added substr derivations with one example

3789381c6cbd0b495c31a3b6c5236d57d4b7eb79

Karl-Johan Alm <karljohan-alm@garage.co.jp>

Signed

public/script.js+1 -1 Viewed

             const data = await response.json();
             if (data) {
                 const { chat_template, chat_template_hash } = data;
                 console.log(`${wantsContextDerivation} ${wantsInstructDerivation} We have chat template ${chat_template.split('\n')[0]}...`);
                 const templates = await deriveTemplatesFromChatTemplate(chat_template, chat_template_hash);
                 if (templates) {
                     const { context, instruct } = templates;

public/scripts/chat-cemplates.js+19 -9 Viewed

 // the hash can be obtained from command line e.g. via: MODEL=path_to_model; python -c "import json, hashlib, sys; print(hashlib.sha256(json.load(open('"$MODEL"/tokenizer_config.json'))['chat_template'].encode()).hexdigest())"
 // note that chat templates must be trimmed to match the llama.cpp metadata value
 const derivationshash_derivations = {
     // Meta
     'e10ca381b1ccc5cf9db52e371f3b6651576caee0a630b452e2816b2d404d4b65':
         // Meta-Llama-3.1-8B-Instruct

+    ,
 };
 +const substr_derivations = {
 +    '<|im_start|>': 'ChatML', // qwen2.5, ...
 +}
++
 +const parse_derivation = derivation => (typeof derivation === 'string') ? {
 +    'context': derivation,
 +    'instruct': derivation,
 +} : derivation;
++
 export async function deriveTemplatesFromChatTemplate(chat_template, hash) {
     if (hash in derivationshash_derivations) {
         const derivation =return derivationsparse_derivation(hash_derivations[hash]);
 -        if (typeof derivation === 'string') {
 +    }
 -            return {
++
 -                'context': derivation,
 +    // heuristics
 -                'instruct': derivation,
 +    for (const [substr, derivation] of Object.entries(substr_derivations) ) {
 -            }
 +        if (chat_template.includes(substr)) {
 +            return parse_derivation(derivation);
+        }
 -        return derivation;
+    }
++
     console.log(`Unknown chat template hash: ${hash} for [${chat_template}]`);
     return null;
+}