{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ed2ab48877fe47178e9e521fae619346",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer_config.json:   0%|          | 0.00/7.34k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b348b72e4f4b412f949efee9dd3da8d2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "vocab.json:   0%|          | 0.00/798k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c3c9cb29e5184e6ba7d0e698209b1dbe",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "merges.txt:   0%|          | 0.00/456k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d7af8afa19f04ebda69594a35b75ebbd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer.json:   0%|          | 0.00/2.11M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b9d9a927b1f446b2818dbb997608fd93",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "added_tokens.json:   0%|          | 0.00/1.08k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "77f52404773949c5b6e792eb2b5259dd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "special_tokens_map.json:   0%|          | 0.00/99.0 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from transformers import AutoTokenizer\n",
    "\n",
    "tokenizer = AutoTokenizer.from_pretrained(\"vikhyatk/moondream2\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "texts = [\n",
    "    \"This is a short text.\",\n",
    "    \"This is a much longer text that will determine the padding length.\",\n",
    "    \"Medium length text here.\"\n",
    "]\n",
    "tokenizer.pad_token = tokenizer.eos_token\n",
    "\n",
    "# Pad to the longest sequence in the batch\n",
    "encoded = tokenizer(\n",
    "    texts,\n",
    "    padding=True,  # or padding=\"longest\"\n",
    "    return_tensors=\"pt\",  # or \"tf\" for TensorFlow\n",
    "    model_max_length=512,\n",
    "\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "transformers.tokenization_utils_base.BatchEncoding"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(encoded)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'input_ids': tensor([[ 1212,   318,   257,  1790,  2420,    13, 50256, 50256, 50256, 50256,\n",
       "         50256, 50256, 50256],\n",
       "        [ 1212,   318,   257,   881,  2392,  2420,   326,   481,  5004,   262,\n",
       "         24511,  4129,    13],\n",
       "        [31205,  4129,  2420,   994,    13, 50256, 50256, 50256, 50256, 50256,\n",
       "         50256, 50256, 50256]]), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],\n",
       "        [1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0]])}"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([3, 13])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded.input_ids.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0])"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded.attention_mask[0] * encoded.attention_mask[0].T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "mask = encoded.attention_mask[0].clone().reshape(-1, 1)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "encoded.attention_mask[0].T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([13, 1])"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mask.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],\n",
       "        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]])"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "real = mask @ mask.T\n",
    "real"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "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": "3.13.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}