-
Notifications
You must be signed in to change notification settings - Fork 8
/
Copy pathREADME.html
700 lines (522 loc) · 16.6 KB
/
README.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
<html>
<head>
<title>Doki Doki AI Edition - Readme</title>
<style>
body {
margin: 0;
width: 100%;
min-width: 320px;
min-height: 100vh;
line-height: 24px;
font-family: 'Arial';
font-size: 16px;
font-weight: 400;
color: white;
background-color: #1b1b1f;
font-synthesis: style;
text-rendering: optimizeLegibility;
-webkit-font-smoothing: antialiased;
}
section {
padding: 50px;
}
h1 {
font-size: 40px;
}
footer {
text-align: center;
color: rgba(235, 235, 245, 0.6);
border-top: 1px solid;
border-color: black;
padding: 50px;
}
p {
color: #cfcfcf
}
.container {
display: flex;
flex-direction: row;
max-width: 1152px;
margin: 0 auto;
}
.main {
display: flex;
flex-direction: column;
}
.name {
max-width: 392px;
letter-spacing: -0.4px;
line-height: 40px;
font-size: 32px;
font-weight: 700;
margin-bottom: 10px;
}
.clip {
background: -webkit-linear-gradient(120deg, #bd34fe 30%, #41d1ff);
-webkit-background-clip: text;
background-clip: text;
-webkit-text-fill-color: transparent
}
.text {
margin-top: 0px;
}
.tagline {
color: rgba(235, 235, 245, 0.6);
line-height: 36px;
font-size: 24px;
font-weight: 550;
margin-top: 0;
}
.actions {
display: flex;
flex-direction: row;
}
.action {
margin-right: 20px;
}
a{
display: inline-block;
border: 1px solid transparent;
text-align: center;
font-weight: 600;
white-space: nowrap;
text-decoration: none;
color: white;
transition: color 0.25s, border-color 0.25s, background-color 0.25s;
}
.started-btn {
background-color: #3e63dd;
border-color: #3e63dd;
color: white;
padding: 0 20px;
border-radius: 20px;
line-height: 38px;
font-size: 14px;
}
.github-btn {
background-color: #32363f;
border-color: #32363f;
color: white;
padding: 0 20px;
border-radius: 20px;
line-height: 38px;
font-size: 14px;
}
.image {
flex-grow: 1;
order: 2;
margin: 0;
min-height: 100%;
}
.image-container {
display: flex;
justify-content: center;
align-items: center;
width: 100%;
height: 100%;
transform: translate(-32px, -32px);
position: relative;
margin: 0 auto;
}
.image-bg {
background-image: linear-gradient(-45deg, #bd34fe 50%, #47caff 50%);
position: absolute;
top: 50%;
left: 50%;
border-radius: 50%;
width: 192px;
height: 192px;
filter: blur(68px);
transform: translate(-50%, -50%);
}
.image-src {
position: absolute;
top: 50%;
left: 50%;
max-width: 320px;
max-height: 320px;
transform: translate(-50%, -50%);
}
.main-option {
background-color: #202127;
height: 150px;
border-radius: 20px;
width: 240px;
padding: 20px;
margin-right: 20px;
}
.main-opt-inside {
display: flex;
flex-direction: column;
}
.icon-img {
background-color: rgba(101, 117, 133, 0.16);
width: 50px;
height: 50px;
display: flex;
justify-content: center;
align-items: center;
margin-bottom: 20px;
border-radius: 6px;
font-size: 24px;
}
.main-opt-inside span{
font-weight: 600;
margin-bottom: 10px;
}
.main-opt-inside a{
font-weight: 200;
color: rgba(235, 235, 245, 0.6);
text-align: left;
}
.l-break {
background-color: #646cff;
height: 1px;
}
.note-container {
background-color: rgba(100, 108, 255, .08);
border-color: #646cff;
border: 1px solid;
display: flex;
flex-direction: column;
color: #bcc0ff;
border-radius: 25px;
width: 65%;
padding: 10px;
font-family: math;
}
.note-container h1 {
font-size: small;
}
.note-container-red {
background-color: rgba(255, 100, 100, 0.08);
border-color: #ff6464;
border: 1px solid;
display: flex;
flex-direction: column;
color: #ffbcbc;
border-radius: 25px;
width: 65%;
padding: 10px;
font-family: system-ui;
}
.note-container-red h1 {
font-size: small;
}
@media (min-width: 960px) {
.container {
flex-direction: row;
}
.main {
max-width: 592px;
}
.name {
line-height: 64px;
font-size: 56px;
}
}
@media (min-width: 640px) {
.name, .text {
max-width: 576px;
line-height: 56px;
font-size: 48px;
}
}
</style>
</head>
<body>
<section>
<div class="container">
<div class="main">
<h1 class="name">
<span class="clip">Doki Doki AI Edition</span>
</h1>
<p class="name text">Experience a unique playthrough</p>
<p class="tagline">Setup your own version of DDAE with ease!</p>
<div class="actions">
<div class="action"><a class="started-btn" href="#setup">Get Started</a></div>
<div class="action"><a class="github-btn" href="https://github.com/doki-doki-ai-edition" target="_blank">View on GitHub</a></div>
</div>
</div>
<div class="image">
<div class="image-container">
<div class="image-bg"></div>
<img class="image-src" src="game/assets/imgs/gui/logo.png">
</div>
</div>
</div>
</section>
<section>
<div class="container">
<div class="main-option">
<div class="main-opt-inside">
<div class="icon-img">
<img style="width: 100%;" src="game/assets/imgs/gui/logo.png" alt="DDAE Logo">
</div>
<span>Mod</span>
<a href="#setup">Getting Started</a>
</div>
</div>
<div class="main-option">
<div class="main-opt-inside">
<div class="icon-img">
<span style="margin-bottom: 1px;" >🎲</span>
</div>
<span>Settings</span>
<a href="#settings">Examples</a>
</div>
</div>
<div class="main-option">
<div class="main-opt-inside">
<div class="icon-img">
<span style="margin-bottom: 1px;">❓</span>
</div>
<span>Frequently Asked Questions</span>
<a href="#faq">FAQ</a>
</div>
</div>
<div class="main-option">
<div class="main-opt-inside">
<div class="icon-img">
<span>🧰</span>
</div>
<span>Helpful Resource</span>
<a target="_blank" href="https://discord.gg/rDA7ehBSq7">Support Server</a>
</div>
</div>
</div>
</section>
<section style="display: flex; justify-content: center;">
<span>
HTML design heavily referenced from <a target="_blank" href="https://vitejs.dev/">https://vitejs.dev/</a>
</span>
</section>
<section>
<div class="container" style="flex-direction: column;">
<h1 id="setup" >Setup</h1>
<ul>
<li>Make sure you have a fresh install of DDLC <a href="https://ddlc.moe/">https://ddlc.moe/</a></li>
<li>Download the latest version of the DDAE mod on this github page <a href="https://github.com/doki-doki-ai-edition/Mod/releases">https://github.com/doki-doki-ai-edition/Mod/releases</a></li>
</ul>
<h2 style="color: #646cff">Ollama</h2>
<p>
<b>Overview:</b> Ollama is an open-source project that allows people to setup local language models on their machine in a far more user friendly way. They have both uncensored and censored models and the models are quantized (compressed) to run on low-end devices.
</p>
<p>
<b>Pros</b>
</p>
<ul>
<li>Run models locally on your machine</li>
<li>Easy to setup</li>
<li>Privacy: Even if your internet is off, it'll still work</li>
<li>Uncensored: Freedom to say whatever</li>
</ul>
<p>
<b>Cons</b>
</p>
<ul>
<li>Very Slow (depending on your hardware)</li>
<li>Quality isn't that great especially compared to what some APIs offer</li>
</ul>
<p>
<b>Let's Start!</b>
</p>
<ul>
<li>Go to <a target="_blank" href="https://ollama.com/download">https://ollama.com/download</a></li>
<li>Select Windows (or whichever operating system you're using)</li>
<li>Click on Download</li>
</ul>
<p>
Once it's downloaded, run the application.
If you're on windows you should see a small icon in your taskbar that looks like this
</p>
<img style="width: 80px;" src="game/assets/imgs/help_page/llama_task.png" alt="llama taskbar">
<ul>
<li>Open up a terminal on your device (eg. cmd+r if you're using windows, then type "cmd" and press enter)</li>
<li>In the terminal type "ollama pull llama3.1" (or any model you find in https://ollama.com/library/)</li>
</ul>
<span style="color: #64ff6e">
You're Done!
</span>
<p>
This should be roughly 4.7 gigabytes and once it's done installing you can run the mod!
</p>
</div>
</section>
<section>
<div class="container" style="flex-direction: column;">
<h1 id="settings">Settings</h1>
<div>
<h2>AI Type</h2>
<span>
<b>Local LLM (Large Language Model)</b>
</span>
<p>
Shows the default local models available in the "Model Name" section.
</p>
</div>
<div class="l-break"></div>
<div>
<h2>Model Name</h2>
<span>
<b>Suggested Models</b>
</span>
<p>
Models that have been tested to work properly in accordance to the default prompt given
</p>
<span>
<b>Other Models</b>
</span>
<p>
Models that may not be as high quality or haven't been frequently tested
</p>
</div>
<div class="l-break"></div>
<div>
<h2>Model Config</h2>
<div>
<span>
<b>Context Window</b>
</span>
<p>
The maximum amount of tokens (messages) that will be remembered.
</p>
<p>
Note that the more context you give it, the slower responses may be and the more memory you use on your device.
</p>
<p>
Make sure to check the context window of a model. For example:
</p>
<p>
Go to <a target="_blank" href="https://ollama.com/library/llama3:8b">https://ollama.com/library/llama3:8b</a>
</p>
<img src="game/assets/imgs/help_page/model_show.png" alt="model display">
<p>
Click on "model" (and wait since it may take a while to load)
</p>
<p>
Look for context_length
</p>
<img src="game/assets/imgs/help_page/model_context_len.png" alt="model context window">
<p>
If you want to use more than a 8192 token limit then you can check out other models such as
</p>
<a target="_blank" href="https://ollama.com/library/llama3-gradient">https://ollama.com/library/llama3-gradient</a>
<p>
which has a 1M context window.
You won't be able to use the full 1M tokens because of the hardware requirement for something like that but depending on your hardware you could use a lot more than 8k.
</p>
</div>
<div>
<span>
<b>Temperature</b>
</span>
<p>
Determines how random/creative the output is.
</p>
<p>
Higher = More random and creative but if it's too high, the responses can be nonsensical and may not even follow the system prompt appropriately.
</p>
<p>
Lower = Less random and predictable. The model chooses the most likely word to respond with but ends up sounding more robotic.
</p>
<p>
The typical default temperature is "1" but I've personally found 0.6 to be a good sweet spot, just high enough to be creative but low enough to give good responses.
</p>
</div>
<div style="margin-bottom: 15px;" class="note-container-red">
<h1><b>WARNING</b></h1>
<span>It is highly advisable to leave the temperature on the default settings because it has been shown to follow the prompt far more accurately when testing compared to other numbers.</span>
</div>
<div>
<span>
<b>Seed</b>
</span>
<p>
Deterministic responses. If set to a specific seed, your outputs will be pretty much exactly the same or relatively similar if you feed it the same input.
</p>
<p>
Eg. If you have a seed of 22 and your input response goes like this:
</p>
<p>
user: "how are you monika?"<br>
monika: "I'm not doing too well to be honest..."
</p>
<p>
Assuming the system prompt is the same and your response is formatted in the exact same way, you'll always get that same output (or something very similar).
</p>
</div>
<div>
<span>
<b>Default Prompt</b>
</span>
<p>
In game/assets/prompts there is a prompt_templates.json file that stores all the default prompts for every character.
</p>
<p>
You can edit the backstory to your liking.
</p>
</div>
</div>
</div>
</section>
<section>
<div class="container" style="flex-direction: column;">
<h1 id="faq">Frequently Asked Questions</h1>
<div>
<h2>Are my chats private?</h2>
<p>
Yes. Your chats are completely localized on your machine, even if you turn your internet off, it'll still work.
</p>
<h2>Are the models uncensored</h2>
<p>
It depends on what model you're using.
You can go to <a href="https://ollama.com/library ">https://ollama.com/library</a> to find uncensored models but they may not follow the system prompt as well which would cause the game to not function correctly.
</p>
<p>
Secondly, they may not be entirely uncensored. There are certain questions that it may refuse, but since it's a local model you can attempt to jailbreak it as much as you want.
</p>
<h2>Can I trust Ollama?</h2>
<p>
As much as you trust all the other apps you use daily.
Jeffrey Morgan is one of the founders and he has a credible background by previously working for docker, twitter and google.
The project is also used by google in firebase <a target="_blank" href="https://firebase.google.com/docs/genkit/plugins/ollama">https://firebase.google.com/docs/genkit/plugins/ollama</a>
And obviously, the project is open-sourced.
</p>
<h2>Why do the models sometime return "ERROR"?</h2>
<p>
This happens when the model doesn't respond with the instructed format. This typically happens on regular models if you say something too provacative.
There's a few solutions to this.
The first is to use an uncensored model.
The second thing you could try is continuing the chat by ignoring the error message and try to force the ai back into a roleplay state
And the 3rd thing you could try is just jailbreaking the ai model by editing the
system prompt to make the ai say anything.
</p>
<h2>Why does the response keep loading?</h2>
<p>
If your hardware isn't that good then it could likely mean that it's just a hardware limitation. HOWEVER. If it's taking abnormally long
check the log.txt file, there may be an error in there which would have cauesed the game to be in a constant "Loading..." state.
Post your log.txt file in the discord #support channel
</p>
<h2>What hardware do I need?</h2>
<p>
You should be able to run the mod on low-end devices like a laptop with 4GB of RAM
and 4GB of VRAM but the speed at which you get a response is going to be slow.
If you want faster responses you need more VRAM. Regular RAM allows you to run models.
</p>
<h2>I keep getting this error</h2>
<img src="game/assets/imgs/help_page/traceback.png" alt="traceback error">
<p>
When in doubt, restart.
This typically means the model didn't properly generate a background when you first loaded the game.
To fix this you can completely close the game and try loading a new one.
You could also edit the chat file for the old world and try loading it again.
</p>
</div>
</div>
</section>
<footer>
<span> <b>Not affiliated with Team Salvato. Please support the official release <a href="https://ddlc.moe">https://ddlc.moe</a></b></span>
</footer>
</body>
</html>