1242 Commits

Author	SHA1	Message	Date
oobabooga	68059d7c23	llama.cpp: minor log change & lint	2023-11-27 10:44:55 -08:00
tsukanov-as	9f7ae6bb2e	fix detection of stopping strings when HTML escaping is used (#4728)	2023-11-27 15:42:08 -03:00
oobabooga	0589ff5b12	Bump llama-cpp-python to 0.2.19 & add min_p and typical_p parameters to llama.cpp loader (#4701)	2023-11-21 20:59:39 -03:00
oobabooga	2769a1fa25	Hide deprecated args from Session tab	2023-11-21 15:15:16 -08:00
oobabooga	a2e6d00128	Use convert_ids_to_tokens instead of decode in logits endpoint ... This preserves the llama tokenizer spaces.	2023-11-19 09:22:08 -08:00
oobabooga	9da7bb203d	Minor LoRA bug fix	2023-11-19 07:59:29 -08:00
oobabooga	a6f1e1bcc5	Fix PEFT LoRA unloading	2023-11-19 07:55:25 -08:00
oobabooga	ab94f0d9bf	Minor style change	2023-11-18 21:11:04 -08:00
oobabooga	5fcee696ea	New feature: enlarge character pictures on click (#4654)	2023-11-19 02:05:17 -03:00
oobabooga	ef6feedeb2	Add --nowebui flag for pure API mode (#4651)	2023-11-18 23:38:39 -03:00
oobabooga	0fa1af296c	Add /v1/internal/logits endpoint (#4650)	2023-11-18 23:19:31 -03:00
oobabooga	8f4f4daf8b	Add --admin-key flag for API (#4649)	2023-11-18 22:33:27 -03:00
Jordan Tucker	baab894759	fix: use system message in chat-instruct mode (#4648)	2023-11-18 20:20:13 -03:00
oobabooga	47d9e2618b	Refresh the Preset menu after saving a preset	2023-11-18 14:03:42 -08:00
oobabooga	83b64e7fc1	New feature: "random preset" button (#4647)	2023-11-18 18:31:41 -03:00
oobabooga	e0ca49ed9c	Bump llama-cpp-python to 0.2.18 (2nd attempt) (#4637) ... * Update requirements*.txt * Add back seed	2023-11-18 00:31:27 -03:00
oobabooga	9d6f79db74	Revert "Bump llama-cpp-python to 0.2.18 (#4611)" ... This reverts commit 923c8e25fb.	2023-11-17 05:14:25 -08:00
oobabooga	13dc3b61da	Update README	2023-11-16 19:57:55 -08:00
oobabooga	8b66d83aa9	Set use_fast=True by default, create --no_use_fast flag ... This increases tokens/second for HF loaders.	2023-11-16 19:55:28 -08:00
oobabooga	6525707a7f	Fix "send instruction template to..." buttons (closes #4625)	2023-11-16 18:16:42 -08:00
oobabooga	510a01ef46	Lint	2023-11-16 18:03:06 -08:00
oobabooga	923c8e25fb	Bump llama-cpp-python to 0.2.18 (#4611)	2023-11-16 22:55:14 -03:00
oobabooga	58c6001be9	Add missing exllamav2 samplers	2023-11-16 07:09:40 -08:00
oobabooga	cd41f8912b	Warn users about n_ctx / max_seq_len	2023-11-15 18:56:42 -08:00
oobabooga	9be48e83a9	Start API when "api" checkbox is checked	2023-11-15 16:35:47 -08:00
oobabooga	a85ce5f055	Add more info messages for truncation / instruction template	2023-11-15 16:20:31 -08:00
oobabooga	883701bc40	Alternative solution to 025da386a0 ... Fixes an error.	2023-11-15 16:04:02 -08:00
oobabooga	8ac942813c	Revert "Fix CPU memory limit error (issue #3763) (#4597)" ... This reverts commit 025da386a0.	2023-11-15 16:01:54 -08:00
oobabooga	e6f44d6d19	Print context length / instruction template to terminal when loading models	2023-11-15 16:00:51 -08:00
oobabooga	e05d8fd441	Style changes	2023-11-15 15:51:37 -08:00
Andy Bao	025da386a0	Fix CPU memory limit error (issue #3763) (#4597) ... get_max_memory_dict() was not properly formatting shared.args.cpu_memory Co-authored-by: oobabooga <112222186+oobabooga@users.noreply.github.com>	2023-11-15 20:27:20 -03:00
oobabooga	4aabff3728	Remove old API, launch OpenAI API with --api	2023-11-10 06:39:08 -08:00
oobabooga	2af7e382b1	Revert "Bump llama-cpp-python to 0.2.14" ... This reverts commit 5c3eb22ce6. The new version has issues: https://github.com/oobabooga/text-generation-webui/issues/4540 https://github.com/abetlen/llama-cpp-python/issues/893	2023-11-09 10:02:13 -08:00
oobabooga	21ed9a260e	Document the new "Custom system message" field	2023-11-08 17:54:10 -08:00
oobabooga	2358706453	Add /v1/internal/model/load endpoint (tentative)	2023-11-07 20:58:06 -08:00
oobabooga	43c53a7820	Refactor the /v1/models endpoint	2023-11-07 19:59:27 -08:00
oobabooga	1b69694fe9	Add types to the encode/decode/token-count endpoints	2023-11-07 19:32:14 -08:00
oobabooga	6e2e0317af	Separate context and system message in instruction formats (#4499)	2023-11-07 20:02:58 -03:00
oobabooga	5c0559da69	Training: fix .txt files now showing in dropdowns	2023-11-07 14:41:11 -08:00
oobabooga	af3d25a503	Disable logits_all in llamacpp_HF (makes processing 3x faster)	2023-11-07 14:35:48 -08:00
oobabooga	5c3eb22ce6	Bump llama-cpp-python to 0.2.14	2023-11-07 14:20:43 -08:00
oobabooga	ec17a5d2b7	Make OpenAI API the default API (#4430)	2023-11-06 02:38:29 -03:00
feng lui	4766a57352	transformers: add use_flash_attention_2 option (#4373)	2023-11-04 13:59:33 -03:00
wouter van der plas	add359379e	fixed two links in the ui (#4452)	2023-11-04 13:41:42 -03:00
oobabooga	aa5d671579	Add temperature_last parameter (#4472)	2023-11-04 13:09:07 -03:00
oobabooga	1ab8700d94	Change frequency/presence penalty ranges	2023-11-03 17:38:19 -07:00
oobabooga	45fcb60e7a	Make truncation_length_max apply to max_seq_len/n_ctx	2023-11-03 11:29:31 -07:00
oobabooga	7f9c1cbb30	Change min_p default to 0.0	2023-11-03 08:25:22 -07:00
oobabooga	4537853e2c	Change min_p default to 1.0	2023-11-03 08:13:50 -07:00
kalomaze	367e5e6e43	Implement Min P as a sampler option in HF loaders (#4449)	2023-11-02 16:32:51 -03:00
oobabooga	fcb7017b7a	Remove a checkbox	2023-11-02 12:24:09 -07:00
Julien Chaumond	fdcaa955e3	transformers: Add a flag to force load from safetensors (#4450)	2023-11-02 16:20:54 -03:00
oobabooga	c0655475ae	Add cache_8bit option	2023-11-02 11:23:04 -07:00
oobabooga	42f816312d	Merge remote-tracking branch 'refs/remotes/origin/dev' into dev	2023-11-02 11:09:26 -07:00
oobabooga	77abd9b69b	Add no_flash_attn option	2023-11-02 11:08:53 -07:00
Julien Chaumond	a56ef2a942	make torch.load a bit safer (#4448)	2023-11-02 14:07:08 -03:00
Mehran Ziadloo	aaf726dbfb	Updating the shared settings object when loading a model (#4425)	2023-11-01 01:29:57 -03:00
oobabooga	9bd0724d85	Change frequency/presence penalty ranges	2023-10-31 20:57:56 -07:00
Meheret	0707ed7677	updated wiki link (#4415)	2023-10-31 19:09:05 -03:00
oobabooga	262f8ae5bb	Use default gr.Dataframe for evaluation table	2023-10-27 06:49:14 -07:00
oobabooga	839a87bac8	Fix is_ccl_available & is_xpu_available imports	2023-10-26 20:27:04 -07:00
Abhilash Majumder	778a010df8	Intel Gpu support initialization (#4340)	2023-10-26 23:39:51 -03:00
oobabooga	92b2f57095	Minor metadata bug fix (second attempt)	2023-10-26 18:57:32 -07:00
tdrussell	72f6fc6923	Rename additive_repetition_penalty to presence_penalty, add frequency_penalty (#4376)	2023-10-25 12:10:28 -03:00
oobabooga	ef1489cd4d	Remove unused parameter in AutoAWQ	2023-10-23 20:45:43 -07:00
oobabooga	1edf321362	Lint	2023-10-23 13:09:03 -07:00
oobabooga	280ae720d7	Organize	2023-10-23 13:07:17 -07:00
oobabooga	49e5eecce4	Merge remote-tracking branch 'refs/remotes/origin/main'	2023-10-23 12:54:05 -07:00
oobabooga	306d764ff6	Minor metadata bug fix	2023-10-23 12:46:24 -07:00
adrianfiedler	4bc411332f	Fix broken links (#4367) ... --------- Co-authored-by: oobabooga <112222186+oobabooga@users.noreply.github.com>	2023-10-23 14:09:57 -03:00
oobabooga	92691ee626	Disable trust_remote_code by default	2023-10-23 09:57:44 -07:00
tdrussell	4440f87722	Add additive_repetition_penalty sampler setting. (#3627)	2023-10-23 02:28:07 -03:00
oobabooga	df90d03e0b	Replace --mul_mat_q with --no_mul_mat_q	2023-10-22 12:23:03 -07:00
Googulator	d0c3b407b3	transformers loader: multi-LoRAs support (#3120)	2023-10-22 16:06:22 -03:00
omo	4405513ca5	Option to select/target additional linear modules/layers in LORA training (#4178)	2023-10-22 15:57:19 -03:00
oobabooga	2d1b3332e4	Ignore warnings on Colab	2023-10-21 21:45:25 -07:00
oobabooga	09f807af83	Use ExLlama_HF for GPTQ models by default	2023-10-21 20:45:38 -07:00
oobabooga	506d05aede	Organize command-line arguments	2023-10-21 18:52:59 -07:00
oobabooga	fbac6d21ca	Add missing exception	2023-10-20 23:53:24 -07:00
Brian Dashore	3345da2ea4	Add flash-attention 2 for windows (#4235)	2023-10-21 03:46:23 -03:00
Johan	1d5a015ce7	Enable special token support for exllamav2 (#4314)	2023-10-21 01:54:06 -03:00
turboderp	ae8cd449ae	ExLlamav2_HF: Convert logits to FP32 (#4310)	2023-10-18 23:16:05 -03:00
oobabooga	f17f7a6913	Increase the evaluation table height	2023-10-16 12:55:35 -07:00
oobabooga	8ea554bc19	Check for torch.xpu.is_available()	2023-10-16 12:53:40 -07:00
oobabooga	188d20e9e5	Reduce the evaluation table height	2023-10-16 10:53:42 -07:00
oobabooga	2d44adbb76	Clear the torch cache while evaluating	2023-10-16 10:52:50 -07:00
oobabooga	71cac7a1b2	Increase the height of the evaluation table	2023-10-15 21:56:40 -07:00
oobabooga	e14bde4946	Minor improvements to evaluation logs	2023-10-15 20:51:43 -07:00
oobabooga	b88b2b74a6	Experimental Intel Arc transformers support (untested)	2023-10-15 20:51:11 -07:00
Forkoz	8cce1f1126	Exllamav2 lora support (#4229) ... --------- Co-authored-by: oobabooga <112222186+oobabooga@users.noreply.github.com>	2023-10-14 16:12:41 -03:00
oobabooga	773c17faec	Fix a warning	2023-10-10 20:53:38 -07:00
oobabooga	f63361568c	Fix safetensors kwarg usage in AutoAWQ	2023-10-10 19:03:09 -07:00
oobabooga	39f16ff83d	Fix default/notebook tabs css	2023-10-10 18:45:12 -07:00
oobabooga	fae8062d39	Bump to latest gradio (3.47) (#4258)	2023-10-10 22:20:49 -03:00
oobabooga	9fab9a1ca6	Minor fix	2023-10-10 14:08:11 -07:00
oobabooga	a49cc69a4a	Ignore rope_freq_base if value is 10000	2023-10-10 13:57:40 -07:00
oobabooga	3a9d90c3a1	Download models with 4 threads by default	2023-10-10 13:52:10 -07:00
Forkoz	35695e18c7	Remove import. (#4247) ... For real this time.	2023-10-09 18:06:11 -03:00
Forkoz	2e471071af	Update llama_attn_hijack.py (#4231)	2023-10-08 15:16:48 -03:00
Brian Dashore	98fa73a974	Text Generation: stop if EOS token is reached (#4213)	2023-10-07 19:46:42 -03:00
Brian Dashore	7743b5e9de	Llamacpp_HF: Fix CFG cache init (#4219) ... Documentation says that model.context_params should be sent when a new context is created. The current code uses model.params which doesn't exist. Signed-off-by: kingbri <bdashore3@proton.me>	2023-10-07 19:38:29 -03:00
turboderp	8a98646a21	Bump ExLlamaV2 to 0.0.5 (#4186)	2023-10-05 19:12:22 -03:00
oobabooga	7ffb424c7b	Add AutoAWQ to README	2023-10-05 09:22:37 -07:00
cal066	cc632c3f33	AutoAWQ: initial support (#3999)	2023-10-05 13:19:18 -03:00
tdrussell	cb26163a20	Fix off-by-one error in exllama_hf caching logic (#4145)	2023-10-05 12:20:56 -03:00
oobabooga	ae4ba3007f	Add grammar to transformers and _HF loaders (#4091)	2023-10-05 10:01:36 -03:00
oobabooga	b6fe6acf88	Add threads_batch parameter	2023-10-01 21:28:00 -07:00
jllllll	41a2de96e5	Bump llama-cpp-python to 0.2.11	2023-10-01 18:08:10 -05:00
oobabooga	f2d82f731a	Add recommended NTKv1 alpha values	2023-09-29 13:48:38 -07:00
oobabooga	abe99cddeb	Extend evaluation slider bounds	2023-09-29 13:06:26 -07:00
oobabooga	96da2e1c0d	Read more metadata (config.json & quantize_config.json)	2023-09-29 06:14:16 -07:00
oobabooga	56b5a4af74	exllamav2 typical_p	2023-09-28 20:10:12 -07:00
oobabooga	f8e9733412	Minor syntax change	2023-09-28 19:32:35 -07:00
oobabooga	f931184b53	Increase truncation limits to 32768	2023-09-28 19:28:22 -07:00
oobabooga	1dd13e4643	Read Transformers config.json metadata	2023-09-28 19:19:47 -07:00
StoyanStAtanasov	7e6ff8d1f0	Enable NUMA feature for llama_cpp_python (#4040)	2023-09-26 22:05:00 -03:00
oobabooga	87ea2d96fd	Add a note about RWKV loader	2023-09-26 17:43:39 -07:00
oobabooga	0c89180966	Another minor fix	2023-09-26 06:54:21 -07:00
oobabooga	365335e1ae	Minor fix	2023-09-26 06:47:19 -07:00
oobabooga	1ca54faaf0	Improve --multi-user mode	2023-09-26 06:42:33 -07:00
oobabooga	019371c0b6	Lint	2023-09-25 20:31:11 -07:00
oobabooga	814520fed1	Extension install improvements	2023-09-25 20:27:06 -07:00
oobabooga	7f1460af29	Change a warning	2023-09-25 20:22:27 -07:00
oobabooga	862b45b1c7	Extension install improvements	2023-09-25 19:48:30 -07:00
oobabooga	c8952cce55	Move documentation from UI to docs/	2023-09-25 12:28:28 -07:00
oobabooga	d0d221df49	Add --use_fast option (closes #3741)	2023-09-25 12:19:43 -07:00
oobabooga	b973b91d73	Automatically filter by loader (closes #4072)	2023-09-25 10:28:35 -07:00
oobabooga	63de9eb24f	Clean up the transformers loader	2023-09-24 20:26:26 -07:00
oobabooga	36c38d7561	Add disable_exllama to Transformers loader (for GPTQ LoRA training)	2023-09-24 20:03:11 -07:00
oobabooga	55a685d999	Minor fixes	2023-09-24 14:15:10 -07:00
oobabooga	08cf150c0c	Add a grammar editor to the UI (#4061)	2023-09-24 18:05:24 -03:00
oobabooga	eb0b7c1053	Fix a minor UI bug	2023-09-24 07:17:33 -07:00
oobabooga	3edac43426	Remove print statement	2023-09-24 07:13:00 -07:00
oobabooga	b227e65d86	Add grammar to llama.cpp loader (closes #4019)	2023-09-24 07:10:45 -07:00
oobabooga	2e7b6b0014	Create alternative requirements.txt with AMD and Metal wheels (#4052)	2023-09-24 09:58:29 -03:00
oobabooga	7a3ca2c68f	Better detect EXL2 models	2023-09-23 13:05:55 -07:00
oobabooga	b1467bd064	Move one-click-installers into the repository (#4028 from oobabooga/one-click)	2023-09-22 17:43:07 -03:00
oobabooga	c075969875	Add instructions	2023-09-22 13:10:03 -07:00
oobabooga	8ab3eca9ec	Add a warning for outdated installations	2023-09-22 09:35:19 -07:00
oobabooga	95976a9d4f	Fix a bug while deleting characters	2023-09-22 06:02:34 -07:00
oobabooga	d5330406fa	Add a rename menu for chat histories	2023-09-21 19:16:51 -07:00
oobabooga	00ab450c13	Multiple histories for each character (#4022)	2023-09-21 17:19:32 -03:00
oobabooga	029da9563f	Avoid redundant function call in llamacpp_hf	2023-09-19 14:14:40 -07:00
oobabooga	869f47fff9	Lint	2023-09-19 13:51:57 -07:00
oobabooga	13ac55fa18	Reorder some functions	2023-09-19 13:51:57 -07:00
oobabooga	03dc69edc5	ExLlama_HF (v1 and v2) prefix matching	2023-09-19 13:12:19 -07:00
oobabooga	5075087461	Fix command-line arguments being ignored	2023-09-19 13:11:46 -07:00
oobabooga	ff5d3d2d09	Add missing import	2023-09-18 16:26:54 -07:00
oobabooga	605ec3c9f2	Add a warning about ExLlamaV2 without flash-attn	2023-09-18 12:26:35 -07:00
oobabooga	f0ef971edb	Remove obsolete warning	2023-09-18 12:25:10 -07:00
oobabooga	745807dc03	Faster llamacpp_HF prefix matching	2023-09-18 11:02:45 -07:00
BadisG	893a72a1c5	Stop generation immediately when using "Maximum tokens/second" (#3952) ... --------- Co-authored-by: oobabooga <112222186+oobabooga@users.noreply.github.com>	2023-09-18 14:27:06 -03:00
Cebtenzzre	8466cf229a	llama.cpp: fix ban_eos_token (#3987)	2023-09-18 12:15:02 -03:00
oobabooga	0ede2965d5	Remove an error message	2023-09-17 18:46:08 -07:00
missionfloyd	cc8eda298a	Move hover menu shortcuts to right side (#3951)	2023-09-17 22:33:00 -03:00
oobabooga	280cca9f66	Merge remote-tracking branch 'refs/remotes/origin/main'	2023-09-17 18:01:27 -07:00
oobabooga	b062d50c45	Remove exllama import that causes problems	2023-09-17 18:00:32 -07:00
James Braza	fee38e0601	Simplified ExLlama cloning instructions and failure message (#3972)	2023-09-17 19:26:05 -03:00
Lu Guanghua	9858acee7b	Fix unexpected extensions load after gradio restart (#3965)	2023-09-17 17:35:43 -03:00
oobabooga	d9b0f2c9c3	Fix llama.cpp double decoding	2023-09-17 13:07:48 -07:00
oobabooga	d71465708c	llamacpp_HF prefix matching	2023-09-17 11:51:01 -07:00
oobabooga	37e2980e05	Recommend mul_mat_q for llama.cpp	2023-09-17 08:27:11 -07:00
oobabooga	a069f3904c	Undo part of ad8ac545a5	2023-09-17 08:12:23 -07:00
oobabooga	ad8ac545a5	Tokenization improvements	2023-09-17 07:02:00 -07:00
saltacc	cd08eb0753	token probs for non HF loaders (#3957)	2023-09-17 10:42:32 -03:00
kalomaze	7c9664ed35	Allow full model URL to be used for download (#3919) ... --------- Co-authored-by: oobabooga <112222186+oobabooga@users.noreply.github.com>	2023-09-16 10:06:13 -03:00
saltacc	ed6b6411fb	Fix exllama tokenizers (#3954) ... --------- Co-authored-by: oobabooga <112222186+oobabooga@users.noreply.github.com>	2023-09-16 09:42:38 -03:00
missionfloyd	2ad6ca8874	Add back chat buttons with --chat-buttons (#3947)	2023-09-16 00:39:37 -03:00
oobabooga	ef04138bc0	Improve the UI tokenizer	2023-09-15 19:30:44 -07:00
oobabooga	c3e4c9fdc2	Add a simple tokenizer to the UI	2023-09-15 19:09:03 -07:00
saltacc	f01b9aa71f	Add customizable ban tokens (#3899)	2023-09-15 18:27:27 -03:00
oobabooga	5b117590ad	Add some scrollbars to Parameters tab	2023-09-15 09:17:37 -07:00
Johan	fdcee0c215	Allow custom tokenizer for llamacpp_HF loader (#3941)	2023-09-15 12:38:38 -03:00
oobabooga	fd7257c7f8	Prevent code blocks from flickering while streaming	2023-09-15 07:46:26 -07:00
oobabooga	a3ecf3bb65	Add cai-chat-square chat style	2023-09-14 16:15:08 -07:00
oobabooga	3d1c0f173d	User config precedence over GGUF metadata	2023-09-14 12:15:52 -07:00
oobabooga	94dc64f870	Add a border	2023-09-14 07:20:36 -07:00
oobabooga	70aafa34dc	Fix blockquote markdown rendering	2023-09-14 05:57:04 -07:00
oobabooga	644a9b8765	Change the chat generate button	2023-09-14 05:16:44 -07:00
oobabooga	ecc90f9f62	Continue on Alt + Enter	2023-09-14 03:59:12 -07:00
oobabooga	1ce3c93600	Allow "Your name" field to be saved	2023-09-14 03:44:35 -07:00
oobabooga	27dbcc59f5	Make the chat input expand upwards (#3920)	2023-09-14 07:06:42 -03:00
oobabooga	6b6af74e14	Keyboard shortcuts without conflicts (hopefully)	2023-09-14 02:33:52 -07:00
oobabooga	fc11d1eff0	Add chat keyboard shortcuts	2023-09-13 19:22:40 -07:00
oobabooga	9f199c7a4c	Use Noto Sans font ... Copied from 6c8bd06308/public/webfonts/NotoSans	2023-09-13 13:48:05 -07:00
oobabooga	8ce94b735c	Show progress on impersonate	2023-09-13 11:22:53 -07:00
oobabooga	7cd437e05c	Properly close the hover menu on mobile	2023-09-13 11:10:46 -07:00
oobabooga	1b47b5c676	Change the Generate/Stop buttons	2023-09-13 09:25:26 -07:00
oobabooga	8ea28cbfe0	Reorder chat buttons	2023-09-13 08:49:11 -07:00
oobabooga	5e3d2f7d44	Reorganize chat buttons (#3892)	2023-09-13 02:36:12 -03:00
Panchovix	34dc7306b8	Fix NTK (alpha) and RoPE scaling for exllamav2 and exllamav2_HF (#3897)	2023-09-13 02:35:09 -03:00
oobabooga	b7adf290fc	Fix ExLlama-v2 path issue	2023-09-12 17:42:22 -07:00
oobabooga	b190676893	Merge remote-tracking branch 'refs/remotes/origin/main'	2023-09-12 15:06:33 -07:00
oobabooga	2f935547c8	Minor changes	2023-09-12 15:05:21 -07:00
oobabooga	18e6b275f3	Add alpha_value/compress_pos_emb to ExLlama-v2	2023-09-12 15:02:47 -07:00
Gennadij	460c40d8ab	Read more GGUF metadata (scale_linear and freq_base) (#3877)	2023-09-12 17:02:42 -03:00
oobabooga	16e1696071	Minor qol change	2023-09-12 10:44:26 -07:00
oobabooga	c2a309f56e	Add ExLlamaV2 and ExLlamav2_HF loaders (#3881)	2023-09-12 14:33:07 -03:00
oobabooga	df123a20fc	Prevent extra keys from being saved to settings.yaml	2023-09-11 20:13:10 -07:00
oobabooga	dae428a967	Revamp cai-chat theme, make it default	2023-09-11 19:30:40 -07:00