KCruM-5", "outputId": "5569c05e-3322-44b8-9bfd-8e0e259ac85e" }, "outputs": [ { "ename": "AttributeError", "evalue": "'Dataset' object has no attribute 'copy'", "output_type": "error", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mAttributeError\u001b[0m Traceback (most recent call last)", "\u001b[0;32m/tmp/ipython-input-3988866434.py\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;31m# If you still have the tokenized dataset before filtering:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mdataset_original\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# Save a copy before filtering\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 3\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;31mAttributeError\u001b[0m: 'Dataset' object has no attribute 'copy'" ] } ], "source": [ "# If you still have the tokenized dataset before filtering:\n", "dataset_original = dataset.copy() # Save a copy before filtering\n", "\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "ol4qkVnIEx1y" }, "outputs": [], "source": [ "max_seq_length = 4096" ] }, { "cell_type": "markdown", "metadata": { "id": "E6NkUCAGIj8N" }, "source": [ "We then tokenize the messages and convert it to a Hugging Face compatible dataset format:" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "x-ZQxA_vtkGY", "outputId": "c694dd62-8235-41e0-ee92-6a1a50ee7553" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "10505" ] }, "metadata": {}, "execution_count": 62 } ], "source": [ "len(dataset)" ] }, { "cell_type": "markdown", "metadata": { "id": "bAQJjQrYKzOk" }, "source": [ "Let's now pre fine-tune the model so it follows our custom GRPO formatting!" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "zo4WDynR6aA-", "outputId": "24f9dac5-afbc-469d-c10f-686b43a9b3f6" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "[\"system: You are given a problem.\\nThink about the problem and provide your working out.\\nPlace it between and .\\nThen, provide your solution between \\n\\nuser: dancing with the stars does how many seasons a year Here are your sources:/Derek Hough (/ˈhʌf/ ; born May 17, 1985) is an American professional Latin and ballroom dancer, choreographer, actor, singer, and songwriter. He is known for his work, since September 2007, on the ABC dance-competition series Dancing with the Stars, on which he has won a record five seasons. Menounos became the second celebrity in Dancing with the Stars history to be eliminated on a night where she came in first place with the judges (the first was Willa Ford in season three). For season 15 (All Stars), Hough was paired with season eight champion, Shawn Johnson./ , /Allyssa Lee. Meryl and Maks beat out Amy and Derek for Season 18's coveted Mirrorball trophy. After 10 weeks, 11 eliminations, and too many sequins to count, the judges’ scores and viewer votes have been tallied, and we have a β€œDancing With the Stars” winner! Season 18’s champs are … Meryl Davis and Maksim Chmerkovskiy! And props to fellow finalists Amy Purdy and Derek Hough and Candace Cameron Bure and Mark Ballas for fighting the good fight, but this was the right choice/ , /For the current season, see Dancing with the Stars (U.S. season 21) . Dancing with the Stars is an American dance competition show airing since 2005 on ABC in the United States, and CTV / CTV Two in Canada. The show is the American version of the British television series Strictly Come Dancing. For the first twenty seasons, Len Goodman served as head judge. Julianne Hough, who was a professional dancing partner (in seasons 4–8), was added as a permanent fourth judge on the panel in season 19, after having been a guest judge in the previous two seasons./ , /The main judges are Carrie Ann Inaba and Bruno Tonioli. For the first twenty seasons, Len Goodman served as head judge. Julianne Hough, who was a professional dancing partner (in seasons 4–8), was added as a permanent fourth judge on the panel in season 19, after having been a guest judge in the previous two seasons./ , /For season eleven, Hough won for the third time with actress Jennifer Grey, known for her film roles on Dirty Dancing and Ferris Bueller's Day Off, on November 23, 2010. Hough did not participate in season 12 in order to be a part of the film COBU 3D (later renamed to Make your Move 3D). Menounos became the second celebrity in Dancing with the Stars history to be eliminated on a night where she came in first place with the judges (the first was Willa Ford in season three). For season 15 (All Stars), Hough was paired with season eight champion, Shawn Johnson./ , /Senior Writer, BuddyTV. Add as BuddyTV Friend Β». I hope you have your dancing shoes on, because Dancing with the Stars is back for its 19th season. This year has a few new twists, but still the same old feel as celebrities and professionals team up to Jive, Cha-Cha and Foxtrot into our living rooms. DWTS season 19 brings in Julianne Hough as the new permanent fourth judge. Whether you like her or not, it's definitely better than the endless parade of unqualified buffoons from last year (I'm looking at you, Redfoo)./ , /Season twenty-one of Dancing with the Stars premiered on September 14, 2015. / , /Best Answer: Yes, DWTS has two season per year. This fall will be season 7. The fall season begins in September and ends in November while the spring season begins in March and ends in May. / , /0 0 0. The professional dancers on Dancing With the Stars give up their lives for the show, working up to 10 weeks of air time, plus about a month of pre-season rehearsals. / , /Dancing With the Stars has been on syndication since 2005. It's first episode was aired on ABC on June 5th. The show has had 19 seasons so far with a season airing on both … the spring and winter TV season. The 19th season began on September 19th, 2014. 1 LED TVs are LCD TVs with an alternative backlighting system. 2 The traditional backlighting system for LCD u…. 3 The Best DWTS Cast You've Ever Seen The one and only Bill Nye the Science Guy will be taking the floor on this season of Dancing With the Stars. He'll be paired with Tyne Stecklein. 4 It was only a matter of…./ , \\n\\nassistant: Two season per year\\n\\n\"]" ] }, "metadata": {}, "execution_count": 63 } ], "source": [ "def format_chat_template(example):\n", " messages = example[\"Messages\"][\"messages\"] # Navigate your nested structure\n", "\n", " formatted_text = \"\"\n", " for message in messages:\n", " role = message[\"role\"]\n", " content = message[\"content\"]\n", " formatted_text += f\"{role}: {content}\\n\\n\"\n", "\n", " return [formatted_text]\n", "\n", "\n", "test_result = format_chat_template(dataset[1])\n", "test_result" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "Zz2BM2nU8Yfz", "outputId": "8e1702af-02e3-478a-aa36-692dd17f41f7" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "[\"system: You are given a problem.\\nThink about the problem and provide your working out.\\nPlace it between and .\\nThen, provide your solution between \\n\\nuser: dancing with the stars does how many seasons a year Here are your sources:/Derek Hough (/ˈhʌf/ ; born May 17, 1985) is an American professional Latin and ballroom dancer, choreographer, actor, singer, and songwriter. He is known for his work, since September 2007, on the ABC dance-competition series Dancing with the Stars, on which he has won a record five seasons. Menounos became the second celebrity in Dancing with the Stars history to be eliminated on a night where she came in first place with the judges (the first was Willa Ford in season three). For season 15 (All Stars), Hough was paired with season eight champion, Shawn Johnson./ , /Allyssa Lee. Meryl and Maks beat out Amy and Derek for Season 18's coveted Mirrorball trophy. After 10 weeks, 11 eliminations, and too many sequins to count, the judges’ scores and viewer votes have been tallied, and we have a β€œDancing With the Stars” winner! Season 18’s champs are … Meryl Davis and Maksim Chmerkovskiy! And props to fellow finalists Amy Purdy and Derek Hough and Candace Cameron Bure and Mark Ballas for fighting the good fight, but this was the right choice/ , /For the current season, see Dancing with the Stars (U.S. season 21) . Dancing with the Stars is an American dance competition show airing since 2005 on ABC in the United States, and CTV / CTV Two in Canada. The show is the American version of the British television series Strictly Come Dancing. For the first twenty seasons, Len Goodman served as head judge. Julianne Hough, who was a professional dancing partner (in seasons 4–8), was added as a permanent fourth judge on the panel in season 19, after having been a guest judge in the previous two seasons./ , /The main judges are Carrie Ann Inaba and Bruno Tonioli. For the first twenty seasons, Len Goodman served as head judge. Julianne Hough, who was a professional dancing partner (in seasons 4–8), was added as a permanent fourth judge on the panel in season 19, after having been a guest judge in the previous two seasons./ , /For season eleven, Hough won for the third time with actress Jennifer Grey, known for her film roles on Dirty Dancing and Ferris Bueller's Day Off, on November 23, 2010. Hough did not participate in season 12 in order to be a part of the film COBU 3D (later renamed to Make your Move 3D). Menounos became the second celebrity in Dancing with the Stars history to be eliminated on a night where she came in first place with the judges (the first was Willa Ford in season three). For season 15 (All Stars), Hough was paired with season eight champion, Shawn Johnson./ , /Senior Writer, BuddyTV. Add as BuddyTV Friend Β». I hope you have your dancing shoes on, because Dancing with the Stars is back for its 19th season. This year has a few new twists, but still the same old feel as celebrities and professionals team up to Jive, Cha-Cha and Foxtrot into our living rooms. DWTS season 19 brings in Julianne Hough as the new permanent fourth judge. Whether you like her or not, it's definitely better than the endless parade of unqualified buffoons from last year (I'm looking at you, Redfoo)./ , /Season twenty-one of Dancing with the Stars premiered on September 14, 2015. / , /Best Answer: Yes, DWTS has two season per year. This fall will be season 7. The fall season begins in September and ends in November while the spring season begins in March and ends in May. / , /0 0 0. The professional dancers on Dancing With the Stars give up their lives for the show, working up to 10 weeks of air time, plus about a month of pre-season rehearsals. / , /Dancing With the Stars has been on syndication since 2005. It's first episode was aired on ABC on June 5th. The show has had 19 seasons so far with a season airing on both … the spring and winter TV season. The 19th season began on September 19th, 2014. 1 LED TVs are LCD TVs with an alternative backlighting system. 2 The traditional backlighting system for LCD u…. 3 The Best DWTS Cast You've Ever Seen The one and only Bill Nye the Science Guy will be taking the floor on this season of Dancing With the Stars. He'll be paired with Tyne Stecklein. 4 It was only a matter of…./ , \\n\\nassistant: Two season per year\\n\\n\"]" ] }, "metadata": {}, "execution_count": 64 } ], "source": [ "format_chat_template(dataset[1])" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "mc_Ytv7x718_", "outputId": "337486db-9aeb-4833-ce89-5726eb9921de" }, "outputs": [ { "output_type": "stream", "name": "stdout", "text": [ "Converting dataset...\n", "Processed 0/10505 items\n", "Processed 1000/10505 items\n", "Processed 2000/10505 items\n", "Processed 3000/10505 items\n", "Processed 4000/10505 items\n", "Processed 5000/10505 items\n", "Processed 6000/10505 items\n", "Processed 7000/10505 items\n", "Processed 8000/10505 items\n", "Processed 9000/10505 items\n", "Processed 10000/10505 items\n", "Successfully converted 10505 items\n" ] } ], "source": [ "print(\"Converting dataset...\")\n", "text_dataset = []\n", "\n", "for i, item in enumerate(dataset):\n", " try:\n", " # Handle your specific structure\n", " if isinstance(item[\"Messages\"], list) and len(item[\"Messages\"]) > 0:\n", " messages = item[\"Messages\"][0][\"messages\"]\n", " elif isinstance(item[\"Messages\"], dict):\n", " messages = item[\"Messages\"][\"messages\"]\n", " else:\n", " print(f\"Skipping item {i}: unknown structure\")\n", " continue\n", "\n", " formatted_text = \"\"\n", " for message in messages:\n", " role = message[\"role\"]\n", " content = message[\"content\"]\n", " formatted_text += f\"{role}: {content}\\n\\n\"\n", "\n", " text_dataset.append({\"text\": formatted_text})\n", "\n", " if i % 1000 == 0:\n", " print(f\"Processed {i}/{len(dataset)} items\")\n", "\n", " except Exception as e:\n", " print(f\"Error processing item {i}: {e}\")\n", " continue\n", "\n", "print(f\"Successfully converted {len(text_dataset)} items\")\n", "dataset = Dataset.from_list(text_dataset)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "B3ZQqaT9x1sy", "outputId": "55bd9d6d-f0c4-4aa3-962e-1dadbfce9124" }, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "{'text': \"system: You are given a problem.\\nThink about the problem and provide your working out.\\nPlace it between and .\\nThen, provide your solution between \\n\\nuser: dancing with the stars does how many seasons a year Here are your sources:/Derek Hough (/ˈhʌf/ ; born May 17, 1985) is an American professional Latin and ballroom dancer, choreographer, actor, singer, and songwriter. He is known for his work, since September 2007, on the ABC dance-competition series Dancing with the Stars, on which he has won a record five seasons. Menounos became the second celebrity in Dancing with the Stars history to be eliminated on a night where she came in first place with the judges (the first was Willa Ford in season three). For season 15 (All Stars), Hough was paired with season eight champion, Shawn Johnson./ , /Allyssa Lee. Meryl and Maks beat out Amy and Derek for Season 18's coveted Mirrorball trophy. After 10 weeks, 11 eliminations, and too many sequins to count, the judges’ scores and viewer votes have been tallied, and we have a β€œDancing With the Stars” winner! Season 18’s champs are … Meryl Davis and Maksim Chmerkovskiy! And props to fellow finalists Amy Purdy and Derek Hough and Candace Cameron Bure and Mark Ballas for fighting the good fight, but this was the right choice/ , /For the current season, see Dancing with the Stars (U.S. season 21) . Dancing with the Stars is an American dance competition show airing since 2005 on ABC in the United States, and CTV / CTV Two in Canada. The show is the American version of the British television series Strictly Come Dancing. For the first twenty seasons, Len Goodman served as head judge. Julianne Hough, who was a professional dancing partner (in seasons 4–8), was added as a permanent fourth judge on the panel in season 19, after having been a guest judge in the previous two seasons./ , /The main judges are Carrie Ann Inaba and Bruno Tonioli. For the first twenty seasons, Len Goodman served as head judge. Julianne Hough, who was a professional dancing partner (in seasons 4–8), was added as a permanent fourth judge on the panel in season 19, after having been a guest judge in the previous two seasons./ , /For season eleven, Hough won for the third time with actress Jennifer Grey, known for her film roles on Dirty Dancing and Ferris Bueller's Day Off, on November 23, 2010. Hough did not participate in season 12 in order to be a part of the film COBU 3D (later renamed to Make your Move 3D). Menounos became the second celebrity in Dancing with the Stars history to be eliminated on a night where she came in first place with the judges (the first was Willa Ford in season three). For season 15 (All Stars), Hough was paired with season eight champion, Shawn Johnson./ , /Senior Writer, BuddyTV. Add as BuddyTV Friend Β». I hope you have your dancing shoes on, because Dancing with the Stars is back for its 19th season. This year has a few new twists, but still the same old feel as celebrities and professionals team up to Jive, Cha-Cha and Foxtrot into our living rooms. DWTS season 19 brings in Julianne Hough as the new permanent fourth judge. Whether you like her or not, it's definitely better than the endless parade of unqualified buffoons from last year (I'm looking at you, Redfoo)./ , /Season twenty-one of Dancing with the Stars premiered on September 14, 2015. / , /Best Answer: Yes, DWTS has two season per year. This fall will be season 7. The fall season begins in September and ends in November while the spring season begins in March and ends in May. / , /0 0 0. The professional dancers on Dancing With the Stars give up their lives for the show, working up to 10 weeks of air time, plus about a month of pre-season rehearsals. / , /Dancing With the Stars has been on syndication since 2005. It's first episode was aired on ABC on June 5th. The show has had 19 seasons so far with a season airing on both … the spring and winter TV season. The 19th season began on September 19th, 2014. 1 LED TVs are LCD TVs with an alternative backlighting system. 2 The traditional backlighting system for LCD u…. 3 The Best DWTS Cast You've Ever Seen The one and only Bill Nye the Science Guy will be taking the floor on this season of Dancing With the Stars. He'll be paired with Tyne Stecklein. 4 It was only a matter of…./ , \\n\\nassistant: Two season per year\\n\\n\"}" ] }, "metadata": {}, "execution_count": 66 } ], "source": [ "dataset[1]" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "oJhNWWdW-YCy" }, "outputs": [], "source": [ "import torch\n", "\n", "model = torch.compile(model, disable=True)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 228 }, "id": "QubUTWD_qtwF", "outputId": "0072d5cf-61b5-46d4-d2e0-db505b348be1" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "=== TRAINING STARTING ===\n" ] }, { "ename": "NameError", "evalue": "name 'trainer' is not defined", "output_type": "error", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", "\u001b[0;32m/tmp/ipython-input-3937035576.py\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;31m# Test if logging works\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"=== TRAINING STARTING ===\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"Logging steps: {trainer.args.logging_steps}\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 4\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"Log level: {trainer.args.log_level}\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"Report to: {trainer.args.report_to}\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;31mNameError\u001b[0m: name 'trainer' is not defined" ] } ], "source": [ "# Test if logging works\n", "print(\"=== TRAINING STARTING ===\")\n", "print(f\"Logging steps: {trainer.args.logging_steps}\")\n", "print(f\"Log level: {trainer.args.log_level}\")\n", "print(f\"Report to: {trainer.args.report_to}\")\n", "\n", "# Force a manual log to test\n", "trainer.log({\"test\": \"logging_works\"})" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 209, "referenced_widgets": [ "3126c2b7cdbb4576a345191f10f85cca", "fb611bc668df4b0aae64036cf5683a73", "2e4b1cef89d245b2838e0b09cf6ec597", "6b69ed70e39b488c8d17537c0733481d", "c7ef3d39776d4cda9fbd031ac52fa5cc", "24174c6a668d4e84bf365bbb08912a9a", "93288f198db2429b81311b218b19d016", "e18f56429f8b44d3be0d2d7d383999b3", "50b91752695e420fbb75d3d10f12fb47", "d2bf4a726a4143a1a8f7c4c28dc9903d", "ca0a0edf786c4778bdc8d35f7a98f492", "6d01a0f99735483aa038ff8514816bc9", "5b918ef5bbcc499bbbb67fe237631e06", "f25f30cbbba241d596f2bbb4e597c70d", "4fc2f75f4ab9419896c4f1f0a260070a", "f1680d9e416e4d91b0c5789effe418ed", "63821ba3c7184046892558a60d3835a5", "88c7860ce9824cb79274de474e75a7c1", "2e140ebfa4eb448da7fcb2c44fa9fa72", "21ce2424a87f4925af372885e800bb38", "ef40befe783c4c999af84baae12c0e1e", "ad8c90092e18420e888a2e8c0585e391" ] }, "id": "LHE3RW1bXn2T", "outputId": "38eae432-93aa-429a-b02b-93efa2175edb" }, "outputs": [ { "output_type": "stream", "name": "stderr", "text": [ "PyTorch: setting up devices\n", "Both warmup_ratio and warmup_steps given, warmup_steps will override any effect of warmup_ratio during training\n" ] }, { "output_type": "display_data", "data": { "text/plain": [ "Unsloth: Tokenizing [\"text\"] (num_proc=16): 0%| | 0/8404 [00:00" ], "text/html": [ "\n", "
\n", " \n", " \n", " [2101/2101 7:27:00, Epoch 1/1]\n", "
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
StepTraining LossValidation Loss
1000.5961000.687546
2000.5156000.673229
3000.8674000.659915
4000.6237000.657596
5000.6748000.644143
6000.5242000.637036
7000.5891000.637502
8000.4699000.630543
9000.6208000.629754
10000.5078000.626087
11000.4759000.624740
12000.6035000.620238
13000.6347000.619928
14000.6651000.615476
15000.4536000.616709
16000.5801000.614273
17000.6512000.613702
18000.4892000.614094
19000.3743000.611262
20000.6913000.608146
21000.5415000.608047

" ] }, "metadata": {} }, { "output_type": "stream", "name": "stderr", "text": [ "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-100\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-200\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-300\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-400\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-500\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-600\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-700\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-800\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-900\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-1000\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-1100\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-1200\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-1300\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-1400\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-1500\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-1600\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-1700\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-1800\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-1900\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-2000\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "The following columns in the Evaluation set don't have a corresponding argument in `PeftModelForCausalLM.forward` and have been ignored: text, attention_mask. If text, attention_mask are not expected by `PeftModelForCausalLM.forward`, you can safely ignore this message.\n", "\n", "***** Running Evaluation *****\n", " Num examples = 800\n", " Batch size = 1\n", "Saving model checkpoint to ./checkpoints/checkpoint-2100\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "/usr/local/lib/python3.12/dist-packages/bitsandbytes/autograd/_functions.py:181: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization\n", " warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\n", "Saving model checkpoint to ./checkpoints/checkpoint-2101\n", "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--FluegelQueen--Coeur-Validation-Continued/snapshots/4374b291fdddffe11bff305001665a8b21aea0b1/config.json\n", "/usr/local/lib/python3.12/dist-packages/transformers/models/cohere2/configuration_cohere2.py:235: FutureWarning: The `sliding_window_pattern` attribute is deprecated and will be removed in v4.55.0.\n", " warnings.warn(\n", "Model config Cohere2Config {\n", " \"_sliding_window_pattern\": 4,\n", " \"architectures\": [\n", " \"Cohere2ForCausalLM\"\n", " ],\n", " \"attention_bias\": false,\n", " \"attention_dropout\": 0.0,\n", " \"bos_token_id\": 5,\n", " \"cache_implementation\": \"hybrid\",\n", " \"dtype\": \"bfloat16\",\n", " \"eos_token_id\": 255001,\n", " \"head_dim\": 128,\n", " \"hidden_act\": \"silu\",\n", " \"hidden_size\": 4096,\n", " \"initializer_range\": 0.02,\n", " \"intermediate_size\": 14336,\n", " \"layer_norm_eps\": 1e-05,\n", " \"layer_switch\": 4,\n", " \"layer_types\": [\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"sliding_attention\",\n", " \"full_attention\"\n", " ],\n", " \"logit_scale\": 0.25,\n", " \"max_position_embeddings\": 132096,\n", " \"model_type\": \"cohere2\",\n", " \"num_attention_heads\": 32,\n", " \"num_hidden_layers\": 32,\n", " \"num_key_value_heads\": 8,\n", " \"order_of_interleaved_layers\": \"local_attn_first\",\n", " \"pad_token_id\": 0,\n", " \"position_embedding_type\": \"rope_gptj\",\n", " \"quantization_config\": {\n", " \"_load_in_4bit\": false,\n", " \"_load_in_8bit\": true,\n", " \"bnb_4bit_compute_dtype\": \"float32\",\n", " \"bnb_4bit_quant_storage\": \"uint8\",\n", " \"bnb_4bit_quant_type\": \"fp4\",\n", " \"bnb_4bit_use_double_quant\": false,\n", " \"llm_int8_enable_fp32_cpu_offload\": false,\n", " \"llm_int8_has_fp16_weight\": false,\n", " \"llm_int8_skip_modules\": null,\n", " \"llm_int8_threshold\": 6.0,\n", " \"load_in_4bit\": false,\n", " \"load_in_8bit\": true,\n", " \"quant_method\": \"bitsandbytes\"\n", " },\n", " \"rope_scaling\": null,\n", " \"rope_theta\": 50000,\n", " \"rotary_pct\": 1.0,\n", " \"sliding_window\": 4096,\n", " \"torch_dtype\": \"bfloat16\",\n", " \"transformers_version\": \"4.55.4\",\n", " \"unsloth_version\": \"2025.9.7\",\n", " \"use_cache\": true,\n", " \"use_embedding_sharing\": true,\n", " \"use_gated_activation\": true,\n", " \"use_parallel_block\": true,\n", " \"use_parallel_embedding\": true,\n", " \"vocab_size\": 256000\n", "}\n", "\n", "\n", "\n", "Training completed. Do not forget to share your model on huggingface.co/models =)\n", "\n", "\n", "Loading best model from ./checkpoints/checkpoint-2100 (score: 0.6080468893051147).\n" ] }, { "output_type": "execute_result", "data": { "text/plain": [ "TrainOutput(global_step=2101, training_loss=0.6234348515338526, metrics={'train_runtime': 26829.0863, 'train_samples_per_second': 0.313, 'train_steps_per_second': 0.078, 'total_flos': 2.677715673695355e+17, 'train_loss': 0.6234348515338526, 'epoch': 1.0})" ] }, "metadata": {}, "execution_count": 69 } ], "source": [ "trainer.train()" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "id": "KOor5b95LfUF" }, "outputs": [], "source": [ "from google.colab import drive\n", "drive.mount('/content/drive')" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 249, "referenced_widgets": [ "f04003e529df48098999d861f7b60cc1", "0470a73fadaa4b5bbffaba218071349a", "caeba9613310491cb89c3a2ba52676c4", "fc4783b25c3644a7aa85e0e22c4b6495", "a7db6263a7e548e1a20da1218e0079da", "b9e717d3f727430c81f52de060a2be90", "4df5f6b2845f4c3f8e4daae473d9d7ca", "e658413ca7f744ba9223b26ad0e5e908", "32509c96053743e98c1b481c74a71fa3", "42efb29fe04d4db2af12bb78d9e24142", "7b9795b3561f40ac8a818a3fc6bc3268", "17c92ae263164e78a11aba313fdf840a", "35e2a490c1214d6f8eeb018959df1493", "b30cd96120554daebc8f9c6b9ae458d3", "a7749ad39902464e82c07a5c0713b8f1", "31da910c187c4e91bbe2e288bd2c2776", "8760a10d096e4c349470718f937bc2cd", "ee27c87555464456a3ed22df5b59cce3", "623153e8d6ab4720a110b72302bff412", "f28eb8ada26b4246abe4f1da9532e200", "79440c6c682a4ffba605e7b207ff5234", "5e1bac4f06d3434cad7a7af8080477b7", "3cd57b8464354723a8039ad2b2bba6be", "32fdcea99cf44df7bc5803dc74ae91b9", "0861e5ca54f7484bac430b57f760e5a2", "f96cb0aba0f24e11a9905df099125836", "17b59976f4b94c0b91c198bdfd2516c6", "d573d695d9c0495b85cb0107be280bf6", "4401542bd6fb4fa7b57ecbc564214b4c", "cca83618b4ae41f990fd036a6eb720a0", "ff2468c428584558a544a0781369c841", "259ca568181b4312ad8d302a464d35f4", "3b4f893619e542cfa32fe9e3c332a4f6", "1af9517c6b35474ebe8a8df9564205cb", "78bc092d93394b91a00fa06bb82ffae8", "b5fb0cee58ba4a5fbc30350e45f5c0de", "d9a2d585b7ba4cf2843289e68b90ac73", "7b9c173871fc4242b01cc4833aa6719b", "4e43271070b8435793225bffd5980af3", "c507648b9f55458bac48a13248439fa3", "d5c155bb612e45f4b7bb6004a0318305", "2a38e497e21b4e949e2a397e9ff09b40", "882f8b244a7c44b3b3e0b1f71e04a822", "cbd3904bd98e431dbcd1ca9d5bc572ab", "7d5ef7ea439a42bf8463ae6872d3aa87", "66a4282941b94d0eb778389eaa96bdee", "d216e829fcc74887aa8478d98f5d52ae", "66470961274042a69f1141db7b1fcdad", "3aff6351e9a74774836d6d475de7b4ba", "4cbaff6e81c24b5296c7a939330d6e4a", "bacf42cdeae243e19a9f43db86f66ed9", "0e94c4a9cc16430193a317b0ce18b947", "7828f3719f8e47d1ac51a88d95d28f8a", "fd89063104054b818e144379bc1f019d", "bc7854d3d2bb4d2d9e51fb2c224fa370" ] }, "id": "BxrVsY7_fJX1", "outputId": "e0cf8d01-f0a4-4e14-8438-5541a74837e7" }, "outputs": [ { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "f04003e529df48098999d861f7b60cc1", "version_major": 2, "version_minor": 0 }, "text/plain": [ "README.md: 0%| | 0.00/599 [00:00