{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "cell-title",
   "metadata": {
    "id": "cell-title"
   },
   "source": [
    "# Base Model vs SFT Model — Comparison\n",
    "\n",
    "This notebook quantitatively benchmarks how **Supervised Fine-Tuning (SFT)** on AWS-CLI traces improves an off-the-shelf code model. Two models go head-to-head:\n",
    "\n",
    "- **Base** — `unsloth/Qwen2.5-Coder-3B-Instruct-bnb-4bit` (no fine-tuning, 4-bit)\n",
    "- **SFT** — same base + LoRA adapter (`Sizzing/aws-rl-sft-qwen25coder3b-adapter`) trained on the AWS RL dataset\n",
    "\n",
    "Two evaluation modes — run one or both:\n",
    "\n",
    "| Mode | What it tests | Needs |\n",
    "|---|---|---|\n",
    "| **Dataset eval** | Static pattern matching on held-out prompts | HF token + dataset access |\n",
    "| **RL env eval** | Live multi-turn task completion against the AWS environment | Dataset eval above + running HF Space |\n",
    "\n",
    "**RL env metrics (per episode)**\n",
    "\n",
    "| Metric | What it measures |\n",
    "|---|---|\n",
    "| `avg_episode_reward` | Mean total reward accumulated per episode |\n",
    "| `completion_rate` | % episodes the model completed the task before hitting max steps |\n",
    "| `avg_steps` | Mean number of AWS commands issued per episode |\n",
    "| `avg_reward_per_step` | Mean per-step reward (efficiency) |\n",
    "| `reward_std` | Reward variance across episodes (consistency) |\n",
    "\n",
    "**Before running:**\n",
    "1. Runtime → Change runtime type → GPU (T4)\n",
    "2. Fill in the Config cell below\n",
    "3. Add `HF_TOKEN` to Colab Secrets (🔑 left sidebar)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-config",
   "metadata": {},
   "source": [
    "## ⚙️ Configuration\n",
    "\n",
    "This block centralises every knob for the run. Two clusters of settings:\n",
    "\n",
    "- **Identity / endpoints** — which model artefacts, dataset, and RL env to talk to (`BASE_MODEL`, `SFT_ADAPTER_REPO`, `DATASET_REPO`, `REPO_URL`, `ENV_BASE_URL`).\n",
    "- **Eval budget** — how much to evaluate. `EVAL_MAX_PER_COMBO=2` means up to 2 rows per (difficulty, source) combo for the dataset eval. `RL_EPISODES_PER_DIFF=3` × 5 tiers = **15 RL episodes per model**. `MAX_EPISODE_STEPS=15` caps how many AWS commands an agent gets per episode before we call it a timeout.\n",
    "\n",
    "`MAX_SEQ_LENGTH=512` is what keeps everything on a 16 GB T4. Two `assert`s at the bottom catch the most common footgun — leaving `YOUR_USERNAME` placeholders in the URLs.\n",
    "\n",
    "> **Output:** `Config OK` confirms placeholders were replaced and the run can proceed.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "cell-config",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cell-config",
    "outputId": "f5ce28ff-dc54-4750-f34d-5d0eda178323"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Config OK\n"
     ]
    }
   ],
   "source": [
    "# ── CONFIG ─────────────────────────────────────────────────────────────────\n",
    "BASE_MODEL        = \"unsloth/Qwen2.5-Coder-3B-Instruct-bnb-4bit\"\n",
    "SFT_ADAPTER_REPO  = \"Sizzing/aws-rl-sft-qwen25coder3b-adapter\"   # HF Hub or local path\n",
    "DATASET_REPO      = \"Sizzing/aws-rl-sft\"\n",
    "REPO_URL          = \"https://github.com/bangar1/aws-rl-env-fork\"  # your fork\n",
    "ENV_BASE_URL      = \"https://bangar-hf-aws-rl-env.hf.space/\"        # HF Space URL\n",
    "\n",
    "# Dataset eval knobs\n",
    "MAX_SEQ_LENGTH     = 512\n",
    "MAX_NEW_TOKENS     = 120\n",
    "EVAL_MAX_PER_COMBO = 2     # rows per (difficulty, source) combo\n",
    "\n",
    "# RL env eval knobs — difficulty tiers: warmup, beginner, intermediate, advanced, expert\n",
    "RL_EPISODES_PER_DIFF = 3   # episodes per difficulty tier\n",
    "MAX_EPISODE_STEPS    = 15  # max AWS commands per episode\n",
    "TEMPERATURE          = 0.7\n",
    "\n",
    "IS_COLAB = True\n",
    "# ───────────────────────────────────────────────────────────────────────────\n",
    "assert \"YOUR_USERNAME\" not in ENV_BASE_URL, \"Set ENV_BASE_URL to your HF Space URL\"\n",
    "assert \"YOUR_USERNAME\" not in REPO_URL,    \"Set REPO_URL to your fork URL\"\n",
    "print(\"Config OK\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-install",
   "metadata": {},
   "source": [
    "## 📦 Install Dependencies\n",
    "\n",
    "Installs the GPU stack we'll need:\n",
    "\n",
    "- `unsloth` — 4-bit LoRA inference (fast, T4-friendly)\n",
    "- `transformers >=4.50, <5.0` — pinned to a working range for unsloth\n",
    "- `trl <0.12.0`, `peft`, `accelerate`, `datasets` — fine-tuning + dataset utilities\n",
    "- `bitsandbytes` — quantization backend\n",
    "- `httpx`, `websockets`, `nest_asyncio` — async client for the RL env\n",
    "- `matplotlib`, `numpy` — plots\n",
    "\n",
    "The `%%capture` magic suppresses pip noise — only failures will surface here.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "cell-install",
   "metadata": {
    "id": "cell-install"
   },
   "outputs": [],
   "source": [
    "%%capture\n",
    "!pip install -q --upgrade pip\n",
    "!pip install -q unsloth\n",
    "!pip install -q --force-reinstall --no-deps \"transformers>=4.50,<5.0\"\n",
    "!pip install -q --upgrade \"trl<0.12.0\" peft accelerate datasets huggingface_hub bitsandbytes\n",
    "!pip install -q matplotlib numpy httpx websockets nest_asyncio"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-clone",
   "metadata": {},
   "source": [
    "## 🧬 Clone the AWS RL Env Repo\n",
    "\n",
    "Pulls a shallow clone of the RL-env fork into `/content/aws-rl-env`, installs it as an editable package (`pip install -e .`), and prepends the path to `sys.path` so later imports like `from client import AwsRlEnv` resolve to this clone.\n",
    "\n",
    "If the directory already exists (e.g. a re-run), it's wiped first to guarantee a clean state.\n",
    "\n",
    "> **Output:** confirms the directory was recreated and the editable install completed. The `Building editable …` lines are normal `pip install -e` chatter.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "cell-clone",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cell-clone",
    "outputId": "c65ab6ce-d038-4367-d94f-81d28b2946ca"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Removed existing directory: /content/aws-rl-env\n",
      "  Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n",
      "  Checking if build backend supports build_editable ... \u001b[?25l\u001b[?25hdone\n",
      "  Getting requirements to build editable ... \u001b[?25l\u001b[?25hdone\n",
      "  Preparing editable metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
      "  Building editable for openenv-aws_rl_env (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
      "Repo ready at /content/aws-rl-env\n"
     ]
    }
   ],
   "source": [
    "import subprocess, sys\n",
    "import shutil\n",
    "\n",
    "REPO_DIR = \"/content/aws-rl-env\"\n",
    "\n",
    "# Remove the directory if it already exists\n",
    "if os.path.exists(REPO_DIR):\n",
    "    shutil.rmtree(REPO_DIR)\n",
    "    print(f\"Removed existing directory: {REPO_DIR}\")\n",
    "\n",
    "subprocess.run([\"git\", \"clone\", \"--depth\", \"1\", REPO_URL, REPO_DIR], check=True)\n",
    "!pip install -q -e /content/aws-rl-env   # ← add this\n",
    "\n",
    "if REPO_DIR not in sys.path:\n",
    "    sys.path.insert(0, REPO_DIR)\n",
    "print(\"Repo ready at\", REPO_DIR)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-gpu",
   "metadata": {},
   "source": [
    "## 🎯 GPU & Runtime Detection\n",
    "\n",
    "Picks the right floating-point format for whatever GPU Colab handed out:\n",
    "\n",
    "- **T4** → `fp16` (no native bfloat16 support)\n",
    "- **Anything newer** (A100, L4, etc.) → `bf16`\n",
    "\n",
    "`nest_asyncio.apply()` lets us drive `async` env-client calls from inside notebook cells (Jupyter already owns the outer event loop). `PYTORCH_ALLOC_CONF=expandable_segments:True` reduces VRAM fragmentation when models load and unload.\n",
    "\n",
    "> **Output:** GPU name, total VRAM (~15.6 GB on T4), and the chosen precision. If you ever see `No GPU` here, change runtime type and re-run.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "cell-gpu",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cell-gpu",
    "outputId": "52260168-fddc-4cde-ed62-d3d0c38178c0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GPU  : Tesla T4\n",
      "VRAM : 15.6 GB\n",
      "Prec : fp16\n"
     ]
    }
   ],
   "source": [
    "import os, gc, time, json, math, asyncio, logging\n",
    "from dataclasses import dataclass, field\n",
    "from typing import List, Tuple\n",
    "import torch\n",
    "import nest_asyncio\n",
    "nest_asyncio.apply()\n",
    "\n",
    "os.environ.setdefault(\"PYTORCH_ALLOC_CONF\", \"expandable_segments:True\")\n",
    "\n",
    "assert torch.cuda.is_available(), \"No GPU — Runtime → Change runtime type → GPU\"\n",
    "gpu     = torch.cuda.get_device_properties(0)\n",
    "IS_T4   = \"T4\" in gpu.name\n",
    "USE_FP16 = IS_T4\n",
    "USE_BF16 = not IS_T4\n",
    "print(f\"GPU  : {gpu.name}\")\n",
    "print(f\"VRAM : {gpu.total_memory / 1e9:.1f} GB\")\n",
    "print(f\"Prec : {'fp16' if USE_FP16 else 'bf16'}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-auth",
   "metadata": {},
   "source": [
    "## 🔐 Hugging Face Authentication\n",
    "\n",
    "Reads `HF_TOKEN` from Colab Secrets, exports it to the env, and authenticates the session so `from_pretrained(...)` calls can pull both the base model and the (potentially gated) SFT adapter.\n",
    "\n",
    "> **Output:** `HF authenticated`. The \"Note: Environment variable `HF_TOKEN` is set …\" line is informational — `huggingface_hub` is just telling you it sees an existing token in the env, not an error.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "cell-auth",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cell-auth",
    "outputId": "a1df87c9-db7b-460c-df5e-7347b83e412b"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Note: Environment variable`HF_TOKEN` is set and is the current active token independently from the token you've just configured.\n",
      "WARNING:huggingface_hub._login:Note: Environment variable`HF_TOKEN` is set and is the current active token independently from the token you've just configured.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HF authenticated\n"
     ]
    }
   ],
   "source": [
    "if IS_COLAB:\n",
    "    from google.colab import userdata\n",
    "    os.environ[\"HF_TOKEN\"] = userdata.get(\"HF_TOKEN\")\n",
    "\n",
    "assert os.environ.get(\"HF_TOKEN\"), \"Set HF_TOKEN in Colab Secrets\"\n",
    "\n",
    "from huggingface_hub import login as hf_login\n",
    "hf_login(token=os.environ[\"HF_TOKEN\"], add_to_git_credential=False)\n",
    "print(\"HF authenticated\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-health",
   "metadata": {},
   "source": [
    "## 🩺 RL Env Health Check\n",
    "\n",
    "Quick HTTP ping to the HF Space hosting the AWS RL environment, mainly to wake the Space if it's been idle.\n",
    "\n",
    "> **Output:** `404` here is **expected** — this Space's root router doesn't expose a literal `/health` endpoint. What matters is that we got *any* response (the Space is awake and answering). The real session is opened later via a websocket inside `AwsRlEnv(...).connect()`.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "cell-health",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cell-health",
    "outputId": "7358686d-101a-4f4e-c7aa-b36869bcc10f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RL env server: 404 {'detail': 'Not Found'}\n"
     ]
    }
   ],
   "source": [
    "import httpx\n",
    "\n",
    "with httpx.Client(timeout=15) as c:\n",
    "    r = c.get(f\"{ENV_BASE_URL}/health\")\n",
    "    print(\"RL env server:\", r.status_code, r.json())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cell-part1-md",
   "metadata": {
    "id": "cell-part1-md"
   },
   "source": [
    "---\n",
    "# Part 1 — Dataset Eval (static)\n",
    "\n",
    "A **fast, deterministic, offline** check. We sample a few held-out prompts from the validation split and ask each model what AWS CLI command it would emit. We never hit the live RL env here — every metric is computed by string-comparing the generated command to the gold-standard one in the dataset.\n",
    "\n",
    "This catches **format drift** (does the model produce parseable single-line commands?) and **service / operation accuracy** (does it pick the right AWS API and verb?).\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-dataset",
   "metadata": {},
   "source": [
    "## 📚 Load the Evaluation Dataset\n",
    "\n",
    "Streams the AWS-RL dataset from the Hub. Three splits:\n",
    "\n",
    "- **train** (1,500 rows) — what the SFT was trained on\n",
    "- **validation** (150 rows) — held out, this is what we score on\n",
    "- **reserve** (200 rows) — kept for downstream RL training\n",
    "\n",
    "Each row contains `task_id`, `difficulty`, `source`, `step_idx`, and a chat-style `messages` list. The last assistant turn is the gold AWS CLI command we'll grade against.\n",
    "\n",
    "> **Output:** the `DatasetDict` printout confirms all three splits loaded with the expected row counts (1500 / 150 / 200).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "cell-dataset",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 491,
     "referenced_widgets": [
      "3cf95de334a14af9b7274dee96b2f20d",
      "41a765ae42f142a5aaf0163612bbe490",
      "0f614193015744f2a0263aa13681b411",
      "68d44bd4ff9a439f958573c8ec16405b",
      "c0c12b0fc46f4b9f8ea2af23e194da23",
      "796833bcdc3845849bf4b44da6379fcc",
      "b22885fcbf3147509191939f5de83602",
      "cc427e51ea1446aaa74d6f24cb17d043",
      "a8816b70cd9f409ba7acfacc2864c043",
      "c71fef24bafe4b45896601a7ae144c2f",
      "548cc14a659b4e6e9c217ff980e34608",
      "9d1f05eb5d3b4bdba57d618f022eaa7b",
      "7ec0191160e04c5397c24bbad3d2b593",
      "bf317a89297c4265938863cc3f74ba8c",
      "616c3d44f0a64e22828c2e472a8c13bc",
      "57cd591f5b3f4d6aa2e401dcb0c3c8fc",
      "51ad078171814b9ca022d88231cf6387",
      "ed06bfd4c13f41598d5e10bbe6139eb8",
      "62ba79bd0e2f47fcbb073cbe15ab2bf6",
      "62a7c1256a3d462ca1f58711b4985852",
      "0cf6a91ad33f4f0a94b890881658b5b2",
      "5178b799d7404fd282f4e5016d616d14",
      "0d2747470d4947849c27775b43ecb54c",
      "fbc88709dcf84cfbb91d7efbb10e0b42",
      "cc4366820b234778b543af95146a4e92",
      "0fba0e0c04c14214ae58cdd3f65fd92f",
      "4dae08a1f54842e49185c4402659df6f",
      "c795b9a1dc834f67ac44a568b9744ed7",
      "1241be723444464fbb334b06d8208410",
      "7f28dc39b40543d9b17bd8ad611e3623",
      "2def96a8b6d646ef8801ce2c1e0a8d14",
      "96abbd4be45f4eef9587a795f10c73f9",
      "6736d9704da04076a134426a47cb15c4",
      "617dcbb6224b49bb9b83dac185420e09",
      "7c93228f55074cf78b1170001545f89d",
      "b1e2bcaf50124b14b7bc70fb403e68b8",
      "c36279d8d0774918812725c906a8af12",
      "ebb9a1dea358429e9e166c470076d4ce",
      "5bbbc6530c014b5296c181526d08cec0",
      "8a2f80c0be5a47b994f6e091cb01d287",
      "fbb8cb04661945c782b1df5e423b37a0",
      "94eb4cb08c8443098267d78b1665514e",
      "f5eb0e8911ad46d19d1d76dfa55fb21c",
      "1161e043c93c4671a8155dce8d9dd701",
      "1dc1444225ce43a888350992cc7f0bbb",
      "9ef9a30c556c4d9da3692fd8a829270f",
      "b14329f3b61447f788d6e836bf934c91",
      "6cff000d81e8431298609a9148cdb793",
      "49de19f8c9e04603980b4a4a8c816545",
      "4dca96c48eb64681b36a76b4f0b01b62",
      "66f989e6acee487faf8f6cc453faa486",
      "052ba95707274c42933cb13fc8ad078c",
      "daf23f97138a44f49f8099d73a39b33b",
      "741c77c00dd34486bd780369d2918d53",
      "b6c49d1dbf184f86a4a94644340b9f90",
      "c9572ada407547acb82b499b4aba9408",
      "2f17b452a3f04d448beff33d60445e2e",
      "5cdf05f2225f499e8574ab64ec9a8766",
      "80d9e1dd1507439088c3d9d8c4c7b0ac",
      "4e6c58ce96404281b880833f88e08413",
      "cd77111d0ee44ecaacb5159f68c88609",
      "89604435feab456888553549728c15a7",
      "66558440f90b42d99278658f7151f043",
      "188a00a386704c7887d139bc284ccc1a",
      "ceadf3915006459fb72ed9141c49dc2f",
      "63b9d68de11b4fbfa5e8acaaa25b4c35",
      "f65560291a834aef846683d3909a9db6",
      "9c1e0ff6a9094793905ddc191e95796d",
      "182361253eee49418048b01fbcf2672f",
      "8af925f00d3541958d318f43591bdd76",
      "c0cdb510c34c4e27b8061a5b4e4b88fa",
      "ea1e771eea51498b9917da282d783427",
      "39929719372644589a9baa7295fd5849",
      "79d5b6596f1b4ee5966b55dad97df3a2",
      "b49e779bf02c49bbb1f4eba5ac2dd1e2",
      "e87a2309352248309f209b71dd050799",
      "1e3372b4af51443f88050b77fedc5bef"
     ]
    },
    "id": "cell-dataset",
    "outputId": "6708e2fd-f2c4-4b6a-baf7-da9410d3cb22"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3cf95de334a14af9b7274dee96b2f20d",
       "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": "9d1f05eb5d3b4bdba57d618f022eaa7b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "data/train-00000-of-00001.parquet:   0%|          | 0.00/1.92M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0d2747470d4947849c27775b43ecb54c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "data/validation-00000-of-00001.parquet:   0%|          | 0.00/194k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "617dcbb6224b49bb9b83dac185420e09",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "data/reserve-00000-of-00001.parquet:   0%|          | 0.00/261k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1dc1444225ce43a888350992cc7f0bbb",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Generating train split:   0%|          | 0/1500 [00:00<?, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c9572ada407547acb82b499b4aba9408",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Generating validation split:   0%|          | 0/150 [00:00<?, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f65560291a834aef846683d3909a9db6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Generating reserve split:   0%|          | 0/200 [00:00<?, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DatasetDict({\n",
      "    train: Dataset({\n",
      "        features: ['task_id', 'difficulty', 'source', 'step_idx', 'messages'],\n",
      "        num_rows: 1500\n",
      "    })\n",
      "    validation: Dataset({\n",
      "        features: ['task_id', 'difficulty', 'source', 'step_idx', 'messages'],\n",
      "        num_rows: 150\n",
      "    })\n",
      "    reserve: Dataset({\n",
      "        features: ['task_id', 'difficulty', 'source', 'step_idx', 'messages'],\n",
      "        num_rows: 200\n",
      "    })\n",
      "})\n"
     ]
    }
   ],
   "source": [
    "from datasets import load_dataset\n",
    "\n",
    "ds = load_dataset(DATASET_REPO, token=os.environ[\"HF_TOKEN\"])\n",
    "print(ds)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-eval-fn",
   "metadata": {},
   "source": [
    "## 🧮 Scoring Helpers\n",
    "\n",
    "These helpers run the eval and break \"correctness\" into five increasingly forgiving metrics so we can see *where* a model fails (formatting? wrong service? wrong flags?):\n",
    "\n",
    "| Metric | Pass condition |\n",
    "|---|---|\n",
    "| `format_pct` | Generation **starts with** `aws ` (no preamble, no markdown fence) |\n",
    "| `format_after_extract_pct` | After stripping fences/prose, *some* line starts with `aws ` |\n",
    "| `exact_pct` | Extracted command matches the gold command character-for-character |\n",
    "| `service_pct` | Same AWS service (e.g. `s3`, `ec2`) as gold |\n",
    "| `operation_pct` | Same service **and** operation (e.g. `s3 ls`) as gold |\n",
    "\n",
    "Plus `avg_latency` and `avg_len` to gauge speed and verbosity.\n",
    "\n",
    "`build_eval_set` picks up to 2 prompts per `(difficulty, source)` combo so the eval is balanced across difficulty tiers — small but representative.\n",
    "\n",
    "> **Output:** `Eval set: 18 prompts across 9 (tier, source) combos` — that's the deterministic eval surface used by both models.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "cell-eval-fn",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cell-eval-fn",
    "outputId": "ac3a5251-d96a-4b52-fc02-8567af3fd4c1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Eval set: 18 prompts across 9 (tier, source) combos\n"
     ]
    }
   ],
   "source": [
    "def extract_command(raw: str) -> str:\n",
    "    text = raw.strip()\n",
    "    if text.startswith(\"```\"):\n",
    "        lines = text.split(\"\\n\")\n",
    "        text = \"\\n\".join(l for l in lines if not l.startswith(\"```\")).strip()\n",
    "    for line in text.split(\"\\n\"):\n",
    "        line = line.strip()\n",
    "        if line.startswith(\"aws \"):\n",
    "            return line\n",
    "    return text\n",
    "\n",
    "\n",
    "def score_row(completion: str, expected: str) -> dict:\n",
    "    extracted  = extract_command(completion)\n",
    "    e_tokens   = extracted.split()\n",
    "    exp_tokens = expected.split()\n",
    "    return {\n",
    "        \"format_ok\":            completion.strip().startswith(\"aws \"),\n",
    "        \"format_after_extract\": extracted.startswith(\"aws \"),\n",
    "        \"exact\":                extracted == expected.strip(),\n",
    "        \"service\":              (len(e_tokens) >= 2 and len(exp_tokens) >= 2\n",
    "                                 and e_tokens[1:2] == exp_tokens[1:2]),\n",
    "        \"operation\":            (len(e_tokens) >= 3 and len(exp_tokens) >= 3\n",
    "                                 and e_tokens[2:3] == exp_tokens[2:3]),\n",
    "    }\n",
    "\n",
    "\n",
    "def build_eval_set(dataset, max_per_combo: int = 2):\n",
    "    seen, picks = {}, []\n",
    "    for r in dataset:\n",
    "        key = (r[\"difficulty\"], r[\"source\"])\n",
    "        seen[key] = seen.get(key, 0) + 1\n",
    "        if seen[key] <= max_per_combo:\n",
    "            picks.append(r)\n",
    "    return picks\n",
    "\n",
    "\n",
    "def dataset_eval(model, tokenizer, eval_set, max_new_tokens: int = 120) -> dict:\n",
    "    results = []\n",
    "    model.eval()\n",
    "    for row in eval_set:\n",
    "        msgs     = row[\"messages\"][:2]\n",
    "        expected = row[\"messages\"][2][\"content\"]\n",
    "        prompt   = tokenizer.apply_chat_template(\n",
    "            msgs, tokenize=False, add_generation_prompt=True\n",
    "        )\n",
    "        inputs = tokenizer(prompt, return_tensors=\"pt\").to(model.device)\n",
    "        t0 = time.time()\n",
    "        with torch.inference_mode():\n",
    "            out_ids = model.generate(\n",
    "                **inputs, max_new_tokens=max_new_tokens,\n",
    "                do_sample=False, temperature=0.0,\n",
    "                pad_token_id=tokenizer.eos_token_id,\n",
    "            )\n",
    "        dt         = time.time() - t0\n",
    "        completion = tokenizer.decode(\n",
    "            out_ids[0, inputs.input_ids.shape[1]:], skip_special_tokens=True\n",
    "        )\n",
    "        s = score_row(completion, expected)\n",
    "        s.update({\"latency\": dt, \"len\": len(completion),\n",
    "                  \"completion\": completion, \"expected\": expected})\n",
    "        results.append(s)\n",
    "\n",
    "    n = len(results)\n",
    "    return {\n",
    "        \"format_pct\":               sum(r[\"format_ok\"]            for r in results) / n,\n",
    "        \"format_after_extract_pct\": sum(r[\"format_after_extract\"] for r in results) / n,\n",
    "        \"exact_pct\":                sum(r[\"exact\"]                for r in results) / n,\n",
    "        \"service_pct\":              sum(r[\"service\"]              for r in results) / n,\n",
    "        \"operation_pct\":            sum(r[\"operation\"]            for r in results) / n,\n",
    "        \"avg_latency\":              sum(r[\"latency\"]              for r in results) / n,\n",
    "        \"avg_len\":                  sum(r[\"len\"]                  for r in results) / n,\n",
    "        \"_per_row\":                 results,\n",
    "    }\n",
    "\n",
    "\n",
    "EVAL_SET = build_eval_set(ds[\"validation\"], max_per_combo=EVAL_MAX_PER_COMBO)\n",
    "combos   = len(set((r[\"difficulty\"], r[\"source\"]) for r in EVAL_SET))\n",
    "print(f\"Eval set: {len(EVAL_SET)} prompts across {combos} (tier, source) combos\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-base-dataset-eval",
   "metadata": {},
   "source": [
    "## 🧪 Base Model — Dataset Eval\n",
    "\n",
    "Loads the un-finetuned `Qwen2.5-Coder-3B-Instruct` in 4-bit, runs the 18-prompt eval at `temperature=0` (greedy, reproducible), then unloads it from VRAM so the SFT model can fit next.\n",
    "\n",
    "> **Output (Base):**\n",
    ">\n",
    "> | Metric | Value | What it means |\n",
    "> |---|---|---|\n",
    "> | `format_pct` | **33.3%** | Only ⅓ of generations start cleanly with `aws ` |\n",
    "> | `format_after_extract_pct` | **100%** | But every output has *some* `aws` line buried inside it |\n",
    "> | `exact_pct` | **38.9%** | Fewer than half match the reference exactly |\n",
    "> | `service_pct` | **77.8%** | Usually picks the right AWS service |\n",
    "> | `operation_pct` | **61.1%** | Right operation about 60% of the time |\n",
    "> | `avg_latency` | 1.90 s | Per-prompt generation time |\n",
    "> | `avg_len` | 85.8 chars | Average response length |\n",
    ">\n",
    "> **Reading:** the base model usually picks the right *service*, but is verbose, often wraps commands in markdown or English prose, and gets the operation/flags wrong about 40% of the time.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "cell-base-dataset-eval",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 686,
     "referenced_widgets": [
      "082ceda90b8d484a81a1de5fead892b6",
      "e8560b1f7c5a42a1b850a24bd22f9a73",
      "5a5d3dac3d2f43419349404bd0c79f4f",
      "4772a353fbb243aca9f472de257d1358",
      "b14fbeb1fb4f4888b63afd9cf4476b28",
      "0ac4cc8e789d4a7389ba2d81f752b2a6",
      "17ad15eed85f493ca0d0d6b8d914ca2c",
      "8c91b59afcd343648321a9ed206531a7",
      "a3710ca3ab474ce59a146b54fea25736",
      "f2247b50901c4a20854c1ac284d76e7f",
      "71c857a576e04c1592352df2553a10f6",
      "44fb962be403416c910b1c4bf25bcdf1",
      "161b757d5f64481a9212dbccd7a884c7",
      "a18521a6e11b4e1ea6e53ce81b2af574",
      "969a9823b8864c0f877eb2c3af383205",
      "753cba0957c240cd8d7fcf171dd2557e",
      "89fdaf865e0e4f7fa1909f032005d1d8",
      "8060b0a254d54032bbb01a56431e98eb",
      "80d066d9f0614ae4b264ade4a304cf49",
      "bdccff3baee846459a9645d014266d2a",
      "f105bcab30d8404a9b3a1ce866480374",
      "9c78bf775e4149cfbe34f1e253c2c9fc",
      "7799ec6190d04928963fbd82d195e2b3",
      "68d465d228c141a7b5c8c97799102790",
      "9f01388850d44b30a155bdd6c1528de4",
      "a5a23a6849e240be90007a200a14e1a3",
      "1a190eca13234bca86da0f33e46a34e1",
      "00ff165127094d2fb481980c130f04c8",
      "d2af9a4571554901a04f1afdd556984d",
      "d06befc9db47458fa696880e19449d90",
      "aadd0d6341714668ae5a85c445f49a09",
      "5329bea02fc2432b84b8bb8eb733eaf9",
      "ba6c2ac768044ec0b688e6573e540f92",
      "e580572f40454f8ebb8d6d50cab3c644",
      "bc4ee14a8b7140edba21f43a0d673045",
      "399a72c355d8478e87995aff1c57f426",
      "ab9aa9b9c93b4ee5900a351cd584cee8",
      "2c51081cf8314bd58a8b7b93cdf2eed7",
      "88f4f5a8f65742eb8cb1228e5d94508c",
      "54c323cb9fe9458fb062b2d4bd5cf927",
      "564a55b31d504a2499140061de7c8354",
      "db821f8b3e1b414590bbde186cc62d59",
      "974d765200034661a14a25e0b241fdeb",
      "41873020c9324e8b96fb6a8a7e432ed2",
      "bab57aecb3f84216af720fa9321f6fb8",
      "2fe800ca1f6948009a218f9f3ce0acbb",
      "9e1d365fa1c84030a4caeead8ef954d3",
      "55a60ff24a23452682aebac4cdcdf086",
      "b71f5888837e4717bc8669603cb587a2",
      "f13c88be577549acad8e2606a8c4e6f2",
      "c1b370ab7fee4dbeb2cb37f47f774eb6",
      "6eb2f500acda438a82e453126b22ae25",
      "9d32149562cf47859390636be9c70d76",
      "923264d53c454d9d96d618b0efd3f59a",
      "8b918bfb9f2f4ee0a71884837b495ebf",
      "3332ae39c1094b3bb3fde04198a3f2a6",
      "71cbf065dfca4450a615fdef7c2114f6",
      "fdadb7a27d07425c8e04dd10df1aeba1",
      "ec590a2a32d045b194dab35ae5273043",
      "826ab03235c94f929ca1e261327a0137",
      "06c2b0aff48f40a886e14723ee788248",
      "0803a7b3995241f493052389046afe98",
      "98dada6fc260427ebe46383263de1e75",
      "5405361924614557867c74ba647b08b4",
      "b3d0061be3d24e22af12225fb08f02ee",
      "9f61db866c9c41e6af53f0b79f55b4e7",
      "ba375ee09b334728b8cfc63975f4302a",
      "4f00645b01294756ae17955d5889514b",
      "358628764f24466198ceadb0f4478487",
      "874de7ba061b4d8783c74e605440394e",
      "01e0602c676442ff952a7a7b3f4024c7",
      "2e707c60d0384e8b8c993c883ad3456a",
      "3af6672587a84a2b886482e68fcfbba9",
      "20e58c5f72e14d47901d1eec0a08c12f",
      "217d16de232c4a1d83a1a4b48ad453d6",
      "2dbe35d37c824c839a2913a0cfbc5266",
      "f9f3f435180b4929a938ac4500216247",
      "31d96fb276614537bab907c316f0ce25",
      "f9ec2b3ae5604344bf3a5f267ce7de38",
      "4772c50a4e454979acb7cc4de466937a",
      "6ee363978e9749d7b39191a4027be66a",
      "b74c48e99fa742ad8b44d5bb7fd34171",
      "c034e1899af543e7ab854b7276a787e2",
      "c545c0cc491b4692b6f7eaaf42e50baf",
      "ef8ace2a258a479db27919c2527a172f",
      "02068d599f844ea5838544754e6d8448",
      "5a09cd99ab3444248c365ba3e1f46436",
      "375ea535f86c4dd193d114568bb3045b"
     ]
    },
    "id": "cell-base-dataset-eval",
    "outputId": "930c1a84-a40f-400b-a419-cb8c67ef404f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "🦥 Unsloth: Will patch your computer to enable 2x faster free finetuning.\n",
      "🦥 Unsloth Zoo will now patch everything to make training faster!\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:unsloth_zoo.log:Unsloth: Could not patch trl.trainer.gkd_trainer: Direct module loading failed for UnslothGKDTrainer: parameter without a default follows parameter with a default (UnslothGKDTrainer.py, line 962)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading base model...\n",
      "==((====))==  Unsloth 2026.4.8: Fast Qwen2 patching. Transformers: 4.57.6.\n",
      "   \\\\   /|    Tesla T4. Num GPUs = 1. Max memory: 14.563 GB. Platform: Linux.\n",
      "O^O/ \\_/ \\    Torch: 2.10.0+cu128. CUDA: 7.5. CUDA Toolkit: 12.8. Triton: 3.6.0\n",
      "\\        /    Bfloat16 = FALSE. FA [Xformers = 0.0.35. FA2 = False]\n",
      " \"-____-\"     Free license: http://github.com/unslothai/unsloth\n",
      "Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "082ceda90b8d484a81a1de5fead892b6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model.safetensors:   0%|          | 0.00/2.05G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "44fb962be403416c910b1c4bf25bcdf1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "generation_config.json:   0%|          | 0.00/266 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7799ec6190d04928963fbd82d195e2b3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer_config.json: 0.00B [00:00, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e580572f40454f8ebb8d6d50cab3c644",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "vocab.json: 0.00B [00:00, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bab57aecb3f84216af720fa9321f6fb8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "merges.txt: 0.00B [00:00, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3332ae39c1094b3bb3fde04198a3f2a6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "added_tokens.json:   0%|          | 0.00/632 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ba375ee09b334728b8cfc63975f4302a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "special_tokens_map.json:   0%|          | 0.00/613 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "31d96fb276614537bab907c316f0ce25",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer.json: 0.00B [00:00, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running base model dataset eval...\n",
      "\n",
      "=== BASE — Dataset Eval ===\n",
      "  format_pct                       33.3%\n",
      "  format_after_extract_pct        100.0%\n",
      "  exact_pct                        38.9%\n",
      "  service_pct                      77.8%\n",
      "  operation_pct                    61.1%\n",
      "  avg_latency                    1.901\n",
      "  avg_len                        85.833\n",
      "\n",
      "Base model unloaded.\n"
     ]
    }
   ],
   "source": [
    "from unsloth import FastLanguageModel\n",
    "\n",
    "print(\"Loading base model...\")\n",
    "base_model, base_tokenizer = FastLanguageModel.from_pretrained(\n",
    "    model_name=BASE_MODEL, max_seq_length=MAX_SEQ_LENGTH,\n",
    "    load_in_4bit=True, dtype=None,\n",
    ")\n",
    "FastLanguageModel.for_inference(base_model)\n",
    "\n",
    "print(\"Running base model dataset eval...\")\n",
    "base_ds_metrics = dataset_eval(base_model, base_tokenizer, EVAL_SET,\n",
    "                               max_new_tokens=MAX_NEW_TOKENS)\n",
    "\n",
    "print(\"\\n=== BASE — Dataset Eval ===\")\n",
    "for k, v in base_ds_metrics.items():\n",
    "    if k == \"_per_row\": continue\n",
    "    print(f\"  {k:<30} {100*v:6.1f}%\" if \"pct\" in k else f\"  {k:<30} {v:.3f}\")\n",
    "\n",
    "del base_model, base_tokenizer\n",
    "gc.collect(); torch.cuda.empty_cache()\n",
    "print(\"\\nBase model unloaded.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-sft-dataset-eval",
   "metadata": {},
   "source": [
    "## 🧪 SFT Model — Dataset Eval\n",
    "\n",
    "Same eval, this time with the LoRA adapter applied to the base.\n",
    "\n",
    "> **Output (SFT):**\n",
    ">\n",
    "> | Metric | Value | Δ vs Base |\n",
    "> |---|---|---|\n",
    "> | `format_pct` | **100%** | **+66.7 pt** |\n",
    "> | `exact_pct` | **88.9%** | **+50.0 pt** |\n",
    "> | `service_pct` | **88.9%** | **+11.1 pt** |\n",
    "> | `operation_pct` | **88.9%** | **+27.8 pt** |\n",
    "> | `avg_latency` | 1.56 s | −0.34 s |\n",
    "> | `avg_len` | 74.7 chars | −11.1 chars |\n",
    ">\n",
    "> **Reading:** SFT teaches the model to **stop chatting and start emitting commands**. Format compliance jumps from 33% → 100%, exact matches more than double, and responses are both shorter *and* faster (less filler ⇒ fewer tokens generated ⇒ lower latency).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "cell-sft-dataset-eval",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 406,
     "referenced_widgets": [
      "68000393fa504271ba1a4c0dfb18a182",
      "ea95e9fdfc1e43ec8c1cb7b35d24f436",
      "5e9776eebdc544159f73ebc7447d54f1",
      "d953a7141b174afdbf2f8912cdf4e04f",
      "2b446f04958943a4adc493cb624ecb7e",
      "2f1ae5179b0945fda20f970c25c93c57",
      "e90dd825fc0b44d2bbe077f7361756af",
      "ee387cb17b58451fab83e2b3241032dc",
      "c402601073394089ab3718d6288660ff",
      "24dadb01295d447ca81ac02a4908fdf1",
      "c07912b3d8974e6891c3d770391aa649"
     ]
    },
    "id": "cell-sft-dataset-eval",
    "outputId": "11069897-2321-47c6-cf66-ffda050561e9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading SFT model...\n",
      "==((====))==  Unsloth 2026.4.8: Fast Qwen2 patching. Transformers: 4.57.6.\n",
      "   \\\\   /|    Tesla T4. Num GPUs = 1. Max memory: 14.563 GB. Platform: Linux.\n",
      "O^O/ \\_/ \\    Torch: 2.10.0+cu128. CUDA: 7.5. CUDA Toolkit: 12.8. Triton: 3.6.0\n",
      "\\        /    Bfloat16 = FALSE. FA [Xformers = 0.0.35. FA2 = False]\n",
      " \"-____-\"     Free license: http://github.com/unslothai/unsloth\n",
      "Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "68000393fa504271ba1a4c0dfb18a182",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "adapter_model.safetensors:   0%|          | 0.00/14.8M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Unsloth 2026.4.8 patched 36 layers with 0 QKV layers, 0 O layers and 0 MLP layers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running SFT model dataset eval...\n",
      "\n",
      "=== SFT — Dataset Eval ===\n",
      "  format_pct                      100.0%\n",
      "  format_after_extract_pct        100.0%\n",
      "  exact_pct                        88.9%\n",
      "  service_pct                      88.9%\n",
      "  operation_pct                    88.9%\n",
      "  avg_latency                    1.559\n",
      "  avg_len                        74.722\n",
      "\n",
      "SFT model unloaded.\n"
     ]
    }
   ],
   "source": [
    "print(\"Loading SFT model...\")\n",
    "sft_model, sft_tokenizer = FastLanguageModel.from_pretrained(\n",
    "    model_name=SFT_ADAPTER_REPO, max_seq_length=MAX_SEQ_LENGTH,\n",
    "    load_in_4bit=True, dtype=None,\n",
    ")\n",
    "FastLanguageModel.for_inference(sft_model)\n",
    "\n",
    "print(\"Running SFT model dataset eval...\")\n",
    "sft_ds_metrics = dataset_eval(sft_model, sft_tokenizer, EVAL_SET,\n",
    "                              max_new_tokens=MAX_NEW_TOKENS)\n",
    "\n",
    "print(\"\\n=== SFT — Dataset Eval ===\")\n",
    "for k, v in sft_ds_metrics.items():\n",
    "    if k == \"_per_row\": continue\n",
    "    print(f\"  {k:<30} {100*v:6.1f}%\" if \"pct\" in k else f\"  {k:<30} {v:.3f}\")\n",
    "\n",
    "del sft_model, sft_tokenizer\n",
    "gc.collect(); torch.cuda.empty_cache()\n",
    "print(\"\\nSFT model unloaded.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cell-part2-md",
   "metadata": {
    "id": "cell-part2-md"
   },
   "source": [
    "---\n",
    "# Part 2 — RL Env Eval (live rollouts)\n",
    "\n",
    "The dataset eval above is necessary but not sufficient — a model can ace single-shot pattern-matching and still fall apart when it has to **plan over multiple turns** and condition on real CLI output.\n",
    "\n",
    "In this section each model plays **N episodes per difficulty tier** (`warmup → beginner → intermediate → advanced → expert`) against the live AWS RL environment. At each step the model sees the task description plus the recent (command, output) history and must emit one CLI command. The env returns observation + reward + `done`. Per-episode metrics:\n",
    "\n",
    "- **Episode reward** — sum of per-step rewards over the episode (env-defined)\n",
    "- **Completion rate** — episodes that hit `done=True` before `MAX_EPISODE_STEPS=15`\n",
    "- **Avg steps & reward/step** — efficiency proxies\n",
    "- **Reward std** — consistency across runs\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-rl-helpers",
   "metadata": {},
   "source": [
    "## 🔧 RL Eval Helpers\n",
    "\n",
    "Defines the rollout machinery:\n",
    "\n",
    "- `SYSTEM_PROMPT` — pins the agent to single-command, no-prose output. This is the same instruction style the SFT data was formatted in.\n",
    "- `build_prompt(...)` — assembles a chat-template prompt with system + task + the **last 4 (command, output) pairs** (each output truncated to 400 chars). Truncating history is what keeps the prompt under 512 tokens — usually.\n",
    "- `_extract_aws_command(...)` — defensively pulls the first `aws ...` line out of whatever the model generated; falls back to `aws help` if none is found (so the env still gets a valid action).\n",
    "- `_generate(...)` — sampled decoding (`temperature=0.7`, `top_p=0.95`), not greedy — RL needs some exploration.\n",
    "- `EpisodeResult` / `run_episode(...)` — drive a single episode end-to-end against `AwsRlEnv`, capturing per-step rewards, total reward, step count, and completion.\n",
    "- `rl_eval(...)` — outer loop that walks all 5 difficulty tiers via `load_tier(...)`, runs `episodes_per_diff` episodes per tier, and aggregates per-tier and global metrics.\n",
    "\n",
    "> **Output:** `RL eval helpers ready.`\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "cell-rl-helpers",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cell-rl-helpers",
    "outputId": "59d6d5da-7d53-4e7d-b031-407e89927efe"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RL eval helpers ready.\n"
     ]
    }
   ],
   "source": [
    "from client import AwsRlEnv\n",
    "from models import AwsRlAction, Task, TaskDifficulty\n",
    "from server.services.curriculum import load_tier\n",
    "\n",
    "logging.basicConfig(level=logging.WARNING)\n",
    "\n",
    "SYSTEM_PROMPT = (\n",
    "    \"You are an expert AWS SRE agent. Emit ONE AWS CLI command per turn starting \"\n",
    "    \"with 'aws '. No explanation, no markdown, no quotes — just the command.\"\n",
    ")\n",
    "\n",
    "# All tiers in progression order\n",
    "ALL_DIFFICULTIES = [\n",
    "    TaskDifficulty.WARMUP,\n",
    "    TaskDifficulty.BEGINNER,\n",
    "    TaskDifficulty.INTERMEDIATE,\n",
    "    TaskDifficulty.ADVANCED,\n",
    "    TaskDifficulty.EXPERT,\n",
    "]\n",
    "\n",
    "\n",
    "def build_prompt(task: Task, history: List[Tuple[str, str]], tokenizer) -> str:\n",
    "    messages = [\n",
    "        {\"role\": \"system\", \"content\": SYSTEM_PROMPT},\n",
    "        {\"role\": \"user\",   \"content\": f\"TASK: {task.description}\"},\n",
    "    ]\n",
    "    for cmd, out in history[-4:]:\n",
    "        messages.append({\"role\": \"assistant\", \"content\": cmd})\n",
    "        messages.append({\"role\": \"user\",      \"content\": f\"OUTPUT:\\n{out[:400]}\"})\n",
    "    return tokenizer.apply_chat_template(\n",
    "        messages, tokenize=False, add_generation_prompt=True\n",
    "    )\n",
    "\n",
    "\n",
    "def _extract_aws_command(raw: str) -> str:\n",
    "    for line in raw.splitlines():\n",
    "        line = line.strip().strip(\"`\").strip()\n",
    "        if line.startswith(\"aws \"):\n",
    "            return line\n",
    "    return \"aws help\"\n",
    "\n",
    "\n",
    "@torch.no_grad()\n",
    "def _generate(model, tokenizer, prompt: str) -> str:\n",
    "    device = next(model.parameters()).device\n",
    "    inputs = tokenizer(prompt, return_tensors=\"pt\").to(device)\n",
    "    out = model.generate(\n",
    "        **inputs,\n",
    "        max_new_tokens=MAX_NEW_TOKENS,\n",
    "        do_sample=True,\n",
    "        temperature=TEMPERATURE,\n",
    "        top_p=0.95,\n",
    "        pad_token_id=tokenizer.eos_token_id,\n",
    "    )\n",
    "    return tokenizer.decode(\n",
    "        out[0, inputs.input_ids.shape[1]:], skip_special_tokens=True\n",
    "    )\n",
    "\n",
    "\n",
    "@dataclass\n",
    "class EpisodeResult:\n",
    "    total_reward: float = 0.0\n",
    "    steps: int = 0\n",
    "    completed: bool = False\n",
    "    per_step_rewards: List[float] = field(default_factory=list)\n",
    "    difficulty: str = \"unknown\"\n",
    "\n",
    "\n",
    "async def run_episode(model, tokenizer, task: Task) -> EpisodeResult:\n",
    "    result = EpisodeResult(difficulty=task.difficulty.value)\n",
    "    env = AwsRlEnv(base_url=ENV_BASE_URL)\n",
    "    await env.connect()\n",
    "    try:\n",
    "        await env.reset(task=task)\n",
    "        history = []\n",
    "        for _ in range(MAX_EPISODE_STEPS):\n",
    "            prompt  = build_prompt(task, history, tokenizer)\n",
    "            text    = await asyncio.to_thread(_generate, model, tokenizer, prompt)\n",
    "            command = _extract_aws_command(text)\n",
    "            step    = await env.step(AwsRlAction(command=command))\n",
    "            r       = float(step.reward)\n",
    "            result.total_reward += r\n",
    "            result.per_step_rewards.append(r)\n",
    "            result.steps += 1\n",
    "            history.append((command, step.observation.command_output or \"\"))\n",
    "            if step.done:\n",
    "                result.completed = True\n",
    "                break\n",
    "    finally:\n",
    "        await env.close()\n",
    "    return result\n",
    "\n",
    "\n",
    "async def rl_eval(model, tokenizer, episodes_per_diff: int = 3) -> dict:\n",
    "    \"\"\"Run episodes across all 5 difficulty tiers and return aggregate metrics.\"\"\"\n",
    "    all_results: List[EpisodeResult] = []\n",
    "\n",
    "    for diff_enum in ALL_DIFFICULTIES:\n",
    "        diff_tasks = load_tier(diff_enum)\n",
    "        if not diff_tasks:\n",
    "            print(f\"  No tasks found for {diff_enum.value}, skipping.\")\n",
    "            continue\n",
    "        diff_label = diff_enum.value\n",
    "        for i in range(episodes_per_diff):\n",
    "            task = diff_tasks[i % len(diff_tasks)]\n",
    "            ep   = await run_episode(model, tokenizer, task)\n",
    "            all_results.append(ep)\n",
    "            print(f\"  [{diff_label:12}] ep {i+1}/{episodes_per_diff}  \"\n",
    "                  f\"reward={ep.total_reward:6.2f}  \"\n",
    "                  f\"steps={ep.steps:2d}  \"\n",
    "                  f\"done={'✓' if ep.completed else '✗'}\")\n",
    "\n",
    "    n          = len(all_results)\n",
    "    rewards    = [r.total_reward for r in all_results]\n",
    "    steps_list = [r.steps        for r in all_results]\n",
    "    flat_sr    = [s for r in all_results for s in r.per_step_rewards]\n",
    "\n",
    "    per_diff = {}\n",
    "    for diff_enum in ALL_DIFFICULTIES:\n",
    "        diff_label = diff_enum.value\n",
    "        sub = [r for r in all_results if r.difficulty == diff_label]\n",
    "        if sub:\n",
    "            per_diff[diff_label] = {\n",
    "                \"avg_reward\":      sum(r.total_reward for r in sub) / len(sub),\n",
    "                \"completion_rate\": sum(r.completed    for r in sub) / len(sub),\n",
    "                \"avg_steps\":       sum(r.steps        for r in sub) / len(sub),\n",
    "            }\n",
    "\n",
    "    mean_r = sum(rewards) / n\n",
    "    return {\n",
    "        \"avg_episode_reward\":  mean_r,\n",
    "        \"reward_std\":          math.sqrt(sum((r - mean_r)**2 for r in rewards) / n),\n",
    "        \"completion_rate\":     sum(r.completed for r in all_results) / n,\n",
    "        \"avg_steps\":           sum(steps_list) / n,\n",
    "        \"avg_reward_per_step\": sum(flat_sr) / len(flat_sr) if flat_sr else 0.0,\n",
    "        \"_rewards\":            rewards,\n",
    "        \"_results\":            all_results,\n",
    "        \"_per_diff\":           per_diff,\n",
    "    }\n",
    "\n",
    "print(\"RL eval helpers ready.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-openenv",
   "metadata": {},
   "source": [
    "## 🔍 Sanity Check — Where is `openenv` Loaded From?\n",
    "\n",
    "A guardrail to confirm the env client is being imported from the version we just cloned (and `pip install -e`'d), not a stale system-wide install. Stale imports are a notorious source of \"I changed the code but nothing's different\" debugging time-sinks.\n",
    "\n",
    "> **Output:** in this run, `openenv` is loaded from the global `/usr/local/lib/python3.12/dist-packages/openenv/` rather than the clone. That's fine here — the public API surface we use (`AwsRlEnv`, `AwsRlAction`, `Task`) is stable — but worth knowing if you start patching env-side behaviour locally and don't see your changes take effect.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "77e7f47e",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "77e7f47e",
    "outputId": "2d52b07c-f695-4249-c692-86a522041919"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "'openenv' module found in sys.modules. Loaded from: /usr/local/lib/python3.12/dist-packages/openenv/__init__.py\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "import os\n",
    "\n",
    "if 'openenv' in sys.modules:\n",
    "    print(f\"'openenv' module found in sys.modules. Loaded from: {sys.modules['openenv'].__file__}\")\n",
    "    # Optionally, verify it's from the REPO_DIR\n",
    "    if os.path.commonpath([sys.modules['openenv'].__file__, REPO_DIR]) == REPO_DIR:\n",
    "        print(\"  (Path confirms it's loaded from the cloned repository)\")\n",
    "else:\n",
    "    print(\"'openenv' module not found in sys.modules.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-base-rl-eval",
   "metadata": {},
   "source": [
    "## 🎮 Base Model — RL Env Rollouts\n",
    "\n",
    "15 episodes total (3 per tier × 5 tiers). Each episode is up to 15 steps; the model must keep emitting AWS commands until the env signals `done=True` or we hit the step cap.\n",
    "\n",
    "> **Output (Base):**\n",
    ">\n",
    "> - **Warmup & Beginner** — solves all 6 episodes in **1 step each** (single-command tasks, easy).\n",
    "> - **Intermediate** — only 1/3 completed; the 15-step misses indicate the model gets stuck looping on wrong commands.\n",
    "> - **Advanced & Expert** — 0/6 completed; reward trickles in (the env hands out partial credit) but episodes always time out.\n",
    "> - **Aggregate:** `avg_reward = 1.187`, `completion_rate = 46.7%`, `avg_steps = 8.6`, `reward_std = 1.137`, `avg_reward_per_step = 0.138`.\n",
    ">\n",
    "> The repeated `Input IDs of length 5xx > max sequence length 512` warnings are Unsloth flagging that, as the (command, output) history accumulates, prompts overflow the context window and have to be left-truncated. This is one mechanical reason the base model degrades on longer tasks — it literally can't see the full conversation any more.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "cell-base-rl-eval",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cell-base-rl-eval",
    "outputId": "25070cec-3f5b-4e1c-b2ea-215f71f5e40d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading base model for RL eval...\n",
      "==((====))==  Unsloth 2026.4.8: Fast Qwen2 patching. Transformers: 4.57.6.\n",
      "   \\\\   /|    Tesla T4. Num GPUs = 1. Max memory: 14.563 GB. Platform: Linux.\n",
      "O^O/ \\_/ \\    Torch: 2.10.0+cu128. CUDA: 7.5. CUDA Toolkit: 12.8. Triton: 3.6.0\n",
      "\\        /    Bfloat16 = FALSE. FA [Xformers = 0.0.35. FA2 = False]\n",
      " \"-____-\"     Free license: http://github.com/unslothai/unsloth\n",
      "Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!\n",
      "Running 3 episodes per difficulty tier...\n",
      "  [warmup      ] ep 1/3  reward=  1.00  steps= 1  done=✓\n",
      "  [warmup      ] ep 2/3  reward=  1.00  steps= 1  done=✓\n",
      "  [warmup      ] ep 3/3  reward=  1.00  steps= 1  done=✓\n",
      "  [beginner    ] ep 1/3  reward=  1.00  steps= 1  done=✓\n",
      "  [beginner    ] ep 2/3  reward=  1.00  steps= 1  done=✓\n",
      "  [beginner    ] ep 3/3  reward=  1.00  steps= 1  done=✓\n",
      "  [intermediate] ep 1/3  reward=  0.00  steps=15  done=✗\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Unsloth: Input IDs of shape torch.Size([1, 523]) with length 523 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 515]) with length 515 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 633]) with length 633 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  [intermediate] ep 2/3  reward=  2.10  steps=15  done=✗\n",
      "  [intermediate] ep 3/3  reward=  2.00  steps= 3  done=✓\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Unsloth: Input IDs of shape torch.Size([1, 570]) with length 570 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 637]) with length 637 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  [advanced    ] ep 1/3  reward=  0.00  steps=15  done=✗\n",
      "  [advanced    ] ep 2/3  reward=  0.56  steps=15  done=✗\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Unsloth: Input IDs of shape torch.Size([1, 527]) with length 527 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 526]) with length 526 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 525]) with length 525 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 567]) with length 567 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  [advanced    ] ep 3/3  reward=  1.83  steps=15  done=✗\n",
      "  [expert      ] ep 1/3  reward=  0.00  steps=15  done=✗\n",
      "  [expert      ] ep 2/3  reward=  4.70  steps=15  done=✗\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Unsloth: Input IDs of shape torch.Size([1, 548]) with length 548 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 623]) with length 623 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  [expert      ] ep 3/3  reward=  0.62  steps=15  done=✗\n",
      "\n",
      "=== BASE — RL Env Eval ===\n",
      "  avg_episode_reward        1.187\n",
      "  reward_std                1.137\n",
      "  completion_rate           46.7%\n",
      "  avg_steps                 8.600\n",
      "  avg_reward_per_step       0.138\n",
      "\n",
      "Base model unloaded.\n"
     ]
    }
   ],
   "source": [
    "print(\"Loading base model for RL eval...\")\n",
    "base_model, base_tokenizer = FastLanguageModel.from_pretrained(\n",
    "    model_name=BASE_MODEL, max_seq_length=MAX_SEQ_LENGTH,\n",
    "    load_in_4bit=True, dtype=None,\n",
    ")\n",
    "FastLanguageModel.for_inference(base_model)\n",
    "\n",
    "print(f\"Running {RL_EPISODES_PER_DIFF} episodes per difficulty tier...\")\n",
    "base_rl_metrics = asyncio.run(\n",
    "    rl_eval(base_model, base_tokenizer, episodes_per_diff=RL_EPISODES_PER_DIFF)\n",
    ")\n",
    "\n",
    "print(\"\\n=== BASE — RL Env Eval ===\")\n",
    "for k, v in base_rl_metrics.items():\n",
    "    if k.startswith(\"_\"): continue\n",
    "    print(f\"  {'completion_rate':<25} {100*v:.1f}%\" if k == \"completion_rate\"\n",
    "          else f\"  {k:<25} {v:.3f}\")\n",
    "\n",
    "del base_model, base_tokenizer\n",
    "gc.collect(); torch.cuda.empty_cache()\n",
    "print(\"\\nBase model unloaded.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-sft-rl-eval",
   "metadata": {},
   "source": [
    "## 🎮 SFT Model — RL Env Rollouts\n",
    "\n",
    "Same 15 episodes against the same tiers, this time with the SFT-tuned adapter.\n",
    "\n",
    "> **Output (SFT):**\n",
    ">\n",
    "> - **Warmup** — 2/3 (one episode hit the step cap; sampling at T=0.7 occasionally produces a bad first command on tiny tasks).\n",
    "> - **Beginner & Intermediate** — **6/6 completed**, all in ≤3 steps each.\n",
    "> - **Advanced** — 1/3 completed, but a missed episode racked up `reward = 4.93` — the model is making real progress, just not closing.\n",
    "> - **Expert** — **2/3 completed** (vs 0/3 for base!).\n",
    "> - **Aggregate:** `avg_reward = 2.011`, `completion_rate = 73.3%`, `avg_steps = 5.733`, `avg_reward_per_step = 0.351`, `reward_std = 1.908`.\n",
    ">\n",
    "> **Compared to base:** **+69%** episode reward, **+27 pp** completion rate, **−33%** steps to solve, **+154%** reward per step. SFT reaches harder tiers and takes fewer turns to get there.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "cell-sft-rl-eval",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cell-sft-rl-eval",
    "outputId": "0959f5c1-6682-4ea0-9bd6-18277dd436ca"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading SFT model for RL eval...\n",
      "==((====))==  Unsloth 2026.4.8: Fast Qwen2 patching. Transformers: 4.57.6.\n",
      "   \\\\   /|    Tesla T4. Num GPUs = 1. Max memory: 14.563 GB. Platform: Linux.\n",
      "O^O/ \\_/ \\    Torch: 2.10.0+cu128. CUDA: 7.5. CUDA Toolkit: 12.8. Triton: 3.6.0\n",
      "\\        /    Bfloat16 = FALSE. FA [Xformers = 0.0.35. FA2 = False]\n",
      " \"-____-\"     Free license: http://github.com/unslothai/unsloth\n",
      "Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!\n",
      "Running 3 episodes per difficulty tier...\n",
      "  [warmup      ] ep 1/3  reward=  0.00  steps=15  done=✗\n",
      "  [warmup      ] ep 2/3  reward=  1.00  steps= 1  done=✓\n",
      "  [warmup      ] ep 3/3  reward=  1.00  steps= 1  done=✓\n",
      "  [beginner    ] ep 1/3  reward=  1.00  steps= 1  done=✓\n",
      "  [beginner    ] ep 2/3  reward=  1.00  steps= 1  done=✓\n",
      "  [beginner    ] ep 3/3  reward=  1.00  steps= 1  done=✓\n",
      "  [intermediate] ep 1/3  reward=  1.50  steps= 2  done=✓\n",
      "  [intermediate] ep 2/3  reward=  1.50  steps= 2  done=✓\n",
      "  [intermediate] ep 3/3  reward=  2.00  steps= 3  done=✓\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Unsloth: Input IDs of shape torch.Size([1, 581]) with length 581 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 622]) with length 622 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 645]) with length 645 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  [advanced    ] ep 1/3  reward=  0.00  steps=15  done=✗\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Unsloth: Input IDs of shape torch.Size([1, 673]) with length 673 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 808]) with length 808 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 815]) with length 815 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 741]) with length 741 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 638]) with length 638 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 517]) with length 517 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  [advanced    ] ep 2/3  reward=  4.93  steps=15  done=✗\n",
      "  [advanced    ] ep 3/3  reward=  3.58  steps= 6  done=✓\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Unsloth: Input IDs of shape torch.Size([1, 558]) with length 558 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 693]) with length 693 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 774]) with length 774 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 593]) with length 593 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 574]) with length 574 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n",
      "Unsloth: Input IDs of shape torch.Size([1, 655]) with length 655 > the model's max sequence length of 512.\n",
      "We shall truncate it ourselves. It's imperative if you correct this issue first.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  [expert      ] ep 1/3  reward=  7.48  steps=15  done=✗\n",
      "  [expert      ] ep 2/3  reward=  2.10  steps= 5  done=✓\n",
      "  [expert      ] ep 3/3  reward=  2.08  steps= 3  done=✓\n",
      "\n",
      "=== SFT — RL Env Eval ===\n",
      "  avg_episode_reward        2.011\n",
      "  reward_std                1.908\n",
      "  completion_rate           73.3%\n",
      "  avg_steps                 5.733\n",
      "  avg_reward_per_step       0.351\n",
      "\n",
      "SFT model unloaded.\n"
     ]
    }
   ],
   "source": [
    "print(\"Loading SFT model for RL eval...\")\n",
    "sft_model, sft_tokenizer = FastLanguageModel.from_pretrained(\n",
    "    model_name=SFT_ADAPTER_REPO, max_seq_length=MAX_SEQ_LENGTH,\n",
    "    load_in_4bit=True, dtype=None,\n",
    ")\n",
    "FastLanguageModel.for_inference(sft_model)\n",
    "\n",
    "print(f\"Running {RL_EPISODES_PER_DIFF} episodes per difficulty tier...\")\n",
    "sft_rl_metrics = asyncio.run(\n",
    "    rl_eval(sft_model, sft_tokenizer, episodes_per_diff=RL_EPISODES_PER_DIFF)\n",
    ")\n",
    "\n",
    "print(\"\\n=== SFT — RL Env Eval ===\")\n",
    "for k, v in sft_rl_metrics.items():\n",
    "    if k.startswith(\"_\"): continue\n",
    "    print(f\"  {'completion_rate':<25} {100*v:.1f}%\" if k == \"completion_rate\"\n",
    "          else f\"  {k:<25} {v:.3f}\")\n",
    "\n",
    "del sft_model, sft_tokenizer\n",
    "gc.collect(); torch.cuda.empty_cache()\n",
    "print(\"\\nSFT model unloaded.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cell-plots-md",
   "metadata": {
    "id": "cell-plots-md"
   },
   "source": [
    "---\n",
    "# 📊 Plots — Dataset Eval + RL Env Eval\n",
    "\n",
    "Two figures summarise the numbers above. **Figure 1** is dataset-eval (static prompt scoring); **Figure 2** is RL-env eval (live multi-turn rollouts). Plot styling is normalised in the first cell so both figures share fonts, colours, and grid behaviour.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-ds-plots",
   "metadata": {},
   "source": [
    "## Figure 1 — Dataset Eval (2×2)\n",
    "\n",
    "Four panels packaged as `compare_dataset.png`:\n",
    "\n",
    "1. **Top-left — Grouped bars:** Base vs SFT side-by-side across the 5 accuracy metrics. Easiest at-a-glance read of where SFT moved the needle most.\n",
    "2. **Top-right — Horizontal latency / length bars:** values normalised to 0–100 so they share a single x-axis; raw values annotated at the bar tips.\n",
    "3. **Bottom-left — Delta bars (SFT − Base, in pp):** green = improvement, red = regression. In this run every category is green.\n",
    "4. **Bottom-right — Radar chart:** \"capability profile\" — bigger filled area = stronger model. Visualises SFT dominance as one combined shape.\n",
    "\n",
    "> **Output:** the figure is rendered inline and saved with `Saved → compare_dataset.png`.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "cell-plots",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "cell-plots",
    "outputId": "31f3c249-0bfc-44db-b465-fd0184378156"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x1680 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved → compare_dataset.png\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.patches as mpatches\n",
    "from matplotlib.gridspec import GridSpec\n",
    "\n",
    "COLOR_BASE = \"#4C72B0\"\n",
    "COLOR_SFT  = \"#DD8452\"\n",
    "COLOR_POS  = \"#55A868\"\n",
    "COLOR_NEG  = \"#C44E52\"\n",
    "\n",
    "plt.rcParams.update({\n",
    "    \"figure.dpi\":        120,\n",
    "    \"font.family\":       \"DejaVu Sans\",\n",
    "    \"font.size\":         11,\n",
    "    \"axes.titlesize\":    12,\n",
    "    \"axes.titleweight\":  \"bold\",\n",
    "    \"axes.labelsize\":    10,\n",
    "    \"axes.spines.top\":   False,\n",
    "    \"axes.spines.right\": False,\n",
    "    \"axes.grid\":         True,\n",
    "    \"grid.alpha\":        0.35,\n",
    "    \"legend.framealpha\": 0.9,\n",
    "    \"legend.fontsize\":   9,\n",
    "    \"xtick.labelsize\":   9,\n",
    "    \"ytick.labelsize\":   9,\n",
    "})\n",
    "\n",
    "\n",
    "# ── helper: annotate bars ───────────────────────────────────────────────────\n",
    "def annotate_bars(ax, bars, fmt=\"{:.1f}\", offset=0.5, color=None, fontsize=8):\n",
    "    for bar in bars:\n",
    "        h = bar.get_height()\n",
    "        ax.text(\n",
    "            bar.get_x() + bar.get_width() / 2, h + offset,\n",
    "            fmt.format(h), ha=\"center\", va=\"bottom\",\n",
    "            fontsize=fontsize, fontweight=\"bold\",\n",
    "            color=color or bar.get_facecolor(),\n",
    "        )\n",
    "\n",
    "\n",
    "# ═══════════════════════════════════════════════════════════════════════════\n",
    "# Figure 1 — Dataset eval (2x2)\n",
    "# ═══════════════════════════════════════════════════════════════════════════\n",
    "acc_keys   = [\"format_pct\", \"format_after_extract_pct\", \"exact_pct\",\n",
    "              \"service_pct\", \"operation_pct\"]\n",
    "acc_labels = [\"Format\", \"Format\\n(extracted)\", \"Exact\", \"Service\", \"Operation\"]\n",
    "base_acc   = [base_ds_metrics[k] * 100 for k in acc_keys]\n",
    "sft_acc    = [sft_ds_metrics[k]  * 100 for k in acc_keys]\n",
    "delta_acc  = [s - b for s, b in zip(sft_acc, base_acc)]\n",
    "\n",
    "lat_labels = [\"Avg Latency (s)\", \"Avg Resp Len\"]\n",
    "base_lat   = [base_ds_metrics[\"avg_latency\"], base_ds_metrics[\"avg_len\"]]\n",
    "sft_lat    = [sft_ds_metrics[\"avg_latency\"],  sft_ds_metrics[\"avg_len\"]]\n",
    "\n",
    "fig1 = plt.figure(figsize=(18, 14))\n",
    "gs1  = GridSpec(2, 2, figure=fig1, hspace=0.48, wspace=0.38)\n",
    "ax1, ax2 = fig1.add_subplot(gs1[0, 0]), fig1.add_subplot(gs1[0, 1])\n",
    "ax3      = fig1.add_subplot(gs1[1, 0])\n",
    "ax4      = fig1.add_subplot(gs1[1, 1], polar=True)\n",
    "\n",
    "# 1a. Grouped bar — accuracy\n",
    "x, w = np.arange(len(acc_labels)), 0.35\n",
    "annotate_bars(ax1, ax1.bar(x - w/2, base_acc, w, color=COLOR_BASE,\n",
    "              label=\"Base\", edgecolor=\"white\", linewidth=0.6), fmt=\"{:.1f}%\")\n",
    "annotate_bars(ax1, ax1.bar(x + w/2, sft_acc,  w, color=COLOR_SFT,\n",
    "              label=\"SFT\",  edgecolor=\"white\", linewidth=0.6), fmt=\"{:.1f}%\")\n",
    "ax1.set(title=\"Dataset Accuracy — Base vs SFT\", ylabel=\"Score (%)\",\n",
    "        xlabel=\"Metric\", ylim=(0, 118))\n",
    "ax1.set_xticks(x); ax1.set_xticklabels(acc_labels)\n",
    "ax1.legend(); ax1.set_axisbelow(True)\n",
    "\n",
    "# 1b. Horizontal bar — latency / length (dual x-axes not needed; normalize)\n",
    "y, h = np.arange(len(lat_labels)), 0.35\n",
    "# Normalize each metric independently so bars fit on one scale\n",
    "max_vals = [max(base_lat[i], sft_lat[i]) for i in range(len(lat_labels))]\n",
    "base_norm = [base_lat[i] / max_vals[i] * 100 for i in range(len(lat_labels))]\n",
    "sft_norm  = [sft_lat[i]  / max_vals[i] * 100 for i in range(len(lat_labels))]\n",
    "hb = ax2.barh(y + h/2, base_norm, h, color=COLOR_BASE, label=\"Base\",\n",
    "              edgecolor=\"white\", linewidth=0.6)\n",
    "hs = ax2.barh(y - h/2, sft_norm,  h, color=COLOR_SFT,  label=\"SFT\",\n",
    "              edgecolor=\"white\", linewidth=0.6)\n",
    "for bar, raw in zip(list(hb) + list(hs),\n",
    "                    [base_lat[0], base_lat[1], sft_lat[0], sft_lat[1]]):\n",
    "    ax2.text(bar.get_width() + 1, bar.get_y() + bar.get_height()/2,\n",
    "             f\"{raw:.2f}\", va=\"center\", fontsize=9, fontweight=\"bold\")\n",
    "ax2.set(title=\"Latency & Response Length\", xlabel=\"% of max (raw value annotated)\")\n",
    "ax2.set_yticks(y); ax2.set_yticklabels(lat_labels)\n",
    "ax2.set_xlim(0, 130); ax2.legend(loc=\"lower right\"); ax2.set_axisbelow(True)\n",
    "\n",
    "# 1c. Delta bar\n",
    "colors_d = [COLOR_POS if d >= 0 else COLOR_NEG for d in delta_acc]\n",
    "bars_d   = ax3.bar(x, delta_acc, 0.5, color=colors_d, edgecolor=\"white\", linewidth=0.6)\n",
    "for bar, d in zip(bars_d, delta_acc):\n",
    "    ax3.text(bar.get_x() + bar.get_width()/2,\n",
    "             bar.get_height() + (0.4 if d >= 0 else -1.2),\n",
    "             f\"{d:+.1f}pt\", ha=\"center\", va=\"bottom\", fontsize=9, fontweight=\"bold\",\n",
    "             color=COLOR_POS if d >= 0 else COLOR_NEG)\n",
    "ax3.axhline(0, color=\"#333\", lw=0.9)\n",
    "ax3.set(title=\"Delta: SFT − Base (dataset, pp)\", ylabel=\"Δ pp\", xlabel=\"Metric\")\n",
    "ax3.set_xticks(x); ax3.set_xticklabels(acc_labels)\n",
    "ax3.legend(handles=[mpatches.Patch(color=COLOR_POS, label=\"Improvement\"),\n",
    "                    mpatches.Patch(color=COLOR_NEG, label=\"Regression\")])\n",
    "ax3.set_axisbelow(True)\n",
    "\n",
    "# 1d. Radar\n",
    "N      = len(acc_labels)\n",
    "angles = np.linspace(0, 2*np.pi, N, endpoint=False).tolist() + [0]\n",
    "ax4.set_theta_offset(np.pi / 2); ax4.set_theta_direction(-1)\n",
    "ax4.set_thetagrids(np.degrees(angles[:-1]), acc_labels, fontsize=8)\n",
    "for vals, color, label in [(base_acc + base_acc[:1], COLOR_BASE, \"Base\"),\n",
    "                            (sft_acc  + sft_acc[:1],  COLOR_SFT,  \"SFT\")]:\n",
    "    ax4.plot(angles, vals, color=color, linewidth=2, label=label)\n",
    "    ax4.fill(angles, vals, color=color, alpha=0.15)\n",
    "ax4.set_ylim(0, 100)\n",
    "ax4.set_yticks([20, 40, 60, 80, 100])\n",
    "ax4.set_yticklabels([\"20%\", \"40%\", \"60%\", \"80%\", \"100%\"], fontsize=7)\n",
    "ax4.set_title(\"Capability Profile (Dataset)\", pad=20, fontweight=\"bold\")\n",
    "ax4.legend(loc=\"upper right\", bbox_to_anchor=(1.35, 1.15))\n",
    "\n",
    "fig1.suptitle(\"Part 1 — Dataset Eval: Base vs SFT\",\n",
    "              fontsize=15, fontweight=\"bold\", y=1.01)\n",
    "plt.savefig(\"compare_dataset.png\", dpi=150, bbox_inches=\"tight\", facecolor=\"white\")\n",
    "plt.show()\n",
    "print(\"Saved → compare_dataset.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-rl-plots",
   "metadata": {},
   "source": [
    "## Figure 2 — RL Env Eval (2×3)\n",
    "\n",
    "Six panels packaged as `compare_rl_env.png`:\n",
    "\n",
    "1. **Top-left — Avg episode reward (±std):** error bars show variance across episodes.\n",
    "2. **Top-middle — Task completion rate:** percentage of episodes the model completed (`done=True`).\n",
    "3. **Top-right — Steps & Reward/Step grouped:** efficiency comparison side-by-side.\n",
    "4. **Bottom-left — Per-tier avg reward:** how each model scales across `warmup → expert`.\n",
    "5. **Bottom-middle — Per-tier completion rate:** where each model hits a wall.\n",
    "6. **Bottom-right — Reward distribution (box + jitter):** every episode's reward as an individual dot, overlaid on a box-and-whisker. The wider SFT box reflects that it tackles harder episodes (which have higher reward ceilings) — variance went up *and* the median moved up.\n",
    "\n",
    "> **Output:** the figure is rendered inline and saved with `Saved → compare_rl_env.png`.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "cell-rl-plots",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 958
    },
    "id": "cell-rl-plots",
    "outputId": "c1ec1dc2-7669-4648-c489-2a40ea4ed8ec"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_1212/2595381142.py:87: MatplotlibDeprecationWarning: The 'labels' parameter of boxplot() has been renamed 'tick_labels' since Matplotlib 3.9; support for the old name will be dropped in 3.11.\n",
      "  bp = rax6.boxplot(\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2400x1440 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved → compare_rl_env.png\n"
     ]
    }
   ],
   "source": [
    "# ═══════════════════════════════════════════════════════════════════════════\n",
    "# Figure 2 — RL env eval (2x3)\n",
    "# ═══════════════════════════════════════════════════════════════════════════\n",
    "TIER_ORDER = [\"warmup\", \"beginner\", \"intermediate\", \"advanced\", \"expert\"]\n",
    "difficulties = [d for d in TIER_ORDER\n",
    "                if d in base_rl_metrics[\"_per_diff\"] and d in sft_rl_metrics[\"_per_diff\"]]\n",
    "\n",
    "fig2 = plt.figure(figsize=(20, 12))\n",
    "gs2  = GridSpec(2, 3, figure=fig2, hspace=0.5, wspace=0.38)\n",
    "rax1 = fig2.add_subplot(gs2[0, 0])   # avg episode reward\n",
    "rax2 = fig2.add_subplot(gs2[0, 1])   # completion rate\n",
    "rax3 = fig2.add_subplot(gs2[0, 2])   # avg steps & reward/step\n",
    "rax4 = fig2.add_subplot(gs2[1, 0])   # per-difficulty reward\n",
    "rax5 = fig2.add_subplot(gs2[1, 1])   # per-difficulty completion\n",
    "rax6 = fig2.add_subplot(gs2[1, 2])   # reward distribution (box)\n",
    "\n",
    "models     = [\"Base\", \"SFT\"]\n",
    "x2         = np.arange(len(models))\n",
    "bar_colors = [COLOR_BASE, COLOR_SFT]\n",
    "\n",
    "# 2a. Avg episode reward ± std\n",
    "avg_rewards = [base_rl_metrics[\"avg_episode_reward\"], sft_rl_metrics[\"avg_episode_reward\"]]\n",
    "reward_stds = [base_rl_metrics[\"reward_std\"],         sft_rl_metrics[\"reward_std\"]]\n",
    "bars_r = rax1.bar(x2, avg_rewards, 0.45, color=bar_colors,\n",
    "                  edgecolor=\"white\", linewidth=0.6,\n",
    "                  yerr=reward_stds, capsize=5, error_kw={\"elinewidth\": 1.5})\n",
    "for bar, v in zip(bars_r, avg_rewards):\n",
    "    rax1.text(bar.get_x() + bar.get_width()/2,\n",
    "              bar.get_height() + max(reward_stds) * 0.1 + 0.05,\n",
    "              f\"{v:.2f}\", ha=\"center\", fontsize=9, fontweight=\"bold\")\n",
    "rax1.set(title=\"Avg Episode Reward (±std)\", ylabel=\"Total Reward\", xlabel=\"Model\")\n",
    "rax1.set_xticks(x2); rax1.set_xticklabels(models); rax1.set_axisbelow(True)\n",
    "\n",
    "# 2b. Task completion rate\n",
    "comp_rates = [base_rl_metrics[\"completion_rate\"] * 100,\n",
    "              sft_rl_metrics[\"completion_rate\"]  * 100]\n",
    "bars_c = rax2.bar(x2, comp_rates, 0.45, color=bar_colors, edgecolor=\"white\", linewidth=0.6)\n",
    "for bar, v in zip(bars_c, comp_rates):\n",
    "    rax2.text(bar.get_x() + bar.get_width()/2, bar.get_height() + 0.8,\n",
    "              f\"{v:.0f}%\", ha=\"center\", fontsize=9, fontweight=\"bold\")\n",
    "rax2.set(title=\"Task Completion Rate\", ylabel=\"% Episodes Completed\",\n",
    "         xlabel=\"Model\", ylim=(0, 115))\n",
    "rax2.set_xticks(x2); rax2.set_xticklabels(models); rax2.set_axisbelow(True)\n",
    "\n",
    "# 2c. Avg steps + reward/step (grouped)\n",
    "step_vals = [base_rl_metrics[\"avg_steps\"],           sft_rl_metrics[\"avg_steps\"]]\n",
    "rps_vals  = [base_rl_metrics[\"avg_reward_per_step\"], sft_rl_metrics[\"avg_reward_per_step\"]]\n",
    "w3 = 0.35\n",
    "for i, (vals, label) in enumerate([(step_vals, \"Avg Steps\"), (rps_vals, \"Reward/Step\")]):\n",
    "    bars_ = rax3.bar(x2 + (i - 0.5) * w3, vals, w3,\n",
    "                     color=bar_colors, edgecolor=\"white\",\n",
    "                     linewidth=0.6, label=label, alpha=0.85 if i else 1.0)\n",
    "    for bar, v in zip(bars_, vals):\n",
    "        rax3.text(bar.get_x() + bar.get_width()/2, bar.get_height() + 0.05,\n",
    "                  f\"{v:.2f}\", ha=\"center\", fontsize=8, fontweight=\"bold\")\n",
    "rax3.set(title=\"Steps & Reward Efficiency\", ylabel=\"Value\", xlabel=\"Model\")\n",
    "rax3.set_xticks(x2); rax3.set_xticklabels(models)\n",
    "rax3.legend(loc=\"upper right\"); rax3.set_axisbelow(True)\n",
    "\n",
    "# 2d. Per-difficulty avg reward (all 5 tiers)\n",
    "xd, wd   = np.arange(len(difficulties)), 0.35\n",
    "base_dr  = [base_rl_metrics[\"_per_diff\"][d][\"avg_reward\"] for d in difficulties]\n",
    "sft_dr   = [sft_rl_metrics[\"_per_diff\"][d][\"avg_reward\"]  for d in difficulties]\n",
    "annotate_bars(rax4, rax4.bar(xd - wd/2, base_dr, wd, color=COLOR_BASE,\n",
    "              label=\"Base\", edgecolor=\"white\"), fmt=\"{:.2f}\", offset=0.02)\n",
    "annotate_bars(rax4, rax4.bar(xd + wd/2, sft_dr,  wd, color=COLOR_SFT,\n",
    "              label=\"SFT\",  edgecolor=\"white\"), fmt=\"{:.2f}\", offset=0.02)\n",
    "rax4.set(title=\"Avg Reward by Difficulty Tier\", ylabel=\"Avg Episode Reward\", xlabel=\"Tier\")\n",
    "rax4.set_xticks(xd)\n",
    "rax4.set_xticklabels([d.capitalize() for d in difficulties], rotation=15, ha=\"right\")\n",
    "rax4.legend(); rax4.set_axisbelow(True)\n",
    "\n",
    "# 2e. Per-difficulty completion rate (all 5 tiers)\n",
    "base_dc = [base_rl_metrics[\"_per_diff\"][d][\"completion_rate\"] * 100 for d in difficulties]\n",
    "sft_dc  = [sft_rl_metrics[\"_per_diff\"][d][\"completion_rate\"]  * 100 for d in difficulties]\n",
    "annotate_bars(rax5, rax5.bar(xd - wd/2, base_dc, wd, color=COLOR_BASE,\n",
    "              label=\"Base\", edgecolor=\"white\"), fmt=\"{:.0f}%\", offset=0.8)\n",
    "annotate_bars(rax5, rax5.bar(xd + wd/2, sft_dc,  wd, color=COLOR_SFT,\n",
    "              label=\"SFT\",  edgecolor=\"white\"), fmt=\"{:.0f}%\", offset=0.8)\n",
    "rax5.set(title=\"Completion Rate by Difficulty Tier\",\n",
    "         ylabel=\"% Episodes Completed\", xlabel=\"Tier\", ylim=(0, 120))\n",
    "rax5.set_xticks(xd)\n",
    "rax5.set_xticklabels([d.capitalize() for d in difficulties], rotation=15, ha=\"right\")\n",
    "rax5.legend(); rax5.set_axisbelow(True)\n",
    "\n",
    "# 2f. Reward distribution — box + jitter\n",
    "bp = rax6.boxplot(\n",
    "    [base_rl_metrics[\"_rewards\"], sft_rl_metrics[\"_rewards\"]],\n",
    "    labels=[\"Base\", \"SFT\"],\n",
    "    patch_artist=True,\n",
    "    medianprops={\"color\": \"white\", \"linewidth\": 2},\n",
    "    whiskerprops={\"linewidth\": 1.5},\n",
    "    capprops={\"linewidth\": 1.5},\n",
    "    flierprops={\"marker\": \"o\", \"markersize\": 5, \"alpha\": 0.6},\n",
    "    widths=0.4,\n",
    ")\n",
    "for patch, color in zip(bp[\"boxes\"], [COLOR_BASE, COLOR_SFT]):\n",
    "    patch.set_facecolor(color); patch.set_alpha(0.85)\n",
    "for i, (rewards, color) in enumerate(\n",
    "    [(base_rl_metrics[\"_rewards\"], COLOR_BASE),\n",
    "     (sft_rl_metrics[\"_rewards\"],  COLOR_SFT)], start=1\n",
    "):\n",
    "    jitter = np.random.uniform(-0.08, 0.08, len(rewards))\n",
    "    rax6.scatter(np.full(len(rewards), i) + jitter, rewards,\n",
    "                 color=color, alpha=0.7, zorder=3, s=30,\n",
    "                 edgecolors=\"white\", linewidths=0.4)\n",
    "rax6.set(title=\"Episode Reward Distribution\",\n",
    "         ylabel=\"Total Episode Reward\", xlabel=\"Model\")\n",
    "rax6.set_axisbelow(True)\n",
    "\n",
    "fig2.suptitle(\n",
    "    f\"Part 2 — RL Env Eval: Base vs SFT  \"\n",
    "    f\"({len(difficulties)} tiers × {RL_EPISODES_PER_DIFF} episodes each)\",\n",
    "    fontsize=15, fontweight=\"bold\", y=1.01,\n",
    ")\n",
    "plt.savefig(\"compare_rl_env.png\", dpi=150, bbox_inches=\"tight\", facecolor=\"white\")\n",
    "plt.show()\n",
    "print(\"Saved → compare_rl_env.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "md-summary",
   "metadata": {},
   "source": [
    "## 📋 Numerical Summary\n",
    "\n",
    "Side-by-side table of every metric in both eval modes, plus a delta column. This is the print-friendly version of what the plots show.\n",
    "\n",
    "> **Output highlights:**\n",
    ">\n",
    "> - Dataset format compliance: **33.3% → 100% (+66.7 pp)**\n",
    "> - Dataset exact match: **38.9% → 88.9% (+50.0 pp)**\n",
    "> - RL env completion rate: **46.7% → 73.3% (+26.7 pp)**\n",
    "> - RL env reward per step: **0.138 → 0.351 (+154%)**\n",
    "> - Latency went **down** in both modes — SFT generates fewer tokens because it stopped padding output with prose.\n",
    ">\n",
    "> The reward standard deviation rose (`+0.77`); that's expected — the SFT model attempts the harder tiers, so per-episode reward swings between low (timeout) and high (jackpot) instead of clustering near the easy-task ceiling. **More variance is a feature here, not a bug** — it means the model is engaging with episodes the base couldn't even start.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "cell-summary-table",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cell-summary-table",
    "outputId": "de39bae0-ef24-4d5c-d84f-0c0134ed671c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "========================================================================\n",
      "DATASET EVAL SUMMARY\n",
      "========================================================================\n",
      "Metric                               Base        SFT        Delta\n",
      "----------------------------------------------------------------\n",
      "format_pct                          33.3%     100.0%       +66.7pt\n",
      "format_after_extract_pct           100.0%     100.0%        +0.0pt\n",
      "exact_pct                           38.9%      88.9%       +50.0pt\n",
      "service_pct                         77.8%      88.9%       +11.1pt\n",
      "operation_pct                       61.1%      88.9%       +27.8pt\n",
      "avg_latency                         1.901      1.559       -0.342\n",
      "avg_len                            85.833     74.722      -11.111\n",
      "\n",
      "========================================================================\n",
      "RL ENV EVAL SUMMARY\n",
      "========================================================================\n",
      "Metric                          Base        SFT        Delta\n",
      "------------------------------------------------------------\n",
      "avg_episode_reward             1.187      2.011       +0.824\n",
      "reward_std                     1.137      1.908       +0.771\n",
      "completion_rate                46.7%      73.3%       +26.7pt\n",
      "avg_steps                      8.600      5.733       -2.867\n",
      "avg_reward_per_step            0.138      0.351       +0.213\n"
     ]
    }
   ],
   "source": [
    "print(\"=\" * 72)\n",
    "print(\"DATASET EVAL SUMMARY\")\n",
    "print(\"=\" * 72)\n",
    "ds_keys = [\"format_pct\", \"format_after_extract_pct\", \"exact_pct\",\n",
    "           \"service_pct\", \"operation_pct\", \"avg_latency\", \"avg_len\"]\n",
    "print(f\"{'Metric':<30} {'Base':>10} {'SFT':>10} {'Delta':>12}\")\n",
    "print(\"-\" * 64)\n",
    "for k in ds_keys:\n",
    "    b, s = base_ds_metrics[k], sft_ds_metrics[k]\n",
    "    if \"pct\" in k:\n",
    "        print(f\"{k:<30} {100*b:9.1f}% {100*s:9.1f}% {100*(s-b):+11.1f}pt\")\n",
    "    else:\n",
    "        print(f\"{k:<30} {b:10.3f} {s:10.3f} {s-b:+12.3f}\")\n",
    "\n",
    "print()\n",
    "print(\"=\" * 72)\n",
    "print(\"RL ENV EVAL SUMMARY\")\n",
    "print(\"=\" * 72)\n",
    "rl_keys = [\"avg_episode_reward\", \"reward_std\", \"completion_rate\",\n",
    "           \"avg_steps\", \"avg_reward_per_step\"]\n",
    "print(f\"{'Metric':<25} {'Base':>10} {'SFT':>10} {'Delta':>12}\")\n",
    "print(\"-\" * 60)\n",
    "for k in rl_keys:\n",
    "    b, s = base_rl_metrics[k], sft_rl_metrics[k]\n",
    "    if k == \"completion_rate\":\n",
    "        print(f\"{k:<25} {100*b:9.1f}% {100*s:9.1f}% {100*(s-b):+11.1f}pt\")\n",
    "    else:\n",
    "        print(f\"{k:<25} {b:10.3f} {s:10.3f} {s-b:+12.3f}\")"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "gpuType": "T4",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.10"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "00ff165127094d2fb481980c130f04c8": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "01e0602c676442ff952a7a7b3f4024c7": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "02068d599f844ea5838544754e6d8448": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "052ba95707274c42933cb13fc8ad078c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "06c2b0aff48f40a886e14723ee788248": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "0803a7b3995241f493052389046afe98": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "082ceda90b8d484a81a1de5fead892b6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_e8560b1f7c5a42a1b850a24bd22f9a73",
       "IPY_MODEL_5a5d3dac3d2f43419349404bd0c79f4f",
       "IPY_MODEL_4772a353fbb243aca9f472de257d1358"
      ],
      "layout": "IPY_MODEL_b14fbeb1fb4f4888b63afd9cf4476b28"
     }
    },
    "0ac4cc8e789d4a7389ba2d81f752b2a6": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "0cf6a91ad33f4f0a94b890881658b5b2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "0d2747470d4947849c27775b43ecb54c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_fbc88709dcf84cfbb91d7efbb10e0b42",
       "IPY_MODEL_cc4366820b234778b543af95146a4e92",
       "IPY_MODEL_0fba0e0c04c14214ae58cdd3f65fd92f"
      ],
      "layout": "IPY_MODEL_4dae08a1f54842e49185c4402659df6f"
     }
    },
    "0f614193015744f2a0263aa13681b411": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_cc427e51ea1446aaa74d6f24cb17d043",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_a8816b70cd9f409ba7acfacc2864c043",
      "value": 1
     }
    },
    "0fba0e0c04c14214ae58cdd3f65fd92f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_96abbd4be45f4eef9587a795f10c73f9",
      "placeholder": "​",
      "style": "IPY_MODEL_6736d9704da04076a134426a47cb15c4",
      "value": " 194k/194k [00:01&lt;00:00, 965kB/s]"
     }
    },
    "1161e043c93c4671a8155dce8d9dd701": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "1241be723444464fbb334b06d8208410": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "161b757d5f64481a9212dbccd7a884c7": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_89fdaf865e0e4f7fa1909f032005d1d8",
      "placeholder": "​",
      "style": "IPY_MODEL_8060b0a254d54032bbb01a56431e98eb",
      "value": "generation_config.json: 100%"
     }
    },
    "17ad15eed85f493ca0d0d6b8d914ca2c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "182361253eee49418048b01fbcf2672f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_79d5b6596f1b4ee5966b55dad97df3a2",
      "max": 200,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_b49e779bf02c49bbb1f4eba5ac2dd1e2",
      "value": 200
     }
    },
    "188a00a386704c7887d139bc284ccc1a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "1a190eca13234bca86da0f33e46a34e1": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1dc1444225ce43a888350992cc7f0bbb": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_9ef9a30c556c4d9da3692fd8a829270f",
       "IPY_MODEL_b14329f3b61447f788d6e836bf934c91",
       "IPY_MODEL_6cff000d81e8431298609a9148cdb793"
      ],
      "layout": "IPY_MODEL_49de19f8c9e04603980b4a4a8c816545"
     }
    },
    "1e3372b4af51443f88050b77fedc5bef": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "20e58c5f72e14d47901d1eec0a08c12f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "217d16de232c4a1d83a1a4b48ad453d6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "24dadb01295d447ca81ac02a4908fdf1": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2b446f04958943a4adc493cb624ecb7e": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2c51081cf8314bd58a8b7b93cdf2eed7": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2dbe35d37c824c839a2913a0cfbc5266": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2def96a8b6d646ef8801ce2c1e0a8d14": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "2e707c60d0384e8b8c993c883ad3456a": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2f17b452a3f04d448beff33d60445e2e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_cd77111d0ee44ecaacb5159f68c88609",
      "placeholder": "​",
      "style": "IPY_MODEL_89604435feab456888553549728c15a7",
      "value": "Generating validation split: 100%"
     }
    },
    "2f1ae5179b0945fda20f970c25c93c57": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2fe800ca1f6948009a218f9f3ce0acbb": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f13c88be577549acad8e2606a8c4e6f2",
      "placeholder": "​",
      "style": "IPY_MODEL_c1b370ab7fee4dbeb2cb37f47f774eb6",
      "value": "merges.txt: "
     }
    },
    "31d96fb276614537bab907c316f0ce25": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_f9ec2b3ae5604344bf3a5f267ce7de38",
       "IPY_MODEL_4772c50a4e454979acb7cc4de466937a",
       "IPY_MODEL_6ee363978e9749d7b39191a4027be66a"
      ],
      "layout": "IPY_MODEL_b74c48e99fa742ad8b44d5bb7fd34171"
     }
    },
    "3332ae39c1094b3bb3fde04198a3f2a6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_71cbf065dfca4450a615fdef7c2114f6",
       "IPY_MODEL_fdadb7a27d07425c8e04dd10df1aeba1",
       "IPY_MODEL_ec590a2a32d045b194dab35ae5273043"
      ],
      "layout": "IPY_MODEL_826ab03235c94f929ca1e261327a0137"
     }
    },
    "358628764f24466198ceadb0f4478487": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_20e58c5f72e14d47901d1eec0a08c12f",
      "max": 613,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_217d16de232c4a1d83a1a4b48ad453d6",
      "value": 613
     }
    },
    "375ea535f86c4dd193d114568bb3045b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "39929719372644589a9baa7295fd5849": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "399a72c355d8478e87995aff1c57f426": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_564a55b31d504a2499140061de7c8354",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_db821f8b3e1b414590bbde186cc62d59",
      "value": 1
     }
    },
    "3af6672587a84a2b886482e68fcfbba9": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "3cf95de334a14af9b7274dee96b2f20d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_41a765ae42f142a5aaf0163612bbe490",
       "IPY_MODEL_0f614193015744f2a0263aa13681b411",
       "IPY_MODEL_68d44bd4ff9a439f958573c8ec16405b"
      ],
      "layout": "IPY_MODEL_c0c12b0fc46f4b9f8ea2af23e194da23"
     }
    },
    "41873020c9324e8b96fb6a8a7e432ed2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "41a765ae42f142a5aaf0163612bbe490": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_796833bcdc3845849bf4b44da6379fcc",
      "placeholder": "​",
      "style": "IPY_MODEL_b22885fcbf3147509191939f5de83602",
      "value": "README.md: "
     }
    },
    "44fb962be403416c910b1c4bf25bcdf1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_161b757d5f64481a9212dbccd7a884c7",
       "IPY_MODEL_a18521a6e11b4e1ea6e53ce81b2af574",
       "IPY_MODEL_969a9823b8864c0f877eb2c3af383205"
      ],
      "layout": "IPY_MODEL_753cba0957c240cd8d7fcf171dd2557e"
     }
    },
    "4772a353fbb243aca9f472de257d1358": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f2247b50901c4a20854c1ac284d76e7f",
      "placeholder": "​",
      "style": "IPY_MODEL_71c857a576e04c1592352df2553a10f6",
      "value": " 2.05G/2.05G [00:14&lt;00:00, 159MB/s]"
     }
    },
    "4772c50a4e454979acb7cc4de466937a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_ef8ace2a258a479db27919c2527a172f",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_02068d599f844ea5838544754e6d8448",
      "value": 1
     }
    },
    "49de19f8c9e04603980b4a4a8c816545": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4dae08a1f54842e49185c4402659df6f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4dca96c48eb64681b36a76b4f0b01b62": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4e6c58ce96404281b880833f88e08413": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4f00645b01294756ae17955d5889514b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_2e707c60d0384e8b8c993c883ad3456a",
      "placeholder": "​",
      "style": "IPY_MODEL_3af6672587a84a2b886482e68fcfbba9",
      "value": "special_tokens_map.json: 100%"
     }
    },
    "5178b799d7404fd282f4e5016d616d14": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "51ad078171814b9ca022d88231cf6387": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5329bea02fc2432b84b8bb8eb733eaf9": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5405361924614557867c74ba647b08b4": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "548cc14a659b4e6e9c217ff980e34608": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "54c323cb9fe9458fb062b2d4bd5cf927": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "55a60ff24a23452682aebac4cdcdf086": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_923264d53c454d9d96d618b0efd3f59a",
      "placeholder": "​",
      "style": "IPY_MODEL_8b918bfb9f2f4ee0a71884837b495ebf",
      "value": " 1.67M/? [00:00&lt;00:00, 43.5MB/s]"
     }
    },
    "564a55b31d504a2499140061de7c8354": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "20px"
     }
    },
    "57cd591f5b3f4d6aa2e401dcb0c3c8fc": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5a09cd99ab3444248c365ba3e1f46436": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5a5d3dac3d2f43419349404bd0c79f4f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_8c91b59afcd343648321a9ed206531a7",
      "max": 2054625552,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_a3710ca3ab474ce59a146b54fea25736",
      "value": 2054625552
     }
    },
    "5bbbc6530c014b5296c181526d08cec0": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5cdf05f2225f499e8574ab64ec9a8766": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_66558440f90b42d99278658f7151f043",
      "max": 150,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_188a00a386704c7887d139bc284ccc1a",
      "value": 150
     }
    },
    "5e9776eebdc544159f73ebc7447d54f1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_ee387cb17b58451fab83e2b3241032dc",
      "max": 14783936,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_c402601073394089ab3718d6288660ff",
      "value": 14783936
     }
    },
    "616c3d44f0a64e22828c2e472a8c13bc": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_0cf6a91ad33f4f0a94b890881658b5b2",
      "placeholder": "​",
      "style": "IPY_MODEL_5178b799d7404fd282f4e5016d616d14",
      "value": " 1.92M/1.92M [00:00&lt;00:00, 9.61MB/s]"
     }
    },
    "617dcbb6224b49bb9b83dac185420e09": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_7c93228f55074cf78b1170001545f89d",
       "IPY_MODEL_b1e2bcaf50124b14b7bc70fb403e68b8",
       "IPY_MODEL_c36279d8d0774918812725c906a8af12"
      ],
      "layout": "IPY_MODEL_ebb9a1dea358429e9e166c470076d4ce"
     }
    },
    "62a7c1256a3d462ca1f58711b4985852": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "62ba79bd0e2f47fcbb073cbe15ab2bf6": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "63b9d68de11b4fbfa5e8acaaa25b4c35": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "66558440f90b42d99278658f7151f043": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "66f989e6acee487faf8f6cc453faa486": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "6736d9704da04076a134426a47cb15c4": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "68000393fa504271ba1a4c0dfb18a182": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_ea95e9fdfc1e43ec8c1cb7b35d24f436",
       "IPY_MODEL_5e9776eebdc544159f73ebc7447d54f1",
       "IPY_MODEL_d953a7141b174afdbf2f8912cdf4e04f"
      ],
      "layout": "IPY_MODEL_2b446f04958943a4adc493cb624ecb7e"
     }
    },
    "68d44bd4ff9a439f958573c8ec16405b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_c71fef24bafe4b45896601a7ae144c2f",
      "placeholder": "​",
      "style": "IPY_MODEL_548cc14a659b4e6e9c217ff980e34608",
      "value": " 4.89k/? [00:00&lt;00:00, 314kB/s]"
     }
    },
    "68d465d228c141a7b5c8c97799102790": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_00ff165127094d2fb481980c130f04c8",
      "placeholder": "​",
      "style": "IPY_MODEL_d2af9a4571554901a04f1afdd556984d",
      "value": "tokenizer_config.json: "
     }
    },
    "6cff000d81e8431298609a9148cdb793": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_741c77c00dd34486bd780369d2918d53",
      "placeholder": "​",
      "style": "IPY_MODEL_b6c49d1dbf184f86a4a94644340b9f90",
      "value": " 1500/1500 [00:00&lt;00:00, 29631.67 examples/s]"
     }
    },
    "6eb2f500acda438a82e453126b22ae25": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "20px"
     }
    },
    "6ee363978e9749d7b39191a4027be66a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_5a09cd99ab3444248c365ba3e1f46436",
      "placeholder": "​",
      "style": "IPY_MODEL_375ea535f86c4dd193d114568bb3045b",
      "value": " 7.03M/? [00:00&lt;00:00, 98.9MB/s]"
     }
    },
    "71c857a576e04c1592352df2553a10f6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "71cbf065dfca4450a615fdef7c2114f6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_06c2b0aff48f40a886e14723ee788248",
      "placeholder": "​",
      "style": "IPY_MODEL_0803a7b3995241f493052389046afe98",
      "value": "added_tokens.json: 100%"
     }
    },
    "741c77c00dd34486bd780369d2918d53": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "753cba0957c240cd8d7fcf171dd2557e": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "7799ec6190d04928963fbd82d195e2b3": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_68d465d228c141a7b5c8c97799102790",
       "IPY_MODEL_9f01388850d44b30a155bdd6c1528de4",
       "IPY_MODEL_a5a23a6849e240be90007a200a14e1a3"
      ],
      "layout": "IPY_MODEL_1a190eca13234bca86da0f33e46a34e1"
     }
    },
    "796833bcdc3845849bf4b44da6379fcc": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "79d5b6596f1b4ee5966b55dad97df3a2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "7c93228f55074cf78b1170001545f89d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_5bbbc6530c014b5296c181526d08cec0",
      "placeholder": "​",
      "style": "IPY_MODEL_8a2f80c0be5a47b994f6e091cb01d287",
      "value": "data/reserve-00000-of-00001.parquet: 100%"
     }
    },
    "7ec0191160e04c5397c24bbad3d2b593": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_51ad078171814b9ca022d88231cf6387",
      "placeholder": "​",
      "style": "IPY_MODEL_ed06bfd4c13f41598d5e10bbe6139eb8",
      "value": "data/train-00000-of-00001.parquet: 100%"
     }
    },
    "7f28dc39b40543d9b17bd8ad611e3623": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "8060b0a254d54032bbb01a56431e98eb": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "80d066d9f0614ae4b264ade4a304cf49": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "80d9e1dd1507439088c3d9d8c4c7b0ac": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_ceadf3915006459fb72ed9141c49dc2f",
      "placeholder": "​",
      "style": "IPY_MODEL_63b9d68de11b4fbfa5e8acaaa25b4c35",
      "value": " 150/150 [00:00&lt;00:00, 4766.72 examples/s]"
     }
    },
    "826ab03235c94f929ca1e261327a0137": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "874de7ba061b4d8783c74e605440394e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_2dbe35d37c824c839a2913a0cfbc5266",
      "placeholder": "​",
      "style": "IPY_MODEL_f9f3f435180b4929a938ac4500216247",
      "value": " 613/613 [00:00&lt;00:00, 59.9kB/s]"
     }
    },
    "88f4f5a8f65742eb8cb1228e5d94508c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "89604435feab456888553549728c15a7": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "89fdaf865e0e4f7fa1909f032005d1d8": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "8a2f80c0be5a47b994f6e091cb01d287": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "8af925f00d3541958d318f43591bdd76": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_e87a2309352248309f209b71dd050799",
      "placeholder": "​",
      "style": "IPY_MODEL_1e3372b4af51443f88050b77fedc5bef",
      "value": " 200/200 [00:00&lt;00:00, 5397.52 examples/s]"
     }
    },
    "8b918bfb9f2f4ee0a71884837b495ebf": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "8c91b59afcd343648321a9ed206531a7": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "923264d53c454d9d96d618b0efd3f59a": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "94eb4cb08c8443098267d78b1665514e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "969a9823b8864c0f877eb2c3af383205": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f105bcab30d8404a9b3a1ce866480374",
      "placeholder": "​",
      "style": "IPY_MODEL_9c78bf775e4149cfbe34f1e253c2c9fc",
      "value": " 266/266 [00:00&lt;00:00, 23.7kB/s]"
     }
    },
    "96abbd4be45f4eef9587a795f10c73f9": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "974d765200034661a14a25e0b241fdeb": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "98dada6fc260427ebe46383263de1e75": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "9c1e0ff6a9094793905ddc191e95796d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_ea1e771eea51498b9917da282d783427",
      "placeholder": "​",
      "style": "IPY_MODEL_39929719372644589a9baa7295fd5849",
      "value": "Generating reserve split: 100%"
     }
    },
    "9c78bf775e4149cfbe34f1e253c2c9fc": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "9d1f05eb5d3b4bdba57d618f022eaa7b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_7ec0191160e04c5397c24bbad3d2b593",
       "IPY_MODEL_bf317a89297c4265938863cc3f74ba8c",
       "IPY_MODEL_616c3d44f0a64e22828c2e472a8c13bc"
      ],
      "layout": "IPY_MODEL_57cd591f5b3f4d6aa2e401dcb0c3c8fc"
     }
    },
    "9d32149562cf47859390636be9c70d76": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "9e1d365fa1c84030a4caeead8ef954d3": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_6eb2f500acda438a82e453126b22ae25",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_9d32149562cf47859390636be9c70d76",
      "value": 1
     }
    },
    "9ef9a30c556c4d9da3692fd8a829270f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_4dca96c48eb64681b36a76b4f0b01b62",
      "placeholder": "​",
      "style": "IPY_MODEL_66f989e6acee487faf8f6cc453faa486",
      "value": "Generating train split: 100%"
     }
    },
    "9f01388850d44b30a155bdd6c1528de4": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_d06befc9db47458fa696880e19449d90",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_aadd0d6341714668ae5a85c445f49a09",
      "value": 1
     }
    },
    "9f61db866c9c41e6af53f0b79f55b4e7": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "a18521a6e11b4e1ea6e53ce81b2af574": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_80d066d9f0614ae4b264ade4a304cf49",
      "max": 266,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_bdccff3baee846459a9645d014266d2a",
      "value": 266
     }
    },
    "a3710ca3ab474ce59a146b54fea25736": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "a5a23a6849e240be90007a200a14e1a3": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_5329bea02fc2432b84b8bb8eb733eaf9",
      "placeholder": "​",
      "style": "IPY_MODEL_ba6c2ac768044ec0b688e6573e540f92",
      "value": " 7.51k/? [00:00&lt;00:00, 521kB/s]"
     }
    },
    "a8816b70cd9f409ba7acfacc2864c043": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "aadd0d6341714668ae5a85c445f49a09": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "ab9aa9b9c93b4ee5900a351cd584cee8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_974d765200034661a14a25e0b241fdeb",
      "placeholder": "​",
      "style": "IPY_MODEL_41873020c9324e8b96fb6a8a7e432ed2",
      "value": " 2.78M/? [00:00&lt;00:00, 55.7MB/s]"
     }
    },
    "b14329f3b61447f788d6e836bf934c91": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_052ba95707274c42933cb13fc8ad078c",
      "max": 1500,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_daf23f97138a44f49f8099d73a39b33b",
      "value": 1500
     }
    },
    "b14fbeb1fb4f4888b63afd9cf4476b28": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b1e2bcaf50124b14b7bc70fb403e68b8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_fbb8cb04661945c782b1df5e423b37a0",
      "max": 260654,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_94eb4cb08c8443098267d78b1665514e",
      "value": 260654
     }
    },
    "b22885fcbf3147509191939f5de83602": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "b3d0061be3d24e22af12225fb08f02ee": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b49e779bf02c49bbb1f4eba5ac2dd1e2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "b6c49d1dbf184f86a4a94644340b9f90": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "b71f5888837e4717bc8669603cb587a2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b74c48e99fa742ad8b44d5bb7fd34171": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ba375ee09b334728b8cfc63975f4302a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_4f00645b01294756ae17955d5889514b",
       "IPY_MODEL_358628764f24466198ceadb0f4478487",
       "IPY_MODEL_874de7ba061b4d8783c74e605440394e"
      ],
      "layout": "IPY_MODEL_01e0602c676442ff952a7a7b3f4024c7"
     }
    },
    "ba6c2ac768044ec0b688e6573e540f92": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "bab57aecb3f84216af720fa9321f6fb8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_2fe800ca1f6948009a218f9f3ce0acbb",
       "IPY_MODEL_9e1d365fa1c84030a4caeead8ef954d3",
       "IPY_MODEL_55a60ff24a23452682aebac4cdcdf086"
      ],
      "layout": "IPY_MODEL_b71f5888837e4717bc8669603cb587a2"
     }
    },
    "bc4ee14a8b7140edba21f43a0d673045": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_88f4f5a8f65742eb8cb1228e5d94508c",
      "placeholder": "​",
      "style": "IPY_MODEL_54c323cb9fe9458fb062b2d4bd5cf927",
      "value": "vocab.json: "
     }
    },
    "bdccff3baee846459a9645d014266d2a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "bf317a89297c4265938863cc3f74ba8c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_62ba79bd0e2f47fcbb073cbe15ab2bf6",
      "max": 1923495,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_62a7c1256a3d462ca1f58711b4985852",
      "value": 1923495
     }
    },
    "c034e1899af543e7ab854b7276a787e2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "c07912b3d8974e6891c3d770391aa649": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "c0c12b0fc46f4b9f8ea2af23e194da23": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "c0cdb510c34c4e27b8061a5b4e4b88fa": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "c1b370ab7fee4dbeb2cb37f47f774eb6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "c36279d8d0774918812725c906a8af12": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f5eb0e8911ad46d19d1d76dfa55fb21c",
      "placeholder": "​",
      "style": "IPY_MODEL_1161e043c93c4671a8155dce8d9dd701",
      "value": " 261k/261k [00:00&lt;00:00, 1.30MB/s]"
     }
    },
    "c402601073394089ab3718d6288660ff": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "c545c0cc491b4692b6f7eaaf42e50baf": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "c71fef24bafe4b45896601a7ae144c2f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "c795b9a1dc834f67ac44a568b9744ed7": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "c9572ada407547acb82b499b4aba9408": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_2f17b452a3f04d448beff33d60445e2e",
       "IPY_MODEL_5cdf05f2225f499e8574ab64ec9a8766",
       "IPY_MODEL_80d9e1dd1507439088c3d9d8c4c7b0ac"
      ],
      "layout": "IPY_MODEL_4e6c58ce96404281b880833f88e08413"
     }
    },
    "cc427e51ea1446aaa74d6f24cb17d043": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "20px"
     }
    },
    "cc4366820b234778b543af95146a4e92": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_7f28dc39b40543d9b17bd8ad611e3623",
      "max": 193593,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_2def96a8b6d646ef8801ce2c1e0a8d14",
      "value": 193593
     }
    },
    "cd77111d0ee44ecaacb5159f68c88609": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ceadf3915006459fb72ed9141c49dc2f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "d06befc9db47458fa696880e19449d90": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "20px"
     }
    },
    "d2af9a4571554901a04f1afdd556984d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "d953a7141b174afdbf2f8912cdf4e04f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_24dadb01295d447ca81ac02a4908fdf1",
      "placeholder": "​",
      "style": "IPY_MODEL_c07912b3d8974e6891c3d770391aa649",
      "value": " 14.8M/14.8M [00:00&lt;00:00, 73.9MB/s]"
     }
    },
    "daf23f97138a44f49f8099d73a39b33b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "db821f8b3e1b414590bbde186cc62d59": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "e580572f40454f8ebb8d6d50cab3c644": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_bc4ee14a8b7140edba21f43a0d673045",
       "IPY_MODEL_399a72c355d8478e87995aff1c57f426",
       "IPY_MODEL_ab9aa9b9c93b4ee5900a351cd584cee8"
      ],
      "layout": "IPY_MODEL_2c51081cf8314bd58a8b7b93cdf2eed7"
     }
    },
    "e8560b1f7c5a42a1b850a24bd22f9a73": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_0ac4cc8e789d4a7389ba2d81f752b2a6",
      "placeholder": "​",
      "style": "IPY_MODEL_17ad15eed85f493ca0d0d6b8d914ca2c",
      "value": "model.safetensors: 100%"
     }
    },
    "e87a2309352248309f209b71dd050799": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e90dd825fc0b44d2bbe077f7361756af": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "ea1e771eea51498b9917da282d783427": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ea95e9fdfc1e43ec8c1cb7b35d24f436": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_2f1ae5179b0945fda20f970c25c93c57",
      "placeholder": "​",
      "style": "IPY_MODEL_e90dd825fc0b44d2bbe077f7361756af",
      "value": "adapter_model.safetensors: 100%"
     }
    },
    "ebb9a1dea358429e9e166c470076d4ce": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ec590a2a32d045b194dab35ae5273043": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_b3d0061be3d24e22af12225fb08f02ee",
      "placeholder": "​",
      "style": "IPY_MODEL_9f61db866c9c41e6af53f0b79f55b4e7",
      "value": " 632/632 [00:00&lt;00:00, 65.0kB/s]"
     }
    },
    "ed06bfd4c13f41598d5e10bbe6139eb8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "ee387cb17b58451fab83e2b3241032dc": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ef8ace2a258a479db27919c2527a172f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "20px"
     }
    },
    "f105bcab30d8404a9b3a1ce866480374": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "f13c88be577549acad8e2606a8c4e6f2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "f2247b50901c4a20854c1ac284d76e7f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "f5eb0e8911ad46d19d1d76dfa55fb21c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "f65560291a834aef846683d3909a9db6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_9c1e0ff6a9094793905ddc191e95796d",
       "IPY_MODEL_182361253eee49418048b01fbcf2672f",
       "IPY_MODEL_8af925f00d3541958d318f43591bdd76"
      ],
      "layout": "IPY_MODEL_c0cdb510c34c4e27b8061a5b4e4b88fa"
     }
    },
    "f9ec2b3ae5604344bf3a5f267ce7de38": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_c034e1899af543e7ab854b7276a787e2",
      "placeholder": "​",
      "style": "IPY_MODEL_c545c0cc491b4692b6f7eaaf42e50baf",
      "value": "tokenizer.json: "
     }
    },
    "f9f3f435180b4929a938ac4500216247": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "fbb8cb04661945c782b1df5e423b37a0": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "fbc88709dcf84cfbb91d7efbb10e0b42": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_c795b9a1dc834f67ac44a568b9744ed7",
      "placeholder": "​",
      "style": "IPY_MODEL_1241be723444464fbb334b06d8208410",
      "value": "data/validation-00000-of-00001.parquet: 100%"
     }
    },
    "fdadb7a27d07425c8e04dd10df1aeba1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_98dada6fc260427ebe46383263de1e75",
      "max": 632,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_5405361924614557867c74ba647b08b4",
      "value": 632
     }
    }
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}