vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

tokenizer tweaks (+2 squashed commit)

Browse source 
Squashed commit:

[18c70621] tokenizer tweaks

[8002f897] handle if localstorage is inaccessible
This commit is contained in:
Concedo 2023-11-06 21:45:26 +08:00
parent 372cfef2c3
commit feb60bc447
1 changed files with 33 additions and 23 deletions
Download patch file Download diff file Expand all files Collapse all files

20
klite.embd
View file

@ -6,7 +6,7 @@ It requires no dependencies, installation or setup.
Just copy this single static HTML file anywhere and open it in a browser, or from a webserver.
Please go to https://github.com/LostRuins/lite.koboldai.net for updates on Kobold Lite.
Kobold Lite is under the AGPL v3.0 License unless otherwise exempted. Please do not remove this line.
Current version: 91
Current version: 92
-Concedo
-->
@ -6282,6 +6282,8 @@ Current version: 91
</div>`;
filetable += entry;
try
{
for(let i=0;i<4;++i)
{
let testslot = localStorage.getItem(STORAGE_PREFIX + "slot_"+i+"_meta");
@ -6297,6 +6299,9 @@ Current version: 91
</div></div>`;
filetable += entry;
}
} catch (e) {
console.log("get slots failed: " + e);
}
document.getElementById("saveloadentries").innerHTML = filetable;
}
function save_to_slot(slot)
@ -7346,7 +7351,7 @@ Current version: 91
let onOk = ()=>{
pending_response_id = "-1";
waiting_for_autosummary = true;
let max_allowed_characters = Math.floor(localsettings.max_context_length * 3.2)-100;
let max_allowed_characters = Math.floor(localsettings.max_context_length * 3.0)-100;
let truncated_context = concat_gametext(true, "");
let max_mem_anote_len = Math.floor(max_allowed_characters*0.9);
@ -7916,7 +7921,7 @@ Current version: 91
truncated_context = truncated_context.replace(/\xA0/g,' '); //replace non breaking space nbsp
//this is a hack since we dont have a proper tokenizer, but we can estimate 1 token per 3 characters
let chars_per_token = 3.2;
let chars_per_token = 3.0;
//we try to detect attempts at coding which tokenize poorly. This usually happens when the average word length is high.
let avgwordlen = (1.0+truncated_context.length)/(1.0+countWords(truncated_context));
if(avgwordlen>=7.8)
@ -7928,7 +7933,7 @@ Current version: 91
//if there is no memory, then we can be a lot of lenient with the character counts since the backend will truncate excess anyway
chars_per_token = 4.8;
}
let max_allowed_characters = Math.max(1, Math.floor((maxctxlen-maxgenamt) * chars_per_token) - 8);
let max_allowed_characters = Math.max(1, Math.floor((maxctxlen-maxgenamt) * chars_per_token) - 12);
//for adventure mode, inject hidden context, even more if there's nothing in memory
if (localsettings.opmode == 2 && localsettings.adventure_context_mod)
@ -8056,7 +8061,7 @@ Current version: 91
//we clip the memory if its too long, taking the last x chars (not the first)
//memory or anote is allowed to be up to 0.9 times of ctx allowance
let max_mem_anote_len = Math.floor(max_allowed_characters*0.9);
let truncated_memory = current_memory.substring(current_memory.length - max_mem_anote_len);
let truncated_memory = substring_to_boundary(current_memory, max_mem_anote_len);
if (truncated_memory != null && truncated_memory != "") {
if(newlineaftermemory)
{
@ -10187,12 +10192,17 @@ Current version: 91
function autosave() {
//autosave
try {
localStorage.setItem(STORAGE_PREFIX + "settings", JSON.stringify(localsettings));
if (localsettings.persist_session) {
let compressedstory = generate_compressed_story(true, true, true);
localStorage.setItem(STORAGE_PREFIX + "story", compressedstory);
}
console.log("autosave done");
} catch (e) {
console.log("autosave failed: " + e);
}
}
function btn_adventure_mode()