152 commits

This branch

This branch

All branches

Author	SHA1	Message	Date
niansa	94e4ca5874	Set default context size limit	2023-04-27 09:47:23 +02:00
niansa	ca4ad5f096	Added context window scrolling with top bar	2023-04-27 09:45:37 +02:00
niansa	219186f4b6	Take const string reference instead of string view in append()	2023-04-27 09:31:22 +02:00
niansa	4e74517bb5	Changed parameter types to some that make more sense	2023-04-27 09:27:09 +02:00
niansa	ea5c5fe7f4	Minor fixes	2023-04-27 09:00:21 +02:00
niansa	bfcda5a3bf	Merge branch 'devel' into 'master' ... Devel See merge request niansa/libjustlm!1	2023-04-27 06:23:45 +00:00
niansa	7aca184dba	Updated llama.cpp	2023-04-27 08:22:51 +02:00
niansa	0661b2e33d	Updated for latest llama.cpp and working gpt-j implementation	2023-04-27 08:21:02 +02:00
niansa	566f8227fd	Should be functional now	2023-04-27 08:00:08 +02:00
niansa	316e8cbf18	Initial GPT-J support	2023-04-26 16:32:45 +02:00
niansa	5f6cf17871	Moved pool functions into seperate file	2023-04-26 11:18:12 +02:00
niansa	1f75673523	Fixed compilation	2023-04-26 11:10:25 +02:00
niansa	aad1bd9ae4	Made Inference class virtual	2023-04-26 10:59:24 +02:00
niansa	55a310b005	Synced top_p default with llama.cpp	2023-04-25 17:08:55 +02:00
niansa	00d72f59f6	Ignore EOS properly	2023-04-25 16:21:19 +02:00
niansa	d88dc5ad98	Fixed InferencePool::store_all() storing empty slots	2023-04-25 15:13:57 +02:00
niansa	97f7f2ebc5	Removed debugging left-over	2023-04-23 18:24:42 +02:00
niansa	138a9bde52	Minor fixes and Python bindings	2023-04-23 18:06:57 +02:00
niansa	7ccdc483ad	Serialize/Deserialize entire state	2023-04-23 16:28:32 +02:00
niansa	0466774286	Fully implemented InferencePool	2023-04-23 15:31:16 +02:00
niansa	414554c69a	Use ios::binary for file streams	2023-04-23 12:40:29 +02:00
niansa	cf2fec84e0	Added experimental InferencePool	2023-04-23 12:31:39 +02:00
niansa	2ec882f4d0	Added serialize()/deserialize() functions	2023-04-23 10:59:55 +02:00
niansa	d09f892120	Updated llama.cpp	2023-04-22 15:34:58 +02:00
niansa	e85556d94d	Added Inference::get_prompt	2023-04-20 21:34:20 +02:00
niansa	677ac470e7	Load/Restore prompt	2023-04-20 21:32:45 +02:00
niansa	06d99a6950	Allow generation without end string and fixed Python bug	2023-04-18 01:17:32 +02:00
niansa	4c9a3a308b	Updated llama.cpp	2023-04-17 23:21:54 +02:00
niansa	f57a729853	Fixed savestates	2023-04-17 23:08:05 +02:00
niansa	f02a189361	Removed debug code	2023-04-17 23:02:09 +02:00
niansa	80d33df458	Added is_valid to Savestate	2023-04-17 22:55:28 +02:00
niansa	139935adb2	Removed gpt2, updated python binding and added "savestates"	2023-04-17 22:46:45 +02:00
niansa	2d97e7b2bd	Updated llama.cpp	2023-04-16 18:02:29 +02:00
niansa	cd429cb3a7	Batch size isn't unused anymore	2023-04-05 19:44:22 +02:00
niansa	6444d30866	No repeat penalty by default	2023-04-05 19:43:13 +02:00
niansa	97d94ea2c9	Updated llama.cpp and python bindings	2023-04-05 19:35:41 +02:00
niansa	e60b1843d0	Allow ignoring EOS x times	2023-04-03 23:25:43 +02:00
niansa	9dc64fbc66	Respect eos and prevent fres cut-off when aborted.	2023-04-03 22:47:12 +02:00
niansa	8b5a375f59	Updated llama.cpp	2023-04-03 10:15:15 +02:00
niansa	7ae1547e5f	Fixed batch processing	2023-04-02 15:29:46 +02:00
niansa	a9a0cce68e	Implement batch evaluation	2023-04-02 15:09:40 +02:00
niansa	51cbd41732	Improved parameter passing	2023-04-01 17:10:03 +02:00
niansa	8ccd11368f	Addded use_mlock	2023-04-01 15:11:16 +02:00
niansa	dc7fe7f9f0	Added header back to CMakeLists.txt	2023-04-01 14:51:14 +02:00
niansa	14410be6e5	Updated llama.cpp	2023-04-01 14:48:42 +02:00
niansa	dc9ab53d01	implemented repeat penalty	2023-04-01 14:48:42 +02:00
niansa	f968b2cd11	Made Inference move-only	2023-03-31 20:52:25 +02:00
niansa	d6f55d0b4f	Simplified project structure	2023-03-31 20:37:45 +02:00
niansa	a13ac36d97	Made params public	2023-03-30 08:25:59 -05:00
niansa	5f4a0b9a92	Moved into namespace and added python bindings	2023-03-30 07:52:53 -05:00