notebooks/Introduction to Galactica Models.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "909676ae",
   "metadata": {},
   "source": [
    "# Introduction to GALACTICA Models\n",
    "\n",
    "Galactica is a family of language models trained on a novel high-quality scientific dataset, making the models capable of working with scientific terminology, math and chemical formulas as well as source codes.\n",
    "\n",
    "The easiest way to use the models is through our library called `galai` which provides convenience utilities to get the models, run generation and work with scientific entites of various types.\n",
    "\n",
    "This document is split into 5 main sections.\n",
    "\n",
    "* Quick Start\n",
    "* The `huge` Model Capabilities\n",
    "  + Citations\n",
    "  + Step-by-Step Reasoning\n",
    "  + Storage Knowledge\n",
    "  + Compositions\n",
    "* Text Generation & Sampling\n",
    "* Working with Large Models\n",
    "* Non-determinism\n",
    "* Pitfalls & Failure Examples\n",
    "\n",
    "\n",
    "**Note:** this notebook is best viewed using jupyter notebook or [nbviewer](https://nbviewer.org/). Other tools might not render all of our custom tokens, such as `<work>` (which should be rendered as `< work >` without spaces) or `</ s>` (which should be rendered as `< / s >` without spaces). You can also view the PDF version of this notebook available in the same directory."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a7c45ca",
   "metadata": {},
   "source": [
    "# Quick Start\n",
    "\n",
    "You can install the `galai` library using `pip` (requires `python>=3.7`):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bd328502",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install galai"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9185f1ee",
   "metadata": {},
   "source": [
    "Let's verify the installation by running generation with the base model (1.3B). We load it with:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4833fedf",
   "metadata": {},
   "outputs": [],
   "source": [
    "import galai as gal\n",
    "from galai.notebook_utils import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9080e5a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = gal.load_model(\"base\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8c30cc34",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'The Transformer architecture [START_REF] Attention is All you Need, Vaswani[END_REF] is a popular choice for sequence-to-sequence models. It consists of a stack of encoder and decoder layers, each of which is composed of a multi-head self-attention mechanism and a feed-forward network. The encoder is used to encode the'"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.generate(\"The Transformer architecture [START_REF]\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ded9905c",
   "metadata": {},
   "source": [
    "We can also generate math:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cb16c212",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "The Riemann zeta function is given by:\n",
       "\n",
       "$$ \\zeta(s)=\\sum_{n=1}^{\\infty}\\frac{1}{n^{s}},\\quad\\Re(s)&gt;1. $$\n",
       "\n",
       "The Riemann hypothesis (RH) states that the zeros of"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt = \"The Riemann zeta function is given by:\\n\\n\\\\[\"\n",
    "output = model.generate(prompt, max_new_tokens=60)\n",
    "display_latex(output)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "545b5294",
   "metadata": {},
   "source": [
    "There are 5 models in total (see more below in Model Selection Section):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6887504b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table><thead><tr><th>Name</th><th>Parameters</th><th>Layers</th><th>Heads</th><th>Head Size</th><th>Vocabulary Size</th><th>Context Size</th></tr></thead><tbody><tr><td><strong>mini</strong></td><td>125.0 M</td><td>12</td><td>12</td><td>64</td><td>50000</td><td>2048</td></tr><tr><td><strong>base</strong></td><td>1.3 B</td><td>24</td><td>32</td><td>64</td><td>50000</td><td>2048</td></tr><tr><td><strong>standard</strong></td><td>6.7 B</td><td>32</td><td>32</td><td>128</td><td>50000</td><td>2048</td></tr><tr><td><strong>large</strong></td><td>30.0 B</td><td>48</td><td>56</td><td>128</td><td>50000</td><td>2048</td></tr><tr><td><strong>huge</strong></td><td>121.3 B</td><td>96</td><td>80</td><td>128</td><td>50000</td><td>2048</td></tr></tbody></table>"
      ],
      "text/plain": [
       "[ModelInfo(name='mini', num_layers=12, num_heads=12, head_size=64, vocab_size=50000, max_positions=2048),\n",
       " ModelInfo(name='base', num_layers=24, num_heads=32, head_size=64, vocab_size=50000, max_positions=2048),\n",
       " ModelInfo(name='standard', num_layers=32, num_heads=32, head_size=128, vocab_size=50000, max_positions=2048),\n",
       " ModelInfo(name='large', num_layers=48, num_heads=56, head_size=128, vocab_size=50000, max_positions=2048),\n",
       " ModelInfo(name='huge', num_layers=96, num_heads=80, head_size=128, vocab_size=50000, max_positions=2048)]"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from galai.utils import ModelInfo\n",
    "ModelInfo.all()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e60c591",
   "metadata": {},
   "source": [
    "# The `huge` Model Capabilities"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13b4ee1c",
   "metadata": {},
   "source": [
    "In this Section we present the capabilities of the Galactica models. We use the `huge` 121 B parameters model with tensor parallelizm (see the Working with Large Models Section for more details):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9ea98dab",
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "model = gal.load_model(\"huge\", parallelize=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "394588b3",
   "metadata": {},
   "source": [
    "## Citations\n",
    "\n",
    "Galactica models are trained on a large corpus comprising more than 360 millions in-context citations and over 50 millions of unique references normalized across a diverse set of sources. This enables Galactica to suggest citations and help discover related papers.\n",
    "\n",
    "Each reference in our corpus is formatted as \"Title, First author\" and wrapped in a pair of `[START_REF]` / `[END_REF]` tokens. The tokens make it easy to steer the models into citing a reference:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "22ed4b21",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Galactica models are based on OPT architecture [START_REF] OPT: Open Pre-trained Transformer Language Models, Zhang[END_REF], which is a variant of the GPT-2 model [START_REF] Language Models are Unsupervised Multitask Learners, Radford[END_REF]. The OPT model is a 12-layer transformer with 12 attention heads and 768'"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.generate(\"Galactica models are based on OPT architecture [START_REF]\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6b26ebe",
   "metadata": {},
   "source": [
    "To make it easier to generate references we provide a convenience function `Model.generate_reference` that automatically handles the `[START_REF]` / `[END_REF]` tokens and avoid generating more output than necessary for faster inference:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "28b2b052",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'On the rapid computation of various polylogarithmic constants, Bailey'"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.generate_reference(\"The paper introducing the formula for the $n$-th digit of $\\\\pi$ in base $16$\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "839d59a6",
   "metadata": {},
   "source": [
    "The call above appends `[START_REF]` token to the prompt, and runs the generation up to the first occurence of `[END_REF]` token.\n",
    "\n",
    "> Please note that while in the example above the returned paper (\"On the rapid computation of various polylogarithmic constants\" by Bailey et al.) matches the description, the generations should be treated as suggestions of papers and should always be verified. Bear in mind that due to the non-determinizm (see Non-deterministic Generation Section for more information) your results might be different."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "93c563bf",
   "metadata": {},
   "source": [
    "The multiple modalities that Galactica is able to work with allows us to query for papers using math, source code, etc.:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1dfbd9d1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p><strong>Prompt</strong>: The paper that presented a novel computing block given by the formula:\n",
       "$$\n",
       "f(Q, K, V) = \\textrm{softmax}\\left(\\frac{QK^T}{\\sqrt{d_k}}\\right)V\n",
       "$$</p>\n",
       "<p><strong>Reference</strong>: Attention is All you Need, Vaswani</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt = \"\"\"The paper that presented a novel computing block given by the formula:\n",
    "\\\\[\n",
    "f(Q, K, V) = \\\\textrm{softmax}\\\\left(\\\\frac{QK^T}{\\\\sqrt{d_k}}\\\\right)V\n",
    "\\\\]\n",
    "\n",
    "\"\"\"\n",
    "reference = model.generate_reference(prompt)\n",
    "display_markdown(f\"**Prompt**: {prompt}\\n\\n**Reference**: {reference}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0ab54dae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p><strong>Prompt</strong>:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"k\">while</span> <span class=\"n\">k</span> <span class=\"o\">&gt;</span> <span class=\"mi\">1</span><span class=\"p\">:</span>\n",
       "    <span class=\"k\">if</span> <span class=\"n\">k</span> <span class=\"o\">%</span> <span class=\"mi\">2</span> <span class=\"o\">==</span> <span class=\"mi\">0</span><span class=\"p\">:</span>\n",
       "        <span class=\"n\">k</span> <span class=\"o\">=</span> <span class=\"n\">k</span> <span class=\"o\">//</span> <span class=\"mi\">2</span>\n",
       "    <span class=\"k\">else</span><span class=\"p\">:</span>\n",
       "        <span class=\"n\">k</span> <span class=\"o\">=</span> <span class=\"mi\">3</span> <span class=\"o\">*</span> <span class=\"n\">k</span> <span class=\"o\">+</span> <span class=\"mi\">1</span>\n",
       "</code></pre></div>\n",
       "\n",
       "<p>A paper studying if the loop above terminates for all positive integers </p>\n",
       "<p><strong>Reference</strong>: On the Collatz $3n+1$ algorithm, Garner</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt = \"\"\"```python\n",
    "while k > 1:\n",
    "    if k % 2 == 0:\n",
    "        k = k // 2\n",
    "    else:\n",
    "        k = 3 * k + 1\n",
    "```\n",
    "\n",
    "A paper studying if the loop above terminates for all positive integers \"\"\"\n",
    "reference = model.generate_reference(prompt)\n",
    "display_markdown(f\"**Prompt**:\\n{prompt}\\n\\n**Reference**: {reference}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7d7b48bf",
   "metadata": {},
   "source": [
    "You can get multiple suggestions of reference for a given prompt by setting `suggestions` parameter. With `suggestions > 1` a beam search decoding is used to try to generate more suggestions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fb5ada04",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Amyloid Hypothesis of Alzheimer's Disease: Progress and Problems on the Road to Therapeutics, Hardy\n",
      "The amyloid cascade hypothesis for Alzheimer's disease: an appraisal for the development of therapeutics, Karran\n",
      "The amyloid hypothesis of Alzheimer's disease at 25 years, Selkoe\n",
      "The amyloid hypothesis of Alzheimer's disease at 25 years, Selkoe\n",
      "The amyloid hypothesis of Alzheimer's disease at 25 years, Selkoe\n"
     ]
    }
   ],
   "source": [
    "for reference in model.generate_reference(\n",
    "    \"A survey paper on the amyloid hypothesis\",\n",
    "    suggestions=5\n",
    "):\n",
    "    print(reference)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e36dbf5d",
   "metadata": {},
   "source": [
    "As apparent from the example above, some of the references may repeat. Setting `diversity_penalty` to a number between `0.0` and `1.0` switches the generation algorithm to [Diverse beam search](https://arxiv.org/pdf/1610.02424.pdf):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e5f81db0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The amyloid hypothesis of Alzheimer's disease at 25 years, Selkoe\n",
      "Alzheimer's disease: the amyloid cascade hypothesis., Hardy\n",
      "The Amyloid Hypothesis of Alzheimer's Disease: Progress and Problems on the Road to Therapeutics, Hardy\n",
      "Amyloid-β and tau: the trigger and bullet in Alzheimer disease pathogenesis., Bloom\n",
      "The amyloid hypothesis of Alzheimer's disease at 25 years, Selkoe\n"
     ]
    }
   ],
   "source": [
    "for reference in model.generate_reference(\n",
    "    \"A survey paper on the amyloid hypothesis\",\n",
    "    suggestions=5, diversity_penalty=0.9\n",
    "):\n",
    "    print(reference)"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "bd545100",
   "metadata": {},
   "source": [
    "### Citation Distribution Bias\n",
    "\n",
    "Language models may encode and amplify biases present in the training corpus. Galactica models are biased towards referencing more frequently cited papers. Even though our analysis shows that as the model size increases the bias get smaller, the difference is still present:\n",
    "\n",
    "![image.png](attachment:image.png)\n",
    "\n",
    "See [our paper](https://galactica.org/static/paper.pdf) for more details."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "af51ab15",
   "metadata": {},
   "source": [
    "## Step-by-Step Reasoning\n",
    "\n",
    "Recent work (f.e., [Wei et al.](https://arxiv.org/abs/2201.11903), [Suzgun et al.](https://arxiv.org/abs/2210.09261)) have shown that chain-of-thought prompting can improve performance of large language models on complex reasoning tasks. In the NatureBook corpus used to train Galactica models we introduced a pair of special tokens - `<work>` and `</work>` to mark sections of fine-grained step-by-step reasoning. Explicit `<work>` token makes it easier to bias the generation into step-by-step reasoning. Compare the two queries:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a83d6d78",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: A bat and a ball cost $\\$1.10$ in total. The bat costs $\\$1.00$ more than the ball. How much does the ball cost?</p>\n",
       "<p>Answer: $\\$0.10$</p>\n",
       "<p>&lt;/work&gt;</p>\n",
       "<p>Ans: $\\$0.10$</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt = f\"Question: A bat and a ball cost $\\\\$1.10$ in total. The bat costs $\\\\$1.00$ more than the ball. How much does the ball cost?\\n\\nAnswer:\"\n",
    "display_markdown(model.generate(prompt, new_doc=True, max_new_tokens=250))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "692bf1fc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: A bat and a ball cost $\\$1.10$ in total. The bat costs $\\$1.00$ more than the ball. How much does the ball cost?</p>\n",
       "<p>&lt;work&gt;</p>\n",
       "<p>Let $x$ represent the ball's cost.</p>\n",
       "<p>The bat costs $x+\\$1.00$.</p>\n",
       "<p>The bat and the ball cost $x+(x+\\$1.00)=\\$1.10$.</p>\n",
       "<p>$2x+\\$1.00=\\$1.10$</p>\n",
       "<p>$2x=\\$0.10$</p>\n",
       "<p>$x=\\$0.05$</p>\n",
       "<p>The ball costs $\\$0.05$.</p>\n",
       "<p>&lt;/work&gt;</p>\n",
       "<p>Ans: The ball costs $\\$0.05$.</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt = f\"Question: A bat and a ball cost $\\\\$1.10$ in total. The bat costs $\\\\$1.00$ more than the ball. How much does the ball cost?\\n\\n<work>\"\n",
    "display_markdown(model.generate(prompt, new_doc=True, max_new_tokens=250))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b946c79b",
   "metadata": {},
   "source": [
    "### Python Evaluation\n",
    "\n",
    "Additionally, the `<work>` section can include a python code used to run external computations. For example,"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bc0c9775",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>What is the $7$-th harmonic number of the second order? Answer with a source code.</p>\n",
       "<p>&lt;work&gt;\n",
       "harmonic.py</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"nv\">ans</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"nv\">sum</span><span class=\"ss\">(</span><span class=\"mi\">1</span><span class=\"o\">/</span><span class=\"nv\">n</span><span class=\"o\">**</span><span class=\"mi\">2</span><span class=\"w\"> </span><span class=\"k\">for</span><span class=\"w\"> </span><span class=\"nv\">n</span><span class=\"w\"> </span><span class=\"nv\">in</span><span class=\"w\"> </span><span class=\"nv\">range</span><span class=\"ss\">(</span><span class=\"mi\">1</span>,<span class=\"w\"> </span><span class=\"mi\">7</span><span class=\"w\"> </span><span class=\"o\">+</span><span class=\"w\"> </span><span class=\"mi\">1</span><span class=\"ss\">))</span><span class=\"w\"></span>\n",
       "\n",
       "<span class=\"nv\">with</span><span class=\"w\"> </span><span class=\"nv\">open</span><span class=\"ss\">(</span><span class=\"s2\">&quot;output.txt&quot;</span>,<span class=\"w\"> </span><span class=\"s2\">&quot;w&quot;</span><span class=\"ss\">)</span><span class=\"w\"> </span><span class=\"nv\">as</span><span class=\"w\"> </span><span class=\"nv\">file</span>:<span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"nv\">file</span>.<span class=\"nv\">write</span><span class=\"ss\">(</span><span class=\"nv\">str</span><span class=\"ss\">(</span><span class=\"nv\">ans</span><span class=\"ss\">))</span><span class=\"w\"></span>\n",
       "</code></pre></div>\n",
       "\n",
       "<p>&lt;&lt;run: \"harmonic.py\"&gt;&gt;</p>\n",
       "<p>&lt;&lt;read: \"output.txt\"&gt;&gt;</p>\n",
       "<p>1.3852941429414294\n",
       "&lt;/work&gt;</p>\n",
       "<p>A: 1.3852941429414294</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_markdown(\n",
    "    model.generate(\n",
    "        \"What is the $7$-th harmonic number of the second order? Answer with a source code.\\n\\n<work>\",\n",
    "        max_new_tokens=300,\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1458bff2",
   "metadata": {},
   "source": [
    "While the numerical answer is incorrect, the generated code correctly implements the formula above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d47e8fdd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.511797052154195"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(1/n**2 for n in range(1, 7 + 1))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "023c2a4d",
   "metadata": {},
   "source": [
    "## Stored knowledge\n",
    "\n",
    "We can use generation to retrieve definitions, formulas, source code and more:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "85fd507e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# Corticosteroid\n",
      "\n",
      " Corticosteroids are a class of steroid hormones that are produced in the adrenal cortex of vertebrates. They are involved in a wide range of physiological processes, including metabolism, immune function, and stress response.[START_REF] Corticosteroids: Mechanisms of Action in Health and Disease, Ramamoorthy[END_REF]\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"# Corticosteroid\\n\", new_doc=True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "42bd5446",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "The $n$-th harmonic number of the second order is given by the formula:\n",
       "\n",
       "$$ H_{n}^{(2)}=\\sum_{k=1}^{n}\\frac{1}{k^{2}}. $$"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_latex(model.generate(\n",
    "    \"The \\\\(n\\\\)-th harmonic number of the second order is given by the formula:\\n\\n\\\\[\",\n",
    "    max_new_tokens=40,\n",
    "))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "12dc2ab3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The IUPAC name of cortisol is: 11β,17α,21-trihydroxypregn-4-ene-3,20-dione.\n",
      "\n",
      "## See also\n",
      "\n",
      "* Cortisone\n",
      "* Corticosterone\n",
      "* Hydrocortisone\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"The IUPAC name of cortisol is:\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "940675f0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "Mixing a kitchen salt with sulfuric acid results in the following reaction:\n",
       "\n",
       "$$ \\ce{NaCl}(aq)+\\ce{H2SO4}(aq)⟶\\ce{NaHSO4}(aq)+\\ce{HCl}(g) $$\n",
       "\n",
       "The hydrogen chloride gas is a strong acid and will react with any base that it comes into contact with.\n",
       "\n",
       "$$ \\ce"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_latex(model.generate(\"Mixing a kitchen salt with sulfuric acid results in the following reaction:\", max_new_tokens=80))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "10e0d13e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Use find to list all PNG files larger than 1 megabyte:\n",
      "\n",
      "```\n",
      "find . -name \"*.png\" -size +1M\n",
      "```\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"Use find to list all PNG files larger than 1 megabyte:\", max_new_tokens=25))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0d9d3804",
   "metadata": {},
   "source": [
    "## Composition\n",
    "\n",
    "Galactica models are able to mix & combine scientific modalities, stored knowledge and generalize to new tasks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "75758fbd",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: Translate the following python code:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"k\">def</span> <span class=\"nf\">cheapestProduct</span><span class=\"p\">(</span><span class=\"n\">products</span><span class=\"p\">:</span> <span class=\"n\">List</span><span class=\"p\">[</span><span class=\"n\">Product</span><span class=\"p\">])</span> <span class=\"o\">-&gt;</span> <span class=\"n\">Product</span><span class=\"p\">:</span>\n",
       "    <span class=\"k\">return</span> <span class=\"nb\">min</span><span class=\"p\">(</span><span class=\"n\">products</span><span class=\"p\">,</span> <span class=\"n\">key</span><span class=\"o\">=</span><span class=\"k\">lambda</span> <span class=\"n\">p</span><span class=\"p\">:</span> <span class=\"n\">p</span><span class=\"o\">.</span><span class=\"n\">price</span><span class=\"p\">)</span>\n",
       "</code></pre></div>\n",
       "\n",
       "<p>into C++.</p>\n",
       "<p>Answer:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"nt\">Product</span><span class=\"w\"> </span><span class=\"nt\">cheapestProduct</span><span class=\"o\">(</span><span class=\"nt\">std</span><span class=\"p\">::</span><span class=\"nd\">vector</span><span class=\"o\">&lt;</span><span class=\"nt\">Product</span><span class=\"o\">&gt;</span><span class=\"w\"> </span><span class=\"nt\">products</span><span class=\"o\">)</span><span class=\"w\"> </span><span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"err\">Product</span><span class=\"w\"> </span><span class=\"err\">min_product</span><span class=\"w\"> </span><span class=\"err\">=</span><span class=\"w\"> </span><span class=\"err\">products</span><span class=\"cp\">[</span><span class=\"mi\">0</span><span class=\"cp\">]</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"err\">for</span><span class=\"w\"> </span><span class=\"err\">(auto</span><span class=\"w\"> </span><span class=\"n\">product</span><span class=\"w\"> </span><span class=\"p\">:</span><span class=\"w\"> </span><span class=\"n\">products</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"err\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">        </span><span class=\"n\">if</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"n\">product</span><span class=\"o\">.</span><span class=\"n\">price</span><span class=\"w\"> </span><span class=\"o\">&lt;</span><span class=\"w\"> </span><span class=\"n\">min_product</span><span class=\"o\">.</span><span class=\"n\">price</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"err\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">            </span><span class=\"n\">min_product</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"n\">product</span><span class=\"p\">;</span><span class=\"w\"></span>\n",
       "<span class=\"w\">        </span><span class=\"p\">}</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"err\">}</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"nt\">return</span><span class=\"w\"> </span><span class=\"nt\">min_product</span><span class=\"o\">;</span><span class=\"w\"></span>\n",
       "<span class=\"err\">}</span><span class=\"w\"></span>\n",
       "</code></pre></div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_markdown(model.generate(\"\"\"Question: Translate the following python code:\n",
    "\n",
    "```python\n",
    "def cheapestProduct(products: List[Product]) -> Product:\n",
    "    return min(products, key=lambda p: p.price)\n",
    "```\n",
    "\n",
    "into C++.\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=150))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3c4a0c4d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: Translate the following math formula:</p>\n",
       "<p>$$\n",
       "  \\zeta(s) = \\sum_{n=1}^{\\infty} n^{-s}\n",
       "$$</p>\n",
       "<p>into plain English.</p>\n",
       "<p>Answer:</p>\n",
       "<p>The zeta function is the sum of the reciprocals of the positive integers raised to the $s$th power.</p>\n",
       "<p>&lt;/work&gt;</p>\n",
       "<p>Answer: $\\zeta(s) = \\sum_{n=1}^{\\infty} n^{-s}$</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_markdown(model.generate(\"\"\"Question: Translate the following math formula:\n",
    "\n",
    "\\\\[\n",
    "  \\\\zeta(s) = \\\\sum_{n=1}^{\\\\infty} n^{-s}\n",
    "\\\\]\n",
    "\n",
    "into plain English.\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a8ad61f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: Translate the following math formula:</p>\n",
       "<p>$$\n",
       "  \\zeta(s) = \\sum_{n=1}^{\\infty} n^{-s}\n",
       "$$</p>\n",
       "<p>into python code.</p>\n",
       "<p>Answer:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"nv\">def</span><span class=\"w\"> </span><span class=\"nv\">zeta</span><span class=\"ss\">(</span><span class=\"nv\">s</span><span class=\"ss\">)</span>:<span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"k\">return</span><span class=\"w\"> </span><span class=\"nv\">sum</span><span class=\"ss\">(</span>[<span class=\"nv\">n</span><span class=\"o\">**-</span><span class=\"nv\">s</span><span class=\"w\"> </span><span class=\"k\">for</span><span class=\"w\"> </span><span class=\"nv\">n</span><span class=\"w\"> </span><span class=\"nv\">in</span><span class=\"w\"> </span><span class=\"nv\">range</span><span class=\"ss\">(</span><span class=\"mi\">1</span>,<span class=\"w\"> </span><span class=\"mi\">100</span><span class=\"ss\">)</span>]<span class=\"ss\">)</span><span class=\"w\"></span>\n",
       "</code></pre></div>\n",
       "\n",
       "<p>The zeta function is a sum of an infinite number of terms. In order to compute it, we need to approximate it with a finite sum. The function above computes the sum of the first 100 terms.</p>\n",
       "<p>The zeta function is a very important function in mathematics. It is used to compute</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_markdown(model.generate(\"\"\"Question: Translate the following math formula:\n",
    "\n",
    "\\\\[\n",
    "  \\\\zeta(s) = \\\\sum_{n=1}^{\\\\infty} n^{-s}\n",
    "\\\\]\n",
    "\n",
    "into python code.\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b67bbc08",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: Translate the following math formula:</p>\n",
       "<p>$$\n",
       "  f(x) = \\int_0^x \\frac{\\cos(2\\cdot t)}{\\sqrt{2\\pi}} dt.\n",
       "$$</p>\n",
       "<p>into python code using sympy package.</p>\n",
       "<p>Answer:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"kn\">from</span> <span class=\"nn\">sympy</span> <span class=\"kn\">import</span> <span class=\"o\">*</span>\n",
       "<span class=\"n\">f</span> <span class=\"o\">=</span> <span class=\"n\">Integral</span><span class=\"p\">(</span><span class=\"n\">cos</span><span class=\"p\">(</span><span class=\"mi\">2</span><span class=\"o\">*</span><span class=\"n\">t</span><span class=\"p\">)</span><span class=\"o\">/</span><span class=\"n\">sqrt</span><span class=\"p\">(</span><span class=\"mi\">2</span><span class=\"o\">*</span><span class=\"n\">pi</span><span class=\"p\">),</span> <span class=\"p\">(</span><span class=\"n\">t</span><span class=\"p\">,</span> <span class=\"mi\">0</span><span class=\"p\">,</span> <span class=\"n\">x</span><span class=\"p\">))</span>\n",
       "</code></pre></div>\n",
       "\n",
       "<h2></h2>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_markdown(model.generate(\"\"\"Question: Translate the following math formula:\n",
    "\n",
    "\\\\[\n",
    "  f(x) = \\\\int_0^x \\\\frac{\\\\cos(2\\cdot t)}{\\\\sqrt{2\\\\pi}} dt.\n",
    "\\\\]\n",
    "\n",
    "into python code using sympy package.\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=45))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2e9b26f6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: What is the expected value of a random variable uniformly distributed over the interval $[a^2, b+c]$?</p>\n",
       "<p>Answer: $\\frac{b+c+a^2}{2}$</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output = model.generate(\"\"\"Question: What is the expected value of a random variable uniformly distributed over the interval \\\\([a^2, b+c]\\\\)?\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=20)\n",
    "display_markdown(output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c07a1ad1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: What is the expected value of a random variable uniformly distributed over the interval $[a^2, b+c]$?</p>\n",
       "<p>Answer: $\\frac{b+c+a^2}{2}$</p>\n",
       "<p>Question: Rewrite the formula above in Mathematica.</p>\n",
       "<p>Answer: $\\text{Expectation}[x,x\\sim\\text{UniformDistribution}[a^2,b+c]]$</p>\n",
       "<p>Question: What is the expected value of a random variable uniformly distributed over the interval $[a, b]</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_markdown(\n",
    "    model.generate(\n",
    "        f\"{output.rstrip()}\\n\\nQuestion: Rewrite the formula above in Mathematica.\\n\\nAnswer:\"\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f39718b4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: Translate the following python code:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"kn\">import</span> <span class=\"nn\">requests</span>\n",
       "<span class=\"kn\">import</span> <span class=\"nn\">re</span>\n",
       "\n",
       "<span class=\"k\">def</span> <span class=\"nf\">get_datasets</span><span class=\"p\">():</span>\n",
       "    <span class=\"n\">req</span> <span class=\"o\">=</span> <span class=\"n\">requests</span><span class=\"o\">.</span><span class=\"n\">get</span><span class=\"p\">(</span><span class=\"s1\">&#39;https://paperswithcode.com/datasets&#39;</span><span class=\"p\">)</span>\n",
       "    <span class=\"k\">if</span> <span class=\"n\">req</span><span class=\"o\">.</span><span class=\"n\">ok</span><span class=\"p\">:</span>\n",
       "        <span class=\"n\">match</span> <span class=\"o\">=</span> <span class=\"n\">re</span><span class=\"o\">.</span><span class=\"n\">search</span><span class=\"p\">(</span><span class=\"sa\">r</span><span class=\"s1\">&#39;(\\d+) dataset results&#39;</span><span class=\"p\">,</span> <span class=\"n\">req</span><span class=\"o\">.</span><span class=\"n\">text</span><span class=\"p\">)</span>\n",
       "        <span class=\"k\">return</span> <span class=\"nb\">int</span><span class=\"p\">(</span><span class=\"n\">match</span><span class=\"o\">.</span><span class=\"n\">group</span><span class=\"p\">(</span><span class=\"mi\">1</span><span class=\"p\">))</span> <span class=\"k\">if</span> <span class=\"n\">match</span> <span class=\"k\">else</span> <span class=\"kc\">None</span>\n",
       "    <span class=\"k\">return</span> <span class=\"kc\">None</span>\n",
       "</code></pre></div>\n",
       "\n",
       "<p>into Javascript.</p>\n",
       "<p>Answer:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"kd\">const</span><span class=\"w\"> </span><span class=\"nx\">getDatasets</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"p\">()</span><span class=\"w\"> </span><span class=\"p\">=&gt;</span><span class=\"w\"> </span><span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"kd\">const</span><span class=\"w\"> </span><span class=\"nx\">req</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"nx\">fetch</span><span class=\"p\">(</span><span class=\"s1\">&#39;https://paperswithcode.com/datasets&#39;</span><span class=\"p\">)</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"k\">if</span><span class=\"w\"> </span><span class=\"p\">(</span><span class=\"nx\">req</span><span class=\"p\">.</span><span class=\"nx\">ok</span><span class=\"p\">)</span><span class=\"w\"> </span><span class=\"p\">{</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"kd\">const</span><span class=\"w\"> </span><span class=\"nx\">match</span><span class=\"w\"> </span><span class=\"o\">=</span><span class=\"w\"> </span><span class=\"sr\">/(\\d+) dataset results/</span><span class=\"p\">.</span><span class=\"nx\">exec</span><span class=\"p\">(</span><span class=\"nx\">req</span><span class=\"p\">.</span><span class=\"nx\">text</span><span class=\"p\">)</span><span class=\"w\"></span>\n",
       "<span class=\"w\">    </span><span class=\"k\">return</span><span class=\"w\"> </span><span class=\"nx\">match</span><span class=\"w\"> </span><span class=\"o\">?</span><span class=\"w\"> </span><span class=\"nb\">parseInt</span><span class=\"p\">(</span><span class=\"nx\">match</span><span class=\"p\">[</span><span class=\"mf\">1</span><span class=\"p\">])</span><span class=\"w\"> </span><span class=\"o\">:</span><span class=\"w\"> </span><span class=\"kc\">null</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"p\">}</span><span class=\"w\"></span>\n",
       "<span class=\"w\">  </span><span class=\"k\">return</span><span class=\"w\"> </span><span class=\"kc\">null</span><span class=\"w\"></span>\n",
       "<span class=\"p\">}</span><span class=\"w\"></span>\n",
       "</code></pre></div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_markdown(model.generate(\"\"\"Question: Translate the following python code:\n",
    "\n",
    "```python\n",
    "import requests\n",
    "import re\n",
    "\n",
    "def get_datasets():\n",
    "    req = requests.get('https://paperswithcode.com/datasets')\n",
    "    if req.ok:\n",
    "        match = re.search(r'(\\\\d+) dataset results', req.text)\n",
    "        return int(match.group(1)) if match else None\n",
    "    return None\n",
    "```\n",
    "\n",
    "into Javascript.\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=150))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "628a3eb5",
   "metadata": {},
   "source": [
    "> **Please note that the generations are not guaranteed to be correct.** In the example above, the model correctly translated the Python regular expression to a Javascript one, parsing an integer value from a string or even matched the common Javascript casing style (`get_datasets` to `getDatasets`). However, the `req` is not handled correctly as a `Promise`. `Request.text` is a function returning a `Promise` as well.\n",
    "See Pitfalls & Failure Examples Section for more details."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "24cd0811",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Use math facts to simplify the following python code:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"k\">def</span> <span class=\"nf\">calc_sum</span><span class=\"p\">(</span><span class=\"n\">n</span><span class=\"p\">):</span>\n",
       "    <span class=\"n\">i</span> <span class=\"o\">=</span> <span class=\"mi\">0</span>\n",
       "    <span class=\"n\">s</span> <span class=\"o\">=</span> <span class=\"mi\">0</span>\n",
       "    <span class=\"k\">while</span> <span class=\"n\">i</span> <span class=\"o\">&lt;=</span> <span class=\"n\">n</span><span class=\"p\">:</span>\n",
       "        <span class=\"n\">s</span> <span class=\"o\">+=</span> <span class=\"n\">i</span>\n",
       "        <span class=\"n\">i</span> <span class=\"o\">+=</span> <span class=\"mi\">1</span>\n",
       "    <span class=\"k\">return</span> <span class=\"n\">s</span>\n",
       "</code></pre></div>\n",
       "\n",
       "<p>&lt;work&gt;</p>\n",
       "<p>Let's look at the body of the <code>while</code> loop:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code>s += i\n",
       "i += 1\n",
       "</code></pre></div>\n",
       "\n",
       "<p>There are two math operations here: <code>+=</code> and <code>+</code>.</p>\n",
       "<p>The <code>+=</code> operator is an assignment operation. It assigns the value of <code>i</code> to <code>s</code>.</p>\n",
       "<p>The <code>+</code> operator is a math operation that adds <code>i</code> to <code>s</code>.</p>\n",
       "<p>The next step is to figure out the order of operations.</p>\n",
       "<p>Assignment operations have the same order of operations as the code that follows.</p>\n",
       "<p>Math operations have the same order of operations as the order of operations in standard math.</p>\n",
       "<p>The order of operations in standard math is:</p>\n",
       "<ul>\n",
       "<li>Exponentiation</li>\n",
       "<li>Multiplication and Division</li>\n",
       "<li>Addition and Subtraction</li>\n",
       "</ul>\n",
       "<p>So, the order of operations in the code is:</p>\n",
       "<ul>\n",
       "<li>Exponentiation</li>\n",
       "<li>Multiplication and Division</li>\n",
       "<li>Addition and Subtraction</li>\n",
       "<li>Assignment</li>\n",
       "</ul>\n",
       "<p>The next step is to figure out what the code is doing:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code>s += i\n",
       "i += 1\n",
       "</code></pre></div>\n",
       "\n",
       "<p>The code is assigning the value of <code>i</code> to <code>s</code> and then adding <code>1</code> to <code>i</code>.</p>\n",
       "<p>The first line can be rewritten as <code>s = s + i</code>.</p>\n",
       "<p>The second line can be rewritten as <code>i = i + 1</code>.</p>\n",
       "<p>The code can be rewritten as:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code>s = s + i\n",
       "i = i + 1\n",
       "</code></pre></div>\n",
       "\n",
       "<p>The next step is to figure out the value of <code>i</code> and <code>s</code> after the loop.</p>\n",
       "<p>The value of <code>i</code> after the loop is <code>n + 1</code>.</p>\n",
       "<p>The value of <code>s</code> after the loop is <code>(n + 1) * (n + 1) / 2</code>.</p>\n",
       "<p>The code can be rewritten as:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code>s = (n + 1) * (n + 1) / 2\n",
       "i = n + 1\n",
       "</code></pre></div>\n",
       "\n",
       "<p>The next step is to figure out the expression that the code is calculating:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code>s = (n + 1) * (n + 1) / 2\n",
       "i = n + 1\n",
       "</code></pre></div>\n",
       "\n",
       "<p>The code is calculating the sum of the numbers from <code>0</code> to <code>n</code>.</p>\n",
       "<p>&lt;/work&gt;</p>\n",
       "<p>Answer:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code>s = (n + 1) * (n + 1) / 2\n",
       "i = n + 1\n",
       "</code></pre></div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_markdown(model.generate(\"\"\"Use math facts to simplify the following python code:\n",
    "\n",
    "```python\n",
    "def calc_sum(n):\n",
    "    i = 0\n",
    "    s = 0\n",
    "    while i <= n:\n",
    "        s += i\n",
    "        i += 1\n",
    "    return s\n",
    "```\n",
    "\n",
    "<work>\"\"\", max_new_tokens=700))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "05cf15fa",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "We can see in this example that the initial error for the value of `s` after the loop is propagated to\n",
    "the final answer. There's an off-by-one error and the correct value should be:\n",
    "\n",
    "```python\n",
    "s = (n + 1) * n / 2\n",
    "```\n",
    "\n",
    "Also, the model output has some incorrect statements, such as:\n",
    "> The code is assigning the value of `i` to `s`"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32f10394",
   "metadata": {},
   "source": [
    "### Few-Shot Prompts\n",
    "\n",
    "We can write a few-shot prompt to try to bias the generation into desired format:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7d9dc0c3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: does \"kayak\" read the same backward as forward? Answer with code.</p>\n",
       "<p>Code:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"k\">def</span> <span class=\"nf\">is_palindrome</span><span class=\"p\">(</span><span class=\"n\">s</span><span class=\"p\">):</span>\n",
       "    <span class=\"k\">return</span> <span class=\"n\">s</span> <span class=\"o\">==</span> <span class=\"n\">s</span><span class=\"p\">[::</span><span class=\"o\">-</span><span class=\"mi\">1</span><span class=\"p\">]</span>\n",
       "</code></pre></div>\n",
       "\n",
       "<p>Answer: <code>is_palindrome(\"kayak\")</code>.</p>\n",
       "<p>Question: An $i$-th Peanut Butter number is given by the formula $pb_i = \\prod_{k=2}^{i} \\frac{1}{1-1/k}$. An $i$-th Jelly number is given by $J_i = \\sum_{k=2}^{i} pb_k$. What is the 6-th Jelly number? Answer with code.</p>\n",
       "<p>Code:</p>\n",
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"k\">def</span> <span class=\"nf\">peanut_butter</span><span class=\"p\">(</span><span class=\"n\">i</span><span class=\"p\">):</span>\n",
       "    <span class=\"k\">return</span> <span class=\"n\">reduce</span><span class=\"p\">(</span><span class=\"k\">lambda</span> <span class=\"n\">x</span><span class=\"p\">,</span> <span class=\"n\">y</span><span class=\"p\">:</span> <span class=\"n\">x</span> <span class=\"o\">*</span> <span class=\"n\">y</span><span class=\"p\">,</span> <span class=\"nb\">map</span><span class=\"p\">(</span><span class=\"k\">lambda</span> <span class=\"n\">k</span><span class=\"p\">:</span> <span class=\"mi\">1</span> <span class=\"o\">/</span> <span class=\"p\">(</span><span class=\"mi\">1</span> <span class=\"o\">-</span> <span class=\"mi\">1</span> <span class=\"o\">/</span> <span class=\"n\">k</span><span class=\"p\">),</span> <span class=\"nb\">range</span><span class=\"p\">(</span><span class=\"mi\">2</span><span class=\"p\">,</span> <span class=\"n\">i</span> <span class=\"o\">+</span> <span class=\"mi\">1</span><span class=\"p\">)))</span>\n",
       "\n",
       "<span class=\"k\">def</span> <span class=\"nf\">jelly</span><span class=\"p\">(</span><span class=\"n\">i</span><span class=\"p\">):</span>\n",
       "    <span class=\"k\">return</span> <span class=\"n\">reduce</span><span class=\"p\">(</span><span class=\"k\">lambda</span> <span class=\"n\">x</span><span class=\"p\">,</span> <span class=\"n\">y</span><span class=\"p\">:</span> <span class=\"n\">x</span> <span class=\"o\">+</span> <span class=\"n\">y</span><span class=\"p\">,</span> <span class=\"nb\">map</span><span class=\"p\">(</span><span class=\"k\">lambda</span> <span class=\"n\">k</span><span class=\"p\">:</span> <span class=\"n\">peanut_butter</span><span class=\"p\">(</span><span class=\"n\">k</span><span class=\"p\">),</span> <span class=\"nb\">range</span><span class=\"p\">(</span><span class=\"mi\">2</span><span class=\"p\">,</span> <span class=\"n\">i</span> <span class=\"o\">+</span> <span class=\"mi\">1</span><span class=\"p\">)))</span>\n",
       "</code></pre></div>\n",
       "\n",
       "<p>Answer: <code>jelly(6)</code>.</p>\n",
       "<p>Question: What is the largest prime factor of $2^{2017}-</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_markdown(model.generate(\"\"\"Question: does \"kayak\" read the same backward as forward? Answer with code.\n",
    "\n",
    "Code:\n",
    "\n",
    "```python\n",
    "def is_palindrome(s):\n",
    "    return s == s[::-1]\n",
    "```\n",
    "\n",
    "Answer: `is_palindrome(\"kayak\")`.\n",
    "\n",
    "Question: An $i$-th Peanut Butter number is given by the formula $pb_i = \\\\prod_{k=2}^{i} \\\\frac{1}{1-1/k}$. An $i$-th Jelly number is given by $J_i = \\\\sum_{k=2}^{i} pb_k$. What is the 6-th Jelly number? Answer with code.\n",
    "\"\"\", max_new_tokens=150))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2921bfa4",
   "metadata": {},
   "source": [
    "# Text Generation & Sampling\n",
    "\n",
    "The `galai` library uses HuggingFace [transformers](https://huggingface.co/docs/transformers/index) to run inference, download checkpoints and efficiently load models. As a result we have an easy access to the comprehensive collection of generation algorithms. In this Section we present how to use the most common ones, supported by `galai`. Additionally we show how to fallback to using `transformers` directly to access additional options."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2d9c4acc",
   "metadata": {},
   "source": [
    "### Greedy Decoding\n",
    "\n",
    "This is the standard algorithm used by `Model.generate`. Using the prompt and already generated tokens, the model computes a probability distribution of the next token over all tokens. The token with the highest score is appended to the generated text and the process is repeated."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5670fe5a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: what are the 10 most common text generation algorithms?</p>\n",
       "<p>Answer:</p>\n",
       "<ul>\n",
       "<li>Beam search</li>\n",
       "<li>Sampling</li>\n",
       "<li>Greedy search</li>\n",
       "<li>Nucleus sampling</li>\n",
       "<li>Diverse beam search</li>\n",
       "<li>Top-k sampling</li>\n",
       "<li>Top-p sampling</li>\n",
       "<li>Repetition penalty</li>\n",
       "<li>Max length penalty</li>\n",
       "<li>Length normalization</li>\n",
       "</ul>\n",
       "<p>Question: what are some categories for text generation algorithms?</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_markdown(model.generate(\"Question: what are the 10 most common text generation algorithms?\\n\\nAnswer:\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27cdf8dc",
   "metadata": {},
   "source": [
    "### Beam Search\n",
    "\n",
    "In Beam Search, the model computes a probability distribution of the next token over all tokens for each of the `num_beams` generated sequences. The `num_beams` sequences with the highest probability are kept and the process is repeated."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ac01597b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"s s-Atom\">def</span> <span class=\"nf\">is_palindrome</span><span class=\"p\">(</span><span class=\"s s-Atom\">word</span><span class=\"p\">)</span><span class=\"o\">:</span>\n",
       "    <span class=\"s2\">&quot;&quot;&quot;</span>\n",
       "<span class=\"s2\">    Check if a word is a palindrome.</span>\n",
       "\n",
       "<span class=\"s2\">    Parameters</span>\n",
       "<span class=\"s2\">    ----------</span>\n",
       "<span class=\"s2\">    word : str</span>\n",
       "<span class=\"s2\">        The word to check.</span>\n",
       "\n",
       "<span class=\"s2\">    Returns</span>\n",
       "<span class=\"s2\">    -------</span>\n",
       "<span class=\"s2\">    bool</span>\n",
       "<span class=\"s2\">        True if the word is a palindrome, False otherwise.</span>\n",
       "\n",
       "<span class=\"s2\">    Examples</span>\n",
       "<span class=\"s2\">    --------</span>\n",
       "<span class=\"s2\">    &gt;&gt;&gt; is_palindrome(&quot;</span><span class=\"s s-Atom\">palindrome</span><span class=\"s2\">&quot;)</span>\n",
       "<span class=\"s2\">    True</span>\n",
       "<span class=\"s2\">    &gt;&gt;&gt; is_palindrome(&quot;</span><span class=\"s s-Atom\">nonpalindrome</span><span class=\"s2\">&quot;)</span>\n",
       "<span class=\"s2\">    False</span>\n",
       "<span class=\"s2\">    &quot;&quot;&quot;</span>\n",
       "    <span class=\"s s-Atom\">return</span> <span class=\"s s-Atom\">word</span> <span class=\"o\">==</span> <span class=\"s s-Atom\">word</span><span class=\"p\">[</span><span class=\"o\">::-</span><span class=\"mi\">1</span><span class=\"p\">]</span>\n",
       "\n",
       "\n",
       "<span class=\"s s-Atom\">def</span> <span class=\"nf\">is_palindromic_word</span><span class=\"p\">(</span><span class=\"s s-Atom\">word</span><span class=\"p\">)</span><span class=\"o\">:</span>\n",
       "    <span class=\"s2\">&quot;&quot;</span><span class=\"err\">&quot;</span>\n",
       "    <span class=\"nv\">Check</span> <span class=\"s s-Atom\">if</span> <span class=\"s s-Atom\">a</span> <span class=\"s s-Atom\">word</span> <span class=\"o\">is</span> <span class=\"s s-Atom\">a</span> <span class=\"s s-Atom\">palindromic</span> <span class=\"s s-Atom\">word</span>\n",
       "</code></pre></div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt = \"def is_palindrome\"\n",
    "# greedy search\n",
    "code = model.generate(prompt, max_new_tokens=150)\n",
    "display_markdown(f\"```\\n{code}\\n```\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "097833c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"codehilite\"><pre><span></span><code><span class=\"s s-Atom\">def</span> <span class=\"nf\">is_palindrome</span><span class=\"p\">(</span><span class=\"s s-Atom\">word</span><span class=\"p\">:</span> <span class=\"s s-Atom\">str</span><span class=\"p\">)</span> <span class=\"s s-Atom\">-&gt;</span> <span class=\"s s-Atom\">bool</span><span class=\"p\">:</span>\n",
       "    <span class=\"s2\">&quot;&quot;&quot;Check if a word is a palindrome.</span>\n",
       "\n",
       "<span class=\"s2\">    Args:</span>\n",
       "<span class=\"s2\">        word (str): The word to check.</span>\n",
       "\n",
       "<span class=\"s2\">    Returns:</span>\n",
       "<span class=\"s2\">        bool: True if the word is a palindrome, False otherwise.</span>\n",
       "<span class=\"s2\">    &quot;&quot;&quot;</span>\n",
       "    <span class=\"s s-Atom\">return</span> <span class=\"s s-Atom\">word</span> <span class=\"o\">==</span> <span class=\"s s-Atom\">word</span><span class=\"p\">[</span><span class=\"o\">::-</span><span class=\"mi\">1</span><span class=\"p\">]</span>\n",
       "\n",
       "\n",
       "<span class=\"s s-Atom\">def</span> <span class=\"nf\">is_palindrome_strict</span><span class=\"p\">(</span><span class=\"s s-Atom\">word</span><span class=\"p\">:</span> <span class=\"s s-Atom\">str</span><span class=\"p\">)</span> <span class=\"s s-Atom\">-&gt;</span> <span class=\"s s-Atom\">bool</span><span class=\"p\">:</span>\n",
       "    <span class=\"s2\">&quot;&quot;</span><span class=\"err\">&quot;</span><span class=\"nv\">Check</span> <span class=\"s s-Atom\">if</span> <span class=\"s s-Atom\">a</span> <span class=\"s s-Atom\">word</span> <span class=\"o\">is</span> <span class=\"s s-Atom\">a</span> <span class=\"s s-Atom\">strict</span> <span class=\"s s-Atom\">palindrome</span><span class=\"p\">.</span>\n",
       "\n",
       "    <span class=\"nv\">Args</span><span class=\"o\">:</span>\n",
       "        <span class=\"nf\">word</span> <span class=\"p\">(</span><span class=\"s s-Atom\">str</span><span class=\"p\">)</span><span class=\"o\">:</span> <span class=\"nv\">The</span> <span class=\"s s-Atom\">word</span> <span class=\"s s-Atom\">to</span> <span class=\"s s-Atom\">check</span><span class=\"p\">.</span>\n",
       "\n",
       "    <span class=\"nv\">Returns</span><span class=\"o\">:</span>\n",
       "        <span class=\"s s-Atom\">bool</span><span class=\"p\">:</span> <span class=\"nv\">True</span> <span class=\"s s-Atom\">if</span> <span class=\"s s-Atom\">the</span> <span class=\"s s-Atom\">word</span> <span class=\"o\">is</span> <span class=\"s s-Atom\">a</span> <span class=\"s s-Atom\">strict</span> <span class=\"s s-Atom\">palindrome</span><span class=\"p\">,</span> <span class=\"nv\">False</span> <span class=\"s s-Atom\">otherwise</span><span class=\"p\">.</span>\n",
       "</code></pre></div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# beam search\n",
    "code = model.generate(prompt, num_beams=5, max_new_tokens=150)\n",
    "display_markdown(f\"```\\n{code}\\n```\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d2ea8a38",
   "metadata": {},
   "source": [
    "You can return up to `num_beams` sequences by specifying `num_return_sequences`."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2fb2dab1",
   "metadata": {},
   "source": [
    "Beam search is slower and requires more memory compared to the Greedy Decoding. The increase in memory consumption is proportional to the number of beams used."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c6f78ad1",
   "metadata": {},
   "source": [
    "### Contrastive Search\n",
    "\n",
    "The contrastive search ([Su et al.](https://arxiv.org/abs/2202.06417), [Su et al.](https://arxiv.org/abs/2210.14140)) algorithm is a novel generation method that aims to produce more natural texts by penalizing repetitions. We can use `transformers` implementation (see more at https://huggingface.co/blog/introducing-csearch) by specifying `penalty_alpha` and `top_k`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3ccaf80e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Title: A Literature Review on Alzheimer's Disease\n",
      "\n",
      "# Abstract\n",
      "\n",
      "Alzheimer's disease (AD) is a neurodegenerative disease that affects millions of people worldwide. The number of people with AD is expected to increase as the population ages. Currently, there is no cure for AD, and the treatments available only slow the progression of the disease. This literature review aims to provide an overview of the pathophysiology of AD, the current treatments available, and the role of exercise in the management of AD.\n",
      "\n",
      "# 1. Introduction\n",
      "\n",
      "\n",
      "Alzheimer's disease (AD) is a neurodegenerative disease that affects millions of people worldwide. The number of people with AD is expected to increase as the population ages []. Currently, there is no cure for AD, and the treatments available only slow the progression of the disease.\n",
      "\n",
      "# 2. Pathophysiology\n",
      "\n",
      "AD is characterized by the presence of amyloid plaques and neurofibrillary tangles in the brain. Amyloid plaques are formed by the accumulation of amyloid-beta (Aβ) peptides, which are produced by the cleavage of amyloid precursor protein (APP) by β-secretase and γ-secretase [[START_REF] A systemic view of Alzheimer disease – insights from amyloid-beta metabolism beyond the brain, Wang[END_REF]]. Neurofibrillary tangles are formed by the aggregation of hyperphosphorylated tau protein, which is involved in the stabilization of microtubules [[START_REF] Tau in Alzheimer disease and related tauopathies., Iqbal<s>abrahams」aja[END_REF]].\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    model.generate(\n",
    "        \"Title: A Literature Review on Alzheimer's Disease\\n\\n# Abstract\\n\",\n",
    "        top_k=4, penalty_alpha=0.6, max_new_tokens=300\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35d92dce",
   "metadata": {},
   "source": [
    "---\n",
    "## Sampling\n",
    "\n",
    "Instead of selecting tokens with the highest scores we can use the scores to model a probability distribution to sample the tokens from.\n",
    "\n",
    "### Nucleus Sampling\n",
    "\n",
    "In Nucleus sampling (see [Holtzman et al.](https://arxiv.org/abs/1904.09751)) the tokens to sample from are limited to most likely tokens which total probability does not exceed `top_p` parameter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ff2c7bc0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " # Image Denoising\n",
      "\n",
      "Image denoising is the process of removing noise from an image, and it is an important task in the field of image processing. Image denoising is a classic ill-posed problem, and its purpose is to reconstruct the original image from the degraded image.\n",
      "\n",
      "We test our method on the benchmark\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\" # Image Denoising\", top_p=0.7))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e6d5f0aa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " # Image Denoising\n",
      "\n",
      "Image denoising is a well-known inverse problem in image processing and computer vision. A lot of works have been done to tackle this problem. The key of the problem is to recover the clean image x from the noisy image y = x + v. In the past decade, there\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\" # Image Denoising\", top_p=0.7))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "142951d0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " # Image Denoising\n",
      "\n",
      "The following section is dedicated to the application of our model to the denoising of images corrupted by additive white Gaussian noise. The task of image denoising consists in removing noise from a given noisy image, where the noise is assumed to be white Gaussian with known standard deviation. In this setting, the forward\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\" # Image Denoising\", top_p=0.7))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9a0e2c73",
   "metadata": {},
   "source": [
    "---\n",
    "With `top_p=1.0` all tokens are included and we get standard sampling."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d633a99d",
   "metadata": {},
   "source": [
    "#### Top-K Sampling\n",
    "\n",
    "In top-k sampling the tokens to sample from are limit to `top_k` most likely tokens."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bac79d01",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " # Image Denoising\n",
      "\n",
      "The task of image denoising is to remove the unwanted signal corruptions from the image. There is a rich body of literature [START_REF] Image restoration: total variation, wavelet frames, and beyond, Cai[END_REF][START_REF] A Review of Image Denoising Algorithms, with a New One, Buades[END_REF][START_REF]\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\" # Image Denoising\", top_k=10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9e4ca7dc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " # Image Denoising\n",
      "\n",
      "In the following section, we apply the proposed method for image denoising and compare with the recent state-of-the-art. The noisy image y, is modeled as\n",
      "\n",
      "y = x + n\n",
      "where x is the original noise free image and n is the additive white\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\" # Image Denoising\", top_k=10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b52efb36",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " # Image Denoising #\n",
      "###############################\n",
      "## # Image Denoising #\n",
      "###############################\n",
      "def denoise_tv_chambolle(noisy_image,\n",
      "                         weight_decay = 0.3,\n",
      "                         weight_gradients = 0.\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\" # Image Denoising\", top_k=10))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "214e0ded",
   "metadata": {},
   "source": [
    "---\n",
    "Both `top_p` and `top_k` can be used at the same time."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1640facd",
   "metadata": {},
   "source": [
    "### Using `transformers` Directly\n",
    "\n",
    "You can generate text with Galactica models directly using HuggingFace `transformers` library. One option is to use the model and tokenizer from the `galai.Model`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "643bc9d1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([1, 50000])\n"
     ]
    }
   ],
   "source": [
    "def transformers_generate(model, prompt, new_doc=False, **options):    \n",
    "    tokens = model._tokenize([prompt], new_doc=new_doc)\n",
    "    out = model.model.generate(\n",
    "        tokens,\n",
    "        **options\n",
    "    )\n",
    "    return out\n",
    "\n",
    "out = transformers_generate(\n",
    "    model,\n",
    "    \"In this paper, we study\",\n",
    "    max_new_tokens=40,\n",
    "    return_dict_in_generate=True,\n",
    "    output_scores=True\n",
    ")\n",
    "print(out.scores[0].shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cabd8bc0",
   "metadata": {},
   "source": [
    "This approach makes sure that the tokenization is done properly: the end-of-document token correctly handles padding and custom sequences are split.\n",
    "\n",
    "You can also use Galactica models soley using `transformers`, for example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "66b9a0bf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The Transformer architecture [START_REF] Attention is All you Need, Vaswani[END_REF] is a popular choice for sequence-to-sequence models\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "from transformers import AutoTokenizer, OPTForCausalLM\n",
    "\n",
    "transformers_tokenizer = AutoTokenizer.from_pretrained(\"facebook/galactica-1.3b\")\n",
    "transformers_model = OPTForCausalLM.from_pretrained(\"facebook/galactica-1.3b\", torch_dtype=torch.float16, device_map=\"auto\")\n",
    "\n",
    "input_text = \"The Transformer architecture [START_REF]\"\n",
    "input_ids = transformers_tokenizer(input_text, return_tensors=\"pt\").input_ids.to(\"cuda\")\n",
    "\n",
    "outputs = transformers_model.generate(input_ids, max_new_tokens=20)\n",
    "print(transformers_tokenizer.decode(outputs[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2ad40d6c",
   "metadata": {},
   "source": [
    "## Tokenization\n",
    "\n",
    "All Galactica models share the same vocabulary of 50000 tokens. The vocabulary was trained on 2% of our training corpus using Byte-Pair Encoding (BPE) tokenization.\n",
    "\n",
    "### Special Tokens\n",
    "\n",
    "Some of the tokens (f.e., the already mentioned `[START_REF]` or `<work>`) are special control tokens that can be used to steer model generation towards a specific type of content.\n",
    "\n",
    "\n",
    "`<unk>` - reserved.\n",
    "\n",
    "`<s>` - reserved.\n",
    "\n",
    "`</s>` - end-of-document token used to split documents during trainig. Prepending this token to prompt (see `new_doc` parameter in `Model.generate`) biases a model into generating a new document.\n",
    "\n",
    "`<pad>` - a standard padding token to align sequences in a batch.\n",
    "\n",
    "`[START_REF]` and `[END_REF]` - markers denoting a reference to a paper. Each paper is represented as `Title, First author name`. F.e., `[START_REF] Backpropagation Applied to Handwritten Zip Code Recognition, LeCun[END_REF]`.\n",
    "\n",
    "`[IMAGE]` - a placeholder for an image removed from a text.\n",
    "\n",
    "`<fragments>` and `</fragments>` - markers denoting fragments in FragmentedGlass dataset.\n",
    "\n",
    "`<work>` and `</work>` - markers denoting step-by-step reasoning (see Step-by-Step Reasoning Section).\n",
    "\n",
    "`[START_SUP]`, `[END_SUP]`, `[START_SUB]` and `[END_SUB]` - markers used to protect superscript and subscript digits from NFKC normaliziation. Our tokenizer uses the standard NFKC rules, which means that `x²⁵` would be tokenized in the same way as `x25`. To prevent this, we encode `x²⁵` as `x[START_SUP]25[END_SUP]`.\n",
    "\n",
    "`[START_DNA]`, `[END_DNA]`, `[START_AMINO]`, `[END_AMINO]`, `[START_SMILES]`, `[END_SMILES]`, `[START_I_SMILES]` and `[END_I_SMILES]` - markers denoting special sequences, respectively: nucleic acids sequences, amino acids sequeqnces, canonical simplified molecular-input line-entry system (SMILES) strings and isometric SMILES strings. Besides marking a sequence of a given type, these tokens force a special tokenization mode in which each character is represented as a single token. F.e., `GATTACA` is tokenized as `G|ATT|ACA`, while `[START_DNA]GATTACA[END_DNA]` is tokenized as `[START_DNA]|G|A|T|T|A|C|A|[END_DNA]`. Note that for this to work you need to transform your prompt with `galai.utils.escape_custom_split_sequence`. All standard text generation functions of `galai.model.Model` do this automatically.\n",
    "\n",
    "The `galai` library takes care of handling of the special tokens. If you are using `tokenizers` directly then most likely you want to keep the special tokens in the output for further processing. Set `skip_special_tokens=False` in `tokenizers.Tokenizer.decode`.\n",
    "\n",
    "### Decoupling of Tokens\n",
    "\n",
    "The BPE training algorithm creates vocabulary based on frequncies of subwords in the training corpus, with more frequent subwords being represented with fewer number of tokens. This means that visually similar subwords may end up having totally different token representations. For example, in the GPT-2 tokenizer (trained before year 2020) each of the numbers `{2000, 2001, ..., 2020}` is encoded with a unique token, and all of the numbers `{2021, 2022, ..., 2030}` are represented as two tokens: `20|21`, `20|22`, etc. Training on a corpus with math, TeX formulas and source code it can happen that a single token encodes multiple independent functions. F.e., `\\(-` can end up being a single token making prompting more difficult and the model less robust to changes in spaces.\n",
    "\n",
    "To prevent this issue we implemented custom splitting rules, presented in the example below. For performance reasons we keep a leading space (i.e., ` text` can be a single token)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "02a6536f",
   "metadata": {},
   "outputs": [],
   "source": [
    "from galai.utils import escape_custom_split_sequence\n",
    "from IPython.display import HTML\n",
    "import html\n",
    "\n",
    "def tokenization_example(tokenizer, text):\n",
    "    text = escape_custom_split_sequence(text)\n",
    "    tokens = [tokenizer.decode([x], skip_special_tokens=False) for x in tokenizer.encode(text).ids]\n",
    "    spans = \"</span><span>\".join([html.escape(t).replace(\" \", \"▁\").replace(\"\\n\", \"\\\\n\") for t in tokens])\n",
    "    style = \"<style>.tok-examp1e > span {border: 1px solid #555; padding: 4px 6px; margin: 2px; background: #f8f8f8}</style>\"\n",
    "    return HTML(style + \"<div class='tok-examp1e' style='display: flex; flex-wrap: wrap'><span>\" + spans + \"</span></div>\")\n",
    "\n",
    "tokenization_example(model.tokenizer, r\"\"\"Tokenization of most of the natural texts is not impacted by the rules.\n",
    "However, most of the non-alphanumeric ASCII characters are split. This is mostly visible in TeX formulas,\n",
    "for example: $\\frac{d}{dx}\\,\\cos(x) = -\\sin(x)$, \\(\\zeta(s)=\\sum_{n=1}^{\\infty} n^{-s}\\). \n",
    "It also impacts source codes, like: x+=((1,2)); \n",
    "As a side-effect, contractions (I'll, you've, it's, etc.) and emoticons (like this Santa Claus *<|:‑) ) are split. \n",
    "This rule makes exception for a repeated sequence of the same character, so f.e., ---------------- is still a single token. \n",
    "Additionally, EOL character is always split, so that \n",
    "\n",
    "\n",
    "\n",
    "\n",
    "are 5 tokens. \n",
    "Numbers are slit into individual digits as before, f.e., $$\\pi=3.14159265\\ldots$$ \n",
    "Note that non-alphanumeric splitting splits space in front as well (f.e., i ++, x <-> y, if ( x <= y )). \n",
    "Special tokens like [START_REF], <work> or [IMAGE] are left intact. \n",
    "The tokenizer additionally supports custom sequence splitting (does not work by default, requires a custom preprocessing step), f.e.: \n",
    "[START_DNA]GATTACA[END_DNA], [START_AMINO]PEPTIDES[END_AMINO], \n",
    "[START_SMILES]CC(=O)NCCC1=CNc2c1cc(OC)cc2[END_SMILES] and [START_I_SMILES]CN1CCC[C@H]1c2cccnc2[END_I_SMILES]\"\"\")\n"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "233b44bf",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03a6526a",
   "metadata": {},
   "source": [
    "## Model Selection\n",
    "\n",
    "There are 5 models in total, ranging in size from 125 million parameter up to 121 billion parameters. The model architecture is practically the same as the architecture of OPT models (see [Zhang et al.](https://arxiv.org/abs/2205.01068))."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2a2255a0",
   "metadata": {},
   "source": [
    "## Working with Large Models\n",
    "\n",
    "### Loading a model\n",
    "\n",
    "There are 5 galactica models to choose from, ranging in size from 125 million to 121 billion parameters:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5ed65f9b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table><thead><tr><th>Name</th><th>Parameters</th><th>Layers</th><th>Heads</th><th>Head Size</th><th>Vocabulary Size</th><th>Context Size</th></tr></thead><tbody><tr><td><strong>mini</strong></td><td>125.0 M</td><td>12</td><td>12</td><td>64</td><td>50000</td><td>2048</td></tr><tr><td><strong>base</strong></td><td>1.3 B</td><td>24</td><td>32</td><td>64</td><td>50000</td><td>2048</td></tr><tr><td><strong>standard</strong></td><td>6.7 B</td><td>32</td><td>32</td><td>128</td><td>50000</td><td>2048</td></tr><tr><td><strong>large</strong></td><td>30.0 B</td><td>48</td><td>56</td><td>128</td><td>50000</td><td>2048</td></tr><tr><td><strong>huge</strong></td><td>121.3 B</td><td>96</td><td>80</td><td>128</td><td>50000</td><td>2048</td></tr></tbody></table>"
      ],
      "text/plain": [
       "[ModelInfo(name='mini', num_layers=12, num_heads=12, head_size=64, vocab_size=50000, max_positions=2048),\n",
       " ModelInfo(name='base', num_layers=24, num_heads=32, head_size=64, vocab_size=50000, max_positions=2048),\n",
       " ModelInfo(name='standard', num_layers=32, num_heads=32, head_size=128, vocab_size=50000, max_positions=2048),\n",
       " ModelInfo(name='large', num_layers=48, num_heads=56, head_size=128, vocab_size=50000, max_positions=2048),\n",
       " ModelInfo(name='huge', num_layers=96, num_heads=80, head_size=128, vocab_size=50000, max_positions=2048)]"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from galai.utils import ModelInfo\n",
    "ModelInfo.all()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa1c3ff5",
   "metadata": {},
   "source": [
    "To load a model use the `load_model()` function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4f68177b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on function load_model in module galai:\n",
      "\n",
      "load_model(name: str, dtype: Union[str, torch.dtype] = None, num_gpus: int = None, parallelize: bool = False)\n",
      "    Utility function for loading the model\n",
      "    \n",
      "    Parameters\n",
      "    ----------\n",
      "    name: str\n",
      "        Name of the model\n",
      "    \n",
      "    dtype: str\n",
      "        Optional dtype; default float32 for all models but 'huge'\n",
      "    \n",
      "    num_gpus : int (optional)\n",
      "        Number of GPUs to use for the inference. If None, all available GPUs are used. If 0 (or if\n",
      "        None and there are no GPUs) only a CPU is used. If a positive number n, then the first n CUDA\n",
      "        devices are used.\n",
      "    \n",
      "    parallelize : bool; default False\n",
      "        Specify if to use model tensor parallelizm. Ignored in CPU or single GPU inference.\n",
      "    \n",
      "        By the default (when parallelize is False) the multi-GPU inference is run using accelerate's\n",
      "        pipeline parallelizm in which each GPU is responsible for evaluating a given subset of\n",
      "        model's layers. In this mode evaluations are run sequentially. This mode is well suited for\n",
      "        developing in model's internals as it is more robust in terms of recovering from exceptions\n",
      "        due to not using additional processes. However, because of the sequential nature of\n",
      "        pipeline parallelizm, at any given time only a single GPU is working.\n",
      "    \n",
      "        If parallelize is True, parallelformers' model tensor parallelizm is used instead.\n",
      "    \n",
      "    Returns\n",
      "    ----------\n",
      "    Model - model object\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(gal.load_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "90efcf2a",
   "metadata": {},
   "source": [
    "#### CPU Inference\n",
    "\n",
    "The default call to `load_model` uses all available CUDA devices. If no device is found the model is loaded to RAM instead. Set `num_gpus=0` to force CPU inference even if CUDA-capable devices are present.\n",
    "\n",
    "#### MPS (Metal Programming Shaders) Inference\n",
    "\n",
    "To run the model on Mac OS on Apple GPUs simply call `model.model.to(\"mps\")` right after loading the model."
   ]
  },
  {
   "attachments": {
    "parallel.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "ac9a3a02",
   "metadata": {},
   "source": [
    "### Multi-GPU Inference\n",
    "\n",
    "We support two types of model parallelizm to enable multi-GPU inference: pipeline parallelizm (using [accelerate](https://huggingface.co/docs/accelerate/)) and tensor parallelizm (using [parallelformers](https://tunib-ai.github.io/parallelformers/)). A greatly simplified comparison of the two modes is depicted below:\n",
    "\n",
    "![parallel.png](attachment:parallel.png)\n",
    "\n",
    "In the pipeline parallel mode (`gal.load_model(..., parallelize=False)`, the default) the model weights are split by layers and the input is processed sequentially. This simplifies the synchronization operations required to run the inference. As a result in this mode it's easier to recover from internal model exceptions (like CUDA OOM), inspect model weights or access internal states. However, because the input is being processed sequentially, at any given time only a single GPU is working.\n",
    "\n",
    "To speed up the inference you can load a model with tensor parallelizm enabled (`gal.load_model(..., parallelize=True)`). In this mode the input is split into parts that are processed in parallel. Underneath we use `parallelformers` library that slices transformer-based decoder modules into [Megatron-LM](https://arxiv.org/pdf/1909.08053.pdf) tensor parallel modules. To process the input in parallel, `parallelformers` spawns one additional process for each GPU. As a side effect of this approach, state changes (such as `torch.no_grad()` or `torch.manual_seed()`) triggered from the main process are not visible inside those processes, unless they are manually propagated.\n",
    "\n",
    "In general, both `accelerate` and `parallelformes` have different characteristics in terms of memory usage, communication overhead, inference speed and ease of use (in case of modifying a model internals), so it's best to compare the two in your particular environment. Below we compare the inference time of the `huge` model in half precision on 8 A100 (40GB VRAM, PCIe), an average of 5 runs after a single warm up run:\n",
    "\n",
    "|   Call     | Batch Size | Prompt length | Generated Tokens | Time (accelerate) | Time (parallelformers) |\n",
    "|:-----------|-----------:|--------------:|-----------------:|------------------:|-----------------------:|\n",
    "| generate() |       4    |           100 |             200  |              48 s |                   18 s |\n",
    "\n",
    "### Disk Space Requirements\n",
    "\n",
    "All of the checkpoints files use float16 weights, so on disk file size in bytes is around two times the number of parameters. F.e., the `standard` 6.7B model requires around 13.7 GB of disk space. You can specify different download location by setting the `TRANSFORMERS_CACHE` environment variable accordingly. Make sure to set the variable before importing `transformers` module (including indirect import through `galai`).\n",
    "\n",
    "### Memory Requirements\n",
    "\n",
    "The memory requirement of the models depends on the inference mode. Loading the model in float16 requires two bytes per parameter. That means that f.e., the `large` 30B model requires around 60 GB of memory. Using the full float32 precision doubles the required memory.\n",
    "\n",
    "Besides the model weights one have to include memory size required to store intermediate activations and cached outputs. The cache size can be computed using `ModelInfo`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "33ca4635",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9.4 GB\n"
     ]
    }
   ],
   "source": [
    "batch_size = 8\n",
    "longest_prompt_length = 100\n",
    "max_new_tokens = 200\n",
    "cache_size = ModelInfo.by_name(\"huge\").memory_per_token(dtype=\"float16\") * batch_size * (longest_prompt_length + max_new_tokens)\n",
    "print(f\"{cache_size / 1e9:.1f} GB\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c1467b01",
   "metadata": {},
   "source": [
    "## Non-deterministic Generation\n",
    "\n",
    "While the outputs presented above are quite robust you might notice some differences depending on the exact environment you are using to run the inference. Additionally, even using the exact same environment the outputs might change due to multiple source of non-determinizm in the generation process. Except for the cases in which non-determinism is by design (i.e., sampling outputs with top_p or top_k) or the standard pytorch and CUDA non-determinizm (see https://pytorch.org/docs/stable/notes/randomness.html), there are various reasons for the outputs to be different between environments or between runs on the same environment. Due to an accumulation of numeric errors, the differences are more likely to occur for bigger models and longer sequences. Below is a list of common sources of non-determinizm:\n",
    "\n",
    "* different dtype used for inference: `float32` vs `float16` vs `bfloat16`.\n",
    "* different transformers version. We recommended using `transformers >= 4.24` to take advantage of stability improvements in OPT models implementation.\n",
    "* different pytorch version. We recommended using `torch >= 1.12` to take advantage of more stable implementation of LayerNorm.\n",
    "* different input shape: batch size and padding.\n",
    "* different parallelizm mode: pipeline parallel vs tensor parallel. Additionally, as noted in Multi-GPU Inference Section, manually setting seed values does not work out of the box with parallelformers.\n",
    "* running inference in the training mode. The model architecture includes dropout regularization in several places, which is turned of in the evaluation mode.\n",
    "* differences in prompts: while larger models should be more robust to subtle changes in a prompt, the slightly different input (f.e., two spaces instead of one, additional new line, using `\\$` `\\$` LaTeX delimiters instead of `\\(` `\\)` or no delimiters at all) might results in totally different output."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "462c4f3d",
   "metadata": {},
   "source": [
    "# Pitfalls & Failure Examples\n",
    "\n",
    "While Galactica language models enable one to analyze and work with scientific data in multiple new ways, it's important to understand the shortcomings of the models. We present here examples of cases in which the models don't work as expected. This section is by no means exhaustive.\n",
    "\n",
    "## Hallucinations\n",
    "\n",
    "The language models are trained with an objective of predicting the next token based on the previous tokens. As a result, the text generated by the models may be non-factual or simply made up:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6ff5326c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# Ignacy Jan Paderewski\n",
      "\n",
      " Ignacy Jan Paderewski (Polish: [iɡˈnatsɨ ˈjan padɛˈrɛfskji]; 12 March 1860 – 13 March 1941) was a Polish pianist, composer, and statesman. He was a leading figure in the international music world of the late 19th and early 20th centuries. He was a virtuoso pianist, composer, and conductor, and a political activist who served as the Prime Minister of\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"# Ignacy Jan Paderewski\\n\", max_new_tokens=120))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "14fbd703",
   "metadata": {},
   "source": [
    "Compare the output with [the wikipedia entry](https://en.wikipedia.org/wiki/Ignacy_Jan_Paderewski):\n",
    "\n",
    "> Ignacy Jan Paderewski (Polish: [iɡˈnatsɨ ˈjan padɛˈrɛfskʲi]; 18 November [O.S. 6 November] 1860 – 29 June 1941) was a Polish pianist and composer who became a spokesman for Polish independence. In 1919, he was the new nation's Prime Minister and foreign minister during which he signed the Treaty of Versailles, which ended World War I."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9480d88",
   "metadata": {},
   "source": [
    "The issue is especially visible in case of prompts with incorrect assumptions, in which a prompt already includes made up statements:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b79f2830",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "The Einstein-Presley-Lewandowski equation is given by:\n",
       "\n",
       "$$ \\displaystyle\\frac{\\partial\\rho}{\\partial t}=-\\frac{i}{\\hbar}[H,\\rho]+\\frac{%\n",
       "\\gamma}{2}\\left(2a\\rho a^{\\dagger}-a^{\\dagger}a\\rho-\\rho a^{\\dagger}a\\right) $$\n",
       "$$ \\displaystyle+\\frac{\\gamma(\\bar{n}+1)}{2}\\left(2a^{\\dagger}\\rho a-aa^{\\dagger}%\n",
       "\\rho-\\rho aa^{\\dagger}\\right) $$\n",
       "$$ \\displaystyle+\\frac{\\gamma\\bar{n}}{2}\\left(2a\\rho a^{\\dagger}-a^{\\dagger}a\\rho%\n",
       "-\\rho"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "display_latex(\n",
    "    model.generate(\n",
    "        \"The Einstein-Presley-Lewandowski equation is given by:\\n\",\n",
    "        max_new_tokens=200,\n",
    "        new_doc=True\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a44b9fed",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: what was the main reason that lead to the duel between Richard Feynman and Jadwiga of Poland?\n",
      "\n",
      "Answer: Feynman's refusal to accept the validity of the Pauli exclusion principle\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    model.generate(\n",
    "        \"Question: what was the main reason that lead to the duel between Richard Feynman and Jadwiga of Poland?\\n\\nAnswer:\"\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8be1d402",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: what is the largest prime number?\n",
      "\n",
      "Answer: 1000000007\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    model.generate(\n",
    "        \"Question: what is the largest prime number?\\n\\nAnswer:\"\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cccce667",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: is there the largest prime number?\n",
      "\n",
      "Answer: No\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    model.generate(\n",
    "        \"Question: is there the largest prime number?\\n\\nAnswer:\"\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff995dc7",
   "metadata": {},
   "source": [
    "---\n",
    "The Galactica models are not multi-lingual by design. Most of the natural language documents in the NatureBook corpus are written in **English**. Prompting in different language results in more random generations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d38f3692",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " # Galaxia\n",
      "Una galaxia es un conjunto de estrellas, galaxias, sistemas planetarios, etc. que se encuentran en una determinada region del universo.\n",
      "\n",
      "Galaxia es una herramienta que permite generar simulaciones de galaxias en un determinado momento del universo.\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Spanish prompt\n",
    "print(model.generate(\" # Galaxia\\nUna galaxia es un conjunto de estrellas,\", new_doc=True, max_new_tokens=65))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0d733c31",
   "metadata": {},
   "source": [
    "---\n",
    "A translation by a native speaker:\n",
    "> A galaxy is a group of stars, galaxies, planetary systems, etc. that are located in a specific region of the universe.\n",
    "Galaxy is a tool to generate galaxy simulations at a specific time of the Universe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ca8507bb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: how do you say 'Good morning' in French?\n",
      "\n",
      "Answer: Bonjour\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"Question: how do you say 'Good morning' in French?\\n\\nAnswer:\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a60710e6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: how do you say 'Good morning' in Polish?\n",
      "\n",
      "Answer: Dziękuję!\n",
      "\n",
      "Answer: Dziękuję!\n",
      "\n",
      "Answer: Dziękuję!\n",
      "\n",
      "Answer: Dziękuję!\n",
      "\n",
      "Answer: Dziękuję!\n",
      "\n",
      "Answer: Dziękuj\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"Question: how do you say 'Good morning' in Polish?\\n\\nAnswer:\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12985b70",
   "metadata": {},
   "source": [
    "---\n",
    "Translation of the answer:\n",
    "> Thank you!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4df5d75d",
   "metadata": {},
   "source": [
    "The NatureBook corpus was assembled in July 2022, so the models have no information about anything that happened after."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e4d11f68",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# Elizabeth II\n",
      "\n",
      " Elizabeth II (Elizabeth Alexandra Mary Windsor; born 21 April 1926) is Queen of the United Kingdom and the other Commonwealth realms, including Canada, Australia, New Zealand, Jamaica, Barbados, and 15 other Commonwealth countries\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"# Elizabeth II\\n\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a8f07507",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: What year is it?\n",
      "\n",
      "Answer: 1997\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"Question: What year is it?\\n\\nAnswer:\", new_doc=True))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9004276f",
   "metadata": {},
   "source": [
    "## Prompt Robustness\n",
    "\n",
    "The model output may depend on seemingly insignificant variations in prompts, especially in case of the smaller models. This Section presents examples of prompts in which small change results in different outputs."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5615092d",
   "metadata": {},
   "source": [
    "### Spelling Errors\n",
    "\n",
    "A large part of the NatureBook corpus consists of documents using a formal and technical language. The model output may change depending on spelling, punctuation and grammatical errors in a prompt."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "63a2a038",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: Write a python function that checks if an input string is a palindrome.\n",
      "\n",
      "Answer:\n",
      "\n",
      "```\n",
      "def is_palindrome(s):\n",
      "    return s == s[::-1]\n",
      "```\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    model.generate(\n",
    "        \"Question: Write a python function that checks if an input string is a palindrome.\\n\\nAnswer:\",\n",
    "        max_new_tokens=30\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bc26491d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: Write python function that check if input string is palindrome.\n",
      "\n",
      "Answer: def is_palindrome(s):\n",
      "    if len(s) < 2:\n",
      "        return True\n",
      "    return s[0] == s[-1] and is_palindrome(s[1:-1])\n",
      "\n",
      "\n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    model.generate(\n",
    "        \"Question: Write python function that check if input string is palindrome.\\n\\nAnswer:\",\n",
    "        max_new_tokens=60\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b1181d9c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# Ignacy Jan Paderewski\n",
      "\n",
      " Ignacy Jan Paderewski (Polish: [iɡˈnatsɨ ˈjan padɛˈrɛfskji]; 12 March 1860 – 13 March 1941) was a Polish p\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"# Ignacy Jan Paderewski\\n\"))  # correct name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6ad1649f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# Ignacy Jan Paderweski\n",
      "\n",
      " Ignacy Jan Paderweski (1770–1830) was a Polish painter, a representative of the Polish school of painting.\n",
      "\n",
      "## Biography\n",
      "\n",
      " He was born in Warsaw, the son of a painter, Franciszek Paderwski. He\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"# Ignacy Jan Paderweski\\n\"))  # a typo in the surname"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ad2aad24",
   "metadata": {},
   "source": [
    "### Whitespace Encoding\n",
    "\n",
    "All of the documents in the NatureBook corpus use `\\n` as the end-of-line character."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ee003e54",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: Write a python function that checks if an input string is a palindrome.\r\n",
      "\r\n",
      "Answer:\n",
      "\n",
      "def is_palindrome(s):\n",
      "    if len(s) == 0:\n",
      "        return True\n",
      "    if s[0] != s[-1]:\n",
      "        return False\n",
      "    return is_palindrome(s[1:])\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    model.generate(\n",
    "        \"Question: Write a python function that checks if an input string is a palindrome.\\r\\n\\r\\nAnswer:\",\n",
    "        max_new_tokens=65\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71b6fbaf",
   "metadata": {},
   "source": [
    "Interestingly using a Beam Search helps to produce a working code:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b0b8ab61",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: Write a python function that checks if an input string is a palindrome.\r\n",
      "\r\n",
      "Answer: ```python\n",
      "def is_palindrome(s):\n",
      "    if len(s) == 0:\n",
      "        return True\n",
      "    else:\n",
      "        return s[0] == s[-1] and is_palindrome(s[1:-1])\n",
      "```\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    model.generate(\n",
    "        \"Question: Write a python function that checks if an input string is a palindrome.\\r\\n\\r\\nAnswer:\",\n",
    "        max_new_tokens=65, num_beams=5\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "58ad6f2d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: Write  a python function that  checks if an input string  is a palindrome.\n",
      "\n",
      "Answer: def is_palindrome(s):\n",
      "    if len(s) < 2:\n",
      "        return True\n",
      "    return s[0] == s[-1] and is_palindrome(s[1:-1])\n"
     ]
    }
   ],
   "source": [
    "# multiple spaces between words\n",
    "print(\n",
    "    model.generate(\n",
    "        \"Question: Write  a python function that  checks if an input string  is a palindrome.\\n\\nAnswer:\",\n",
    "        max_new_tokens=55\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80d5e160",
   "metadata": {},
   "source": [
    "---\n",
    "The solution is correct, just different from the one obtained using prompt with normalized spaces.\n",
    "\n",
    "It's a good practice not to include a trailing space in the prompt."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "16b910b0",
   "metadata": {},
   "source": [
    "### TeX formula markers\n",
    "\n",
    "Most of the documents in the NatureBook corpus use `\\(` and `\\)` for inline TeX formulas and `\\[` and `\\]` for display mode maths, but some of the data sources use `$` and `$$` instead."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "13d9a6e0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: What is the expected value of a random variable uniformly distributed over the interval $[a^2, b+c]$?</p>\n",
       "<p>Answer: $\\frac{b+c+a^2}{2}$</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# using \\( \\)\n",
    "display_markdown(\n",
    "    model.generate(\n",
    "\"\"\"Question: What is the expected value of a random variable uniformly distributed over the interval \\\\([a^2, b+c]\\\\)?\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=20)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7a31bff7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: What is the expected value of a random variable uniformly distributed over the interval $$[a^2, b+c]$$?</p>\n",
       "<p>Answer: $\\frac{a^2+b+c}{2}$</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# using \\[ \\]\n",
    "display_markdown(\n",
    "    model.generate(\n",
    "\"\"\"Question: What is the expected value of a random variable uniformly distributed over the interval \\\\[[a^2, b+c]\\\\]?\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=20)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c14d9ae5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: What is the expected value of a random variable uniformly distributed over the interval $[a^2, b+c]$?</p>\n",
       "<p>Answer: $\\begin{aligned} \\operatorname{E}(X) &amp;= \\int_{x=a^2}^{b+c} x \\cdot \\dfrac{1}{b+c-a^2} \\, dx \\\\ &amp;= \\dfrac{1}{b+c-a^2} \\cdot \\left[\\dfrac{x^2}{2} \\right]_{x=a^2}^{b+c} \\\\ &amp;= \\dfrac{1}{b+c-a^2} \\cdot \\left(\\dfrac{(b+c)^2}{2} - \\dfrac{(a^2)^2}{2} \\right) \\\\ &amp;= \\dfrac{1}{b+c-a^2} \\cdot \\left(\\dfrac{b^2+2bc+c^2}{2} - \\dfrac{a^4}{2} \\right) \\\\ &amp;= \\dfrac{1}{b+c-a^2} \\cdot \\left(\\dfrac{b^2+c^2}{2} + bc - \\dfrac{a^4}{2} \\right) \\end{aligned}$</p>\n",
       "<p>In conclusion, the expected value of $X$ is $\\dfrac{1}{b+c-a^2} \\cdot \\left(\\dfrac{b^2+c^2}{2} + bc - \\dfrac{a^4}{2} \\right)$.</p>\n",
       "<p>&lt;/work&gt;</p>\n",
       "<p>Ans: the expected value of $X$ is $\\dfrac{1}{b+c-a^2} \\cdot \\left(\\dfrac{b^2+c^2}{2} + bc - \\dfrac{a^4}{2} \\right)$.</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# using $\n",
    "display_markdown(\n",
    "    model.generate(\n",
    "\"\"\"Question: What is the expected value of a random variable uniformly distributed over the interval $[a^2, b+c]$?\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=500)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3092c2cc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: What is the expected value of a random variable uniformly distributed over the interval $$[a^2, b+c]$$?</p>\n",
       "<p>Answer: </p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# using $$\n",
    "display_markdown(\n",
    "    model.generate(\n",
    "\"\"\"Question: What is the expected value of a random variable uniformly distributed over the interval $$[a^2, b+c]$$?\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=40)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "141bf95d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: What is the expected value of a random variable uniformly distributed over the interval [a^2, b+c]?</p>\n",
       "<p>Answer: $\\frac{b+c+a^{2}}{2}$</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# plaintext math\n",
    "display_markdown(\n",
    "    model.generate(\n",
    "\"\"\"Question: What is the expected value of a random variable uniformly distributed over the interval [a^2, b+c]?\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=22)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "64e17893",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>Question: What is the expected value of a random variable uniformly distributed over the interval $[a^2, b+c]$?</p>\n",
       "<p>Answer: $\\dfrac{a^2+b+c}{2}$</p>\n",
       "<p>&lt;/work&gt;</p>\n",
       "<p>Answer: $\\dfrac{a^2+b+c}{2}$</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# using $, beam search\n",
    "display_markdown(\n",
    "    model.generate(\n",
    "\"\"\"Question: What is the expected value of a random variable uniformly distributed over the interval $[a^2, b+c]$?\n",
    "\n",
    "Answer:\"\"\", max_new_tokens=50, num_beams=5)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca469fb3",
   "metadata": {},
   "source": [
    "Please note that `display_latex` and `display_markdown` convert the markers to `$` and `$$` only for the display purposes."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f08f6e3e",
   "metadata": {},
   "source": [
    "### Letter-case"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6c03be57",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: what is Alzheimer's Disease?\n",
      "\n",
      "Answer: Alzheimer's disease (AD) is a neurodegenerative disorder that is characterized by progressive cognitive decline and memory loss. The disease is the most common cause of dementia in the elderly. The neuropathological hallmarks of AD are the presence of extracellular amyloid plaques and intracellular neurofibrillary tangles (\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    model.generate(\"Question: what is Alzheimer's Disease?\\n\\n\")\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3f34a25e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: what is alzheimer's disease?\n",
      "\n",
      "Answer: Alzheimer's disease (AD) is a neurodegenerative disorder characterized by progressive cognitive decline and memory loss. The neuropathological hallmarks of AD are the presence of extracellular amyloid plaques and intracellular neurofibrillary tangles (NFTs). Amyloid plaques are composed of amyloid-β (Aβ\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    model.generate(\"Question: what is alzheimer's disease?\\n\\n\")\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0f71ced5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question: what is ALZHEIMER'S DISEASE?\n",
      "\n",
      "Answer: Alzheimer's disease (AD) is a progressive neurodegenerative disorder that is the most common cause of dementia in the elderly. It is characterized by the presence of two types of protein deposits in the brain: extracellular amyloid plaques and intracellular neurofibrillary tangles. The amyloid plaques are composed of\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    model.generate(\"Question: what is ALZHEIMER'S DISEASE?\\n\\n\")\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d6258b4",
   "metadata": {},
   "source": [
    "### New document mode\n",
    "\n",
    "To make the training efficient, all the documents in the training corpus were concatenated into a single sequence of tokens, with `</s>` token as a document boundary marker.\n",
    "\n",
    "If you want autocomplete based functionality, it is often good to experiment with turning off the new document mode by setting `new_doc=False` in calls to `generate`. This puts the generation into continuation mode, which means that the prompt may be in the middle of a document, as opposed to the beginning."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a307b07b",
   "metadata": {},
   "source": [
    "## Other Limitations"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2f99ffb",
   "metadata": {},
   "source": [
    "### [START_REF] position\n",
    "\n",
    "As described above we use the `[START_REF]` token to generate in-context references. The token is a form of \"insert citation here\" instruction to the model. This means that the word order may impact what paper is generated."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc12a702",
   "metadata": {},
   "source": [
    "### Correct Answer with Incorrect reasoning\n",
    "\n",
    "In the example below we use the question answering template to get a solution for a probability question:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "67a91956",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p><strong>Prompt</strong>: Question: We flip a fair coin $3$ times. What is the probability of getting an even number of heads?</p>\n",
       "<p>Answer:</p>\n",
       "<p><strong>Answer</strong>: Question: We flip a fair coin $3$ times. What is the probability of getting an even number of heads?</p>\n",
       "<p>Answer: $\\dfrac{1}{2}$</p>\n",
       "<p>&lt;/work&gt;</p>\n",
       "<p>Ans: $\\dfrac{1}{2}$</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt = f\"Question: We flip a fair coin $3$ times. What is the probability of getting an even number of heads?\\n\\nAnswer:\"\n",
    "answer = model.generate(prompt, new_doc=True)\n",
    "display_markdown(f\"**Prompt**: {prompt}\\n\\n**Answer**: {answer}\\n\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "606aef41",
   "metadata": {},
   "source": [
    "The model provides a correct answer. Trying to get reasoning with <work> token we get different answer:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2f5f16e7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p><strong>Prompt</strong>: Question: We flip a fair coin $3$ times. What is the probability of getting an even number of heads?</p>\n",
       "<p>&lt;work&gt;</p>\n",
       "<p><strong>Answer</strong>: Question: We flip a fair coin $3$ times. What is the probability of getting an even number of heads?</p>\n",
       "<p>&lt;work&gt;</p>\n",
       "<p>Flipping a fair coin $3$ times has the same probability as picking a card from a standard deck of $52$ cards and then randomly putting it back.</p>\n",
       "<p>There are $2$ cards that have an even number of heads: $\\text{HHH}$ and $\\text{TTT}$.</p>\n",
       "<p>There are $2^3=8$ total ways to flip a coin $3$ times.</p>\n",
       "<p>The probability of getting an even number of heads is $\\dfrac{2}{8}=\\dfrac{1}{4}$.</p>\n",
       "<p>&lt;/work&gt;</p>\n",
       "<p>Ans: The probability of getting an even number of heads is </p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt = f\"Question: We flip a fair coin $3$ times. What is the probability of getting an even number of heads?\\n\\n<work>\"\n",
    "answer = model.generate(prompt, new_doc=True, max_new_tokens=150)\n",
    "display_markdown(f\"**Prompt**: {prompt}\\n\\n**Answer**: {answer}\\n\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b51fa923",
   "metadata": {},
   "source": [
    "Sometimes a prompt might be \"too robust\". Let's reconsider the dot-product attention example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a2d26f3d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>The paper that presented a novel computing block given by the formula:\n",
       "$$\n",
       "f(Q, K, V) = \\textrm{softmax}\\left(\\frac{QK^T}{\\sqrt{d_k}}\\right)V\n",
       "$$</p>\n",
       "<p><strong>Reference</strong>: Attention is All you Need, Vaswani</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt1 = \"\"\"The paper that presented a novel computing block given by the formula:\n",
    "\\\\[\n",
    "f(Q, K, V) = \\\\textrm{softmax}\\\\left(\\\\frac{QK^T}{\\\\sqrt{d_k}}\\\\right)V\n",
    "\\\\]\n",
    "\n",
    "\"\"\"\n",
    "ref = model.generate_reference(prompt1)\n",
    "display_markdown(f\"{prompt1}\\n**Reference**: {ref}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5e137c4",
   "metadata": {},
   "source": [
    "How much we can change the formula to still get that reference? Because of the hallucination issue described above the formula can be changed a lot:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3308cfe0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p>The paper that presented a novel computing block given by the formula:\n",
       "$$\n",
       "f(X) = \\cos\\left(\\frac{X}{d_k}\\right)\n",
       "$$</p>\n",
       "<p><strong>Reference</strong>: Attention is All you Need, Vaswani</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt2 = \"\"\"The paper that presented a novel computing block given by the formula:\n",
    "\\\\[\n",
    "f(X) = \\\\cos\\\\left(\\\\frac{X}{d_k}\\\\right)\n",
    "\\\\]\n",
    "\n",
    "\"\"\"\n",
    "ref = model.generate_reference(prompt2)\n",
    "display_markdown(f\"{prompt2}\\n**Reference**: {ref}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5040f8a1",
   "metadata": {},
   "source": [
    "In the example above the model is forced into reference generation with a false premise that such a paper introducing the formula exists. One option to mitigate the issue is to provide a few-shot prompt with examples in which the answer can be \"no such paper\". Another option is to set up a generation threshold on model's logits:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4dca6877",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "\n",
    "def score_generation(model, prompt):\n",
    "    tokens = model._tokenize([prompt], new_doc=False)\n",
    "    out = model.model.generate(\n",
    "        tokens,\n",
    "        max_new_tokens=40,\n",
    "        return_dict_in_generate=True,\n",
    "        output_scores=True\n",
    "    )\n",
    "    generation = out.sequences[0, len(tokens[0]):].tolist()\n",
    "    scores = []\n",
    "    end_ref_id = model.tokenizer.token_to_id(\"[END_REF]\")\n",
    "    for pos, token_id in enumerate(generation):\n",
    "        log_probs = torch.nn.functional.log_softmax(out.scores[pos], dim=-1)\n",
    "        scores.append(log_probs[0, token_id].item())\n",
    "        if token_id == end_ref_id:\n",
    "            break\n",
    "    text = model.tokenizer.decode(generation[:pos], skip_special_tokens=False)\n",
    "    scores = torch.tensor(scores)\n",
    "    return display_markdown(f\"\"\"**Prompt**: {prompt}\n",
    "\n",
    "**Predicted reference**: {text}\n",
    "\n",
    "**Min score**: {scores.min().item():.2f}\n",
    "\n",
    "**Sum score**: {scores.sum().item():.2f}\n",
    "\n",
    "\"\"\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "56d44aaa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p><strong>Prompt</strong>: The paper that presented a novel computing block given by the formula:\n",
       "$$\n",
       "f(Q, K, V) = \\textrm{softmax}\\left(\\frac{QK^T}{\\sqrt{d_k}}\\right)V\n",
       "$$</p>\n",
       "<p>[START_REF]</p>\n",
       "<p><strong>Predicted reference</strong>:  Attention is All you Need, Vaswani</p>\n",
       "<p><strong>Min score</strong>: -0.29</p>\n",
       "<p><strong>Sum score</strong>: -0.35</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score_generation(model, prompt1 + \"[START_REF]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c2d8ec13",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p><strong>Prompt</strong>: The paper that presented a novel computing block given by the formula:\n",
       "$$\n",
       "f(X) = \\cos\\left(\\frac{X}{d_k}\\right)\n",
       "$$</p>\n",
       "<p>[START_REF]</p>\n",
       "<p><strong>Predicted reference</strong>:  Attention is All you Need, Vaswani</p>\n",
       "<p><strong>Min score</strong>: -1.45</p>\n",
       "<p><strong>Sum score</strong>: -1.52</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score_generation(model, prompt2 + \"[START_REF]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5a58d329",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p><strong>Prompt</strong>: The $E=m c^2$ paper [START_REF]</p>\n",
       "<p><strong>Predicted reference</strong>:  Ist die Trägheit eines Körpers von seinem Energieinhalt abhängig, Einstein</p>\n",
       "<p><strong>Min score</strong>: -0.42</p>\n",
       "<p><strong>Sum score</strong>: -0.58</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score_generation(model, \"The $E=m c^2$ paper [START_REF]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3336be1e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<p><strong>Prompt</strong>: The $E=m c^3$ paper [START_REF]</p>\n",
       "<p><strong>Predicted reference</strong>:  Ist die Trägheit eines Körpers von seinem Energieinhalt abhängig, Einstein</p>\n",
       "<p><strong>Min score</strong>: -2.01</p>\n",
       "<p><strong>Sum score</strong>: -2.15</p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score_generation(model, \"The $E=m c^3$ paper [START_REF]\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d19da87",
   "metadata": {},
   "source": [
    "---\n",
    "It can happen, especially with few-shot prompts, that the models continue to generate text after the expected answer. For example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "02877652",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The IUPAC name of cortisol is: 11β,17α,21-trihydroxypregn-4-ene-3,20-dione.\n",
      "\n",
      "## See also\n",
      "\n",
      "* Cortisone\n",
      "* Corticosterone\n",
      "* Hydrocortisone\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"The IUPAC name of cortisol is:\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f2d48da",
   "metadata": {},
   "source": [
    "For this reason some of the generations above were specifying `max_new_tokens` manually to make the examples more readable and easier to follow. In practice a post-processing step may be needed, depending on use case."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d8d734c9",
   "metadata": {},
   "source": [
    "The generation might assume a different type of document than expected:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d24486e3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The IUPAC name of cortisol is:\n",
      "\n",
      "A. 17-Hydroxyprogesterone\n",
      "B. 11-Deoxycortisol\n",
      "C. 11-Deoxycorticosterone\n",
      "D. 17-Hydroxycorticosterone\n",
      "\n",
      "Answer: B\n"
     ]
    }
   ],
   "source": [
    "print(model.generate(\"The IUPAC name of cortisol is:\\n\\n\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d3e070d",
   "metadata": {},
   "source": [
    "For more details see [our paper](https://galactica.org/paper.pdf)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f1028002",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.9.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}