chatterbox-indic-lora / tokenizer /brahmic_init_map.json
reenigne314
Move files to repo root
09417e3
Raw
History Blame Contribute Delete
14.4 kB
{
"description": "Brahmic cross-script embedding initialization map",
"format": "new_token_id (str) → devanagari_token_id (int)",
"original_vocab_size": 2602,
"extended_vocab_size": 2871,
"languages_added": [
"bn",
"ta",
"ml",
"mr",
"gu"
],
"n_mappings": 223,
"map": {
"2608": 1671,
"2609": 1672,
"2610": 1673,
"2611": 1674,
"2612": 1675,
"2613": 1676,
"2614": 1677,
"2615": 1681,
"2616": 1682,
"2617": 1685,
"2618": 1686,
"2619": 1687,
"2620": 1688,
"2621": 1689,
"2622": 1690,
"2623": 1691,
"2624": 1692,
"2625": 1693,
"2626": 1694,
"2627": 1695,
"2628": 1696,
"2629": 1697,
"2630": 1698,
"2632": 1700,
"2633": 1701,
"2634": 1702,
"2635": 1703,
"2636": 1704,
"2637": 1705,
"2638": 1706,
"2639": 1708,
"2640": 1709,
"2641": 1710,
"2642": 1711,
"2643": 1712,
"2644": 1713,
"2645": 1714,
"2646": 1715,
"2647": 1718,
"2648": 1719,
"2649": 1720,
"2650": 1721,
"2652": 1725,
"2653": 1726,
"2654": 1727,
"2655": 1728,
"2656": 1729,
"2657": 1730,
"2658": 1734,
"2659": 1735,
"2660": 1738,
"2661": 1739,
"2606": 1668,
"2607": 1669,
"2662": 1740,
"2675": 1671,
"2676": 1672,
"2677": 1673,
"2678": 1674,
"2679": 1675,
"2680": 1676,
"2681": 1681,
"2682": 1682,
"2683": 1682,
"2684": 1685,
"2685": 1686,
"2686": 1686,
"2687": 1687,
"2688": 1691,
"2689": 1692,
"2691": 1696,
"2692": 1697,
"2693": 1701,
"2694": 1702,
"2695": 1706,
"2697": 1708,
"2698": 1712,
"2699": 1713,
"2700": 1714,
"2702": 1715,
"2705": 1717,
"2704": 1716,
"2703": 1716,
"2701": 1714,
"2696": 1706,
"2690": 1694,
"2706": 1719,
"2707": 1720,
"2708": 1721,
"2709": 1725,
"2710": 1726,
"2711": 1727,
"2712": 1728,
"2713": 1729,
"2714": 1734,
"2715": 1735,
"2716": 1735,
"2717": 1738,
"2718": 1739,
"2719": 1739,
"2673": 1668,
"2674": 1669,
"2720": 1740,
"2733": 1671,
"2734": 1672,
"2735": 1673,
"2736": 1674,
"2737": 1675,
"2738": 1676,
"2739": 1677,
"2740": 1681,
"2741": 1682,
"2742": 1682,
"2743": 1685,
"2744": 1686,
"2745": 1686,
"2746": 1687,
"2747": 1688,
"2748": 1689,
"2749": 1690,
"2750": 1691,
"2751": 1692,
"2752": 1693,
"2753": 1694,
"2754": 1695,
"2755": 1696,
"2756": 1697,
"2757": 1698,
"2758": 1699,
"2759": 1700,
"2760": 1701,
"2761": 1702,
"2762": 1703,
"2763": 1704,
"2764": 1705,
"2765": 1706,
"2766": 1708,
"2767": 1709,
"2768": 1710,
"2769": 1711,
"2770": 1712,
"2771": 1713,
"2772": 1714,
"2773": 1715,
"2774": 1716,
"2775": 1717,
"2776": 1718,
"2777": 1719,
"2778": 1720,
"2779": 1721,
"2780": 1725,
"2781": 1726,
"2782": 1727,
"2783": 1728,
"2784": 1729,
"2785": 1730,
"2786": 1734,
"2787": 1735,
"2788": 1735,
"2789": 1738,
"2790": 1739,
"2791": 1739,
"2731": 1668,
"2732": 1669,
"2792": 1740,
"2805": 1671,
"2806": 1672,
"2807": 1673,
"2808": 1674,
"2809": 1675,
"2810": 1676,
"2811": 1677,
"2812": 1681,
"2813": 1682,
"2814": 1685,
"2815": 1686,
"2816": 1687,
"2817": 1688,
"2818": 1689,
"2819": 1690,
"2820": 1691,
"2821": 1692,
"2822": 1693,
"2823": 1694,
"2824": 1695,
"2825": 1696,
"2826": 1697,
"2827": 1698,
"2828": 1699,
"2829": 1700,
"2830": 1701,
"2831": 1702,
"2832": 1703,
"2833": 1704,
"2834": 1705,
"2835": 1706,
"2836": 1708,
"2837": 1709,
"2838": 1710,
"2839": 1711,
"2840": 1712,
"2841": 1713,
"2842": 1714,
"2843": 1715,
"2844": 1716,
"2845": 1717,
"2846": 1718,
"2847": 1719,
"2848": 1720,
"2849": 1721,
"2850": 1725,
"2851": 1726,
"2852": 1727,
"2853": 1728,
"2854": 1729,
"2855": 1730,
"2856": 1734,
"2857": 1735,
"2858": 1738,
"2859": 1739,
"2803": 1668,
"2804": 1669,
"2860": 1740
},
"readable": {
"অ (bn)": "अ (hi) [id 1671→2608]",
"আ (bn)": "आ (hi) [id 1672→2609]",
"ই (bn)": "इ (hi) [id 1673→2610]",
"ঈ (bn)": "ई (hi) [id 1674→2611]",
"উ (bn)": "उ (hi) [id 1675→2612]",
"ঊ (bn)": "ऊ (hi) [id 1676→2613]",
"ঋ (bn)": "ऋ (hi) [id 1677→2614]",
"এ (bn)": "ए (hi) [id 1681→2615]",
"ঐ (bn)": "ऐ (hi) [id 1682→2616]",
"ও (bn)": "ओ (hi) [id 1685→2617]",
"ঔ (bn)": "औ (hi) [id 1686→2618]",
"ক (bn)": "क (hi) [id 1687→2619]",
"খ (bn)": "ख (hi) [id 1688→2620]",
"গ (bn)": "ग (hi) [id 1689→2621]",
"ঘ (bn)": "घ (hi) [id 1690→2622]",
"ঙ (bn)": "ङ (hi) [id 1691→2623]",
"চ (bn)": "च (hi) [id 1692→2624]",
"ছ (bn)": "छ (hi) [id 1693→2625]",
"জ (bn)": "ज (hi) [id 1694→2626]",
"ঝ (bn)": "झ (hi) [id 1695→2627]",
"ঞ (bn)": "ञ (hi) [id 1696→2628]",
"ট (bn)": "ट (hi) [id 1697→2629]",
"ঠ (bn)": "ठ (hi) [id 1698→2630]",
"ঢ (bn)": "ढ (hi) [id 1700→2632]",
"ণ (bn)": "ण (hi) [id 1701→2633]",
"ত (bn)": "त (hi) [id 1702→2634]",
"থ (bn)": "थ (hi) [id 1703→2635]",
"দ (bn)": "द (hi) [id 1704→2636]",
"ধ (bn)": "ध (hi) [id 1705→2637]",
"ন (bn)": "न (hi) [id 1706→2638]",
"প (bn)": "प (hi) [id 1708→2639]",
"ফ (bn)": "फ (hi) [id 1709→2640]",
"ব (bn)": "ब (hi) [id 1710→2641]",
"ভ (bn)": "भ (hi) [id 1711→2642]",
"ম (bn)": "म (hi) [id 1712→2643]",
"য (bn)": "य (hi) [id 1713→2644]",
"র (bn)": "र (hi) [id 1714→2645]",
"ল (bn)": "ल (hi) [id 1715→2646]",
"শ (bn)": "श (hi) [id 1718→2647]",
"ষ (bn)": "ष (hi) [id 1719→2648]",
"স (bn)": "स (hi) [id 1720→2649]",
"হ (bn)": "ह (hi) [id 1721→2650]",
"া (bn)": "ा (hi) [id 1725→2652]",
"ি (bn)": "ि (hi) [id 1726→2653]",
"ী (bn)": "ी (hi) [id 1727→2654]",
"ু (bn)": "ु (hi) [id 1728→2655]",
"ূ (bn)": "ू (hi) [id 1729→2656]",
"ৃ (bn)": "ृ (hi) [id 1730→2657]",
"ে (bn)": "े (hi) [id 1734→2658]",
"ৈ (bn)": "ै (hi) [id 1735→2659]",
"ো (bn)": "ो (hi) [id 1738→2660]",
"ৌ (bn)": "ौ (hi) [id 1739→2661]",
"ং (bn)": "ं (hi) [id 1668→2606]",
"ঃ (bn)": "ः (hi) [id 1669→2607]",
"্ (bn)": "् (hi) [id 1740→2662]",
"அ (ta)": "अ (hi) [id 1671→2675]",
"ஆ (ta)": "आ (hi) [id 1672→2676]",
"இ (ta)": "इ (hi) [id 1673→2677]",
"ஈ (ta)": "ई (hi) [id 1674→2678]",
"உ (ta)": "उ (hi) [id 1675→2679]",
"ஊ (ta)": "ऊ (hi) [id 1676→2680]",
"எ (ta)": "ए (hi) [id 1681→2681]",
"ஏ (ta)": "ऐ (hi) [id 1682→2682]",
"ஐ (ta)": "ऐ (hi) [id 1682→2683]",
"ஒ (ta)": "ओ (hi) [id 1685→2684]",
"ஓ (ta)": "औ (hi) [id 1686→2685]",
"ஔ (ta)": "औ (hi) [id 1686→2686]",
"க (ta)": "क (hi) [id 1687→2687]",
"ங (ta)": "ङ (hi) [id 1691→2688]",
"ச (ta)": "च (hi) [id 1692→2689]",
"ஞ (ta)": "ञ (hi) [id 1696→2691]",
"ட (ta)": "ट (hi) [id 1697→2692]",
"ண (ta)": "ण (hi) [id 1701→2693]",
"த (ta)": "त (hi) [id 1702→2694]",
"ந (ta)": "न (hi) [id 1706→2695]",
"ப (ta)": "प (hi) [id 1708→2697]",
"ம (ta)": "म (hi) [id 1712→2698]",
"ய (ta)": "य (hi) [id 1713→2699]",
"ர (ta)": "र (hi) [id 1714→2700]",
"ல (ta)": "ल (hi) [id 1715→2702]",
"வ (ta)": "व (hi) [id 1717→2705]",
"ழ (ta)": "ळ (hi) [id 1716→2704]",
"ள (ta)": "ळ (hi) [id 1716→2703]",
"ற (ta)": "र (hi) [id 1714→2701]",
"ன (ta)": "न (hi) [id 1706→2696]",
"ஜ (ta)": "ज (hi) [id 1694→2690]",
"ஷ (ta)": "ष (hi) [id 1719→2706]",
"ஸ (ta)": "स (hi) [id 1720→2707]",
"ஹ (ta)": "ह (hi) [id 1721→2708]",
"ா (ta)": "ा (hi) [id 1725→2709]",
"ி (ta)": "ि (hi) [id 1726→2710]",
"ீ (ta)": "ी (hi) [id 1727→2711]",
"ு (ta)": "ु (hi) [id 1728→2712]",
"ூ (ta)": "ू (hi) [id 1729→2713]",
"ெ (ta)": "े (hi) [id 1734→2714]",
"ே (ta)": "ै (hi) [id 1735→2715]",
"ை (ta)": "ै (hi) [id 1735→2716]",
"ொ (ta)": "ो (hi) [id 1738→2717]",
"ோ (ta)": "ौ (hi) [id 1739→2718]",
"ௌ (ta)": "ौ (hi) [id 1739→2719]",
"ஂ (ta)": "ं (hi) [id 1668→2673]",
"ஃ (ta)": "ः (hi) [id 1669→2674]",
"் (ta)": "् (hi) [id 1740→2720]",
"അ (ml)": "अ (hi) [id 1671→2733]",
"ആ (ml)": "आ (hi) [id 1672→2734]",
"ഇ (ml)": "इ (hi) [id 1673→2735]",
"ഈ (ml)": "ई (hi) [id 1674→2736]",
"ഉ (ml)": "उ (hi) [id 1675→2737]",
"ഊ (ml)": "ऊ (hi) [id 1676→2738]",
"ഋ (ml)": "ऋ (hi) [id 1677→2739]",
"എ (ml)": "ए (hi) [id 1681→2740]",
"ഏ (ml)": "ऐ (hi) [id 1682→2741]",
"ഐ (ml)": "ऐ (hi) [id 1682→2742]",
"ഒ (ml)": "ओ (hi) [id 1685→2743]",
"ഓ (ml)": "औ (hi) [id 1686→2744]",
"ഔ (ml)": "औ (hi) [id 1686→2745]",
"ക (ml)": "क (hi) [id 1687→2746]",
"ഖ (ml)": "ख (hi) [id 1688→2747]",
"ഗ (ml)": "ग (hi) [id 1689→2748]",
"ഘ (ml)": "घ (hi) [id 1690→2749]",
"ങ (ml)": "ङ (hi) [id 1691→2750]",
"ച (ml)": "च (hi) [id 1692→2751]",
"ഛ (ml)": "छ (hi) [id 1693→2752]",
"ജ (ml)": "ज (hi) [id 1694→2753]",
"ഝ (ml)": "झ (hi) [id 1695→2754]",
"ഞ (ml)": "ञ (hi) [id 1696→2755]",
"ട (ml)": "ट (hi) [id 1697→2756]",
"ഠ (ml)": "ठ (hi) [id 1698→2757]",
"ഡ (ml)": "ड (hi) [id 1699→2758]",
"ഢ (ml)": "ढ (hi) [id 1700→2759]",
"ണ (ml)": "ण (hi) [id 1701→2760]",
"ത (ml)": "त (hi) [id 1702→2761]",
"ഥ (ml)": "थ (hi) [id 1703→2762]",
"ദ (ml)": "द (hi) [id 1704→2763]",
"ധ (ml)": "ध (hi) [id 1705→2764]",
"ന (ml)": "न (hi) [id 1706→2765]",
"പ (ml)": "प (hi) [id 1708→2766]",
"ഫ (ml)": "फ (hi) [id 1709→2767]",
"ബ (ml)": "ब (hi) [id 1710→2768]",
"ഭ (ml)": "भ (hi) [id 1711→2769]",
"മ (ml)": "म (hi) [id 1712→2770]",
"യ (ml)": "य (hi) [id 1713→2771]",
"ര (ml)": "र (hi) [id 1714→2772]",
"ല (ml)": "ल (hi) [id 1715→2773]",
"ള (ml)": "ळ (hi) [id 1716→2774]",
"വ (ml)": "व (hi) [id 1717→2775]",
"ശ (ml)": "श (hi) [id 1718→2776]",
"ഷ (ml)": "ष (hi) [id 1719→2777]",
"സ (ml)": "स (hi) [id 1720→2778]",
"ഹ (ml)": "ह (hi) [id 1721→2779]",
"ാ (ml)": "ा (hi) [id 1725→2780]",
"ി (ml)": "ि (hi) [id 1726→2781]",
"ീ (ml)": "ी (hi) [id 1727→2782]",
"ു (ml)": "ु (hi) [id 1728→2783]",
"ൂ (ml)": "ू (hi) [id 1729→2784]",
"ൃ (ml)": "ृ (hi) [id 1730→2785]",
"െ (ml)": "े (hi) [id 1734→2786]",
"േ (ml)": "ै (hi) [id 1735→2787]",
"ൈ (ml)": "ै (hi) [id 1735→2788]",
"ൊ (ml)": "ो (hi) [id 1738→2789]",
"ോ (ml)": "ौ (hi) [id 1739→2790]",
"ൌ (ml)": "ौ (hi) [id 1739→2791]",
"ം (ml)": "ं (hi) [id 1668→2731]",
"ഃ (ml)": "ः (hi) [id 1669→2732]",
"് (ml)": "् (hi) [id 1740→2792]",
"અ (gu)": "अ (hi) [id 1671→2805]",
"આ (gu)": "आ (hi) [id 1672→2806]",
"ઇ (gu)": "इ (hi) [id 1673→2807]",
"ઈ (gu)": "ई (hi) [id 1674→2808]",
"ઉ (gu)": "उ (hi) [id 1675→2809]",
"ઊ (gu)": "ऊ (hi) [id 1676→2810]",
"ઋ (gu)": "ऋ (hi) [id 1677→2811]",
"એ (gu)": "ए (hi) [id 1681→2812]",
"ઐ (gu)": "ऐ (hi) [id 1682→2813]",
"ઓ (gu)": "ओ (hi) [id 1685→2814]",
"ઔ (gu)": "औ (hi) [id 1686→2815]",
"ક (gu)": "क (hi) [id 1687→2816]",
"ખ (gu)": "ख (hi) [id 1688→2817]",
"ગ (gu)": "ग (hi) [id 1689→2818]",
"ઘ (gu)": "घ (hi) [id 1690→2819]",
"ઙ (gu)": "ङ (hi) [id 1691→2820]",
"ચ (gu)": "च (hi) [id 1692→2821]",
"છ (gu)": "छ (hi) [id 1693→2822]",
"જ (gu)": "ज (hi) [id 1694→2823]",
"ઝ (gu)": "झ (hi) [id 1695→2824]",
"ઞ (gu)": "ञ (hi) [id 1696→2825]",
"ટ (gu)": "ट (hi) [id 1697→2826]",
"ઠ (gu)": "ठ (hi) [id 1698→2827]",
"ડ (gu)": "ड (hi) [id 1699→2828]",
"ઢ (gu)": "ढ (hi) [id 1700→2829]",
"ણ (gu)": "ण (hi) [id 1701→2830]",
"ત (gu)": "त (hi) [id 1702→2831]",
"થ (gu)": "थ (hi) [id 1703→2832]",
"દ (gu)": "द (hi) [id 1704→2833]",
"ધ (gu)": "ध (hi) [id 1705→2834]",
"ન (gu)": "न (hi) [id 1706→2835]",
"પ (gu)": "प (hi) [id 1708→2836]",
"ફ (gu)": "फ (hi) [id 1709→2837]",
"બ (gu)": "ब (hi) [id 1710→2838]",
"ભ (gu)": "भ (hi) [id 1711→2839]",
"મ (gu)": "म (hi) [id 1712→2840]",
"ય (gu)": "य (hi) [id 1713→2841]",
"ર (gu)": "र (hi) [id 1714→2842]",
"લ (gu)": "ल (hi) [id 1715→2843]",
"ળ (gu)": "ळ (hi) [id 1716→2844]",
"વ (gu)": "व (hi) [id 1717→2845]",
"શ (gu)": "श (hi) [id 1718→2846]",
"ષ (gu)": "ष (hi) [id 1719→2847]",
"સ (gu)": "स (hi) [id 1720→2848]",
"હ (gu)": "ह (hi) [id 1721→2849]",
"ા (gu)": "ा (hi) [id 1725→2850]",
"િ (gu)": "ि (hi) [id 1726→2851]",
"ી (gu)": "ी (hi) [id 1727→2852]",
"ુ (gu)": "ु (hi) [id 1728→2853]",
"ૂ (gu)": "ू (hi) [id 1729→2854]",
"ૃ (gu)": "ृ (hi) [id 1730→2855]",
"ે (gu)": "े (hi) [id 1734→2856]",
"ૈ (gu)": "ै (hi) [id 1735→2857]",
"ો (gu)": "ो (hi) [id 1738→2858]",
"ૌ (gu)": "ौ (hi) [id 1739→2859]",
"ં (gu)": "ं (hi) [id 1668→2803]",
"ઃ (gu)": "ः (hi) [id 1669→2804]",
"્ (gu)": "् (hi) [id 1740→2860]"
}
}