Commit Graph

  • a2918176ea Update LLaMA-v2-model.md (thanks Panchovix) oobabooga 2023-07-18 13:21:18 -0700
  • e0631e309f
    Create instruction template for Llama-v2 (#3194) oobabooga 2023-07-18 17:19:18 -0300
  • d52ab78626 Apply 4096 context length automatically oobabooga 2023-07-18 13:18:35 -0700
  • 8ef6eb86cb Don't use the Meta system message oobabooga 2023-07-18 13:16:20 -0700
  • 0757c16b3c Use the default system message provided by Meta oobabooga 2023-07-18 13:11:12 -0700
  • 9ad44ba38d Fix the template, update models/config.yaml oobabooga 2023-07-18 13:04:32 -0700
  • f258817dee Typo oobabooga 2023-07-18 12:58:55 -0700
  • 72481e373b Create Llama-v2 instruction template oobabooga 2023-07-18 12:55:21 -0700
  • a714f68b9f
    Proposal, augment context FartyPants 2023-07-18 13:54:50 -0400
  • c8306d8f37 Multi-LoRA loading: Handle edge cases Gábor Stefanik 2023-07-18 19:54:34 +0200
  • 603c596616 Add LLaMA-v2 conversion instructions oobabooga 2023-07-18 10:29:56 -0700
  • c535f14e5f
    Bump bitsandbytes Windows wheel to 0.40.2 (#3186) jllllll 2023-07-18 09:39:43 -0500
  • d7a14174a2
    Remove auto-loading when only one model is available (#3187) jllllll 2023-07-18 09:39:08 -0500
  • 66c8556c98
    Update script.py sammyf 2023-07-18 12:36:06 +0200
  • 9845c7f94d
    Remove auto-loading when only one model is available jllllll 2023-07-18 03:47:37 -0500
  • bad922c897
    Bump bitsandbytes Windows wheel to 0.40.2 jllllll 2023-07-18 03:44:58 -0500
  • 96e6e809a1 Added the logic for starchat model series GiganticPrime 2023-07-18 08:04:43 +0000
  • 2a16c3bf4a Add support for llamacpp_HF Shouyi Wang 2023-07-18 14:46:17 +1000
  • a69955377a
    [GGML] Support for customizable RoPE (#3083) randoentity 2023-07-18 03:32:37 +0200
  • 0ae2990e3b Add comma oobabooga 2023-07-17 17:49:43 -0700
  • 3f264ff038 Remove from server.py oobabooga 2023-07-17 17:48:46 -0700
  • 04b3102232 Merge branch 'dev' into randoentity-custom_rope oobabooga 2023-07-17 17:47:18 -0700
  • 356dceb4cc Reuse the exllama parameters oobabooga 2023-07-17 17:47:08 -0700
  • 89e0d15cf5
    Use 'torch.backends.mps.is_available' to check if mps is supported (#3164) appe233 2023-07-18 08:27:18 +0800
  • 234c58ccd1
    Bump bitsandbytes from 0.40.1.post1 to 0.40.2 (#3178) dependabot[bot] 2023-07-17 21:24:51 -0300
  • 49a5389bd3
    Bump accelerate from 0.20.3 to 0.21.0 oobabooga 2023-07-17 21:23:59 -0300
  • 2d26a07aaf Change [0.0] to None for default tensor_split Shouyi Wang 2023-07-18 10:08:56 +1000
  • 8c1c2e0fae Increase max_new_tokens upper limit oobabooga 2023-07-17 17:08:22 -0700
  • 5e5d926d2b Prevent lists from flickering in chat mode while streaming oobabooga 2023-07-17 17:00:49 -0700
  • 23b5e0a2e1 improve default chat instruction template Matthew Ashton 2023-07-17 18:35:54 -0400
  • e03b41aea1
    Bump gradio from 3.33.1 to 3.37.0 dependabot[bot] 2023-07-17 20:18:40 +0000
  • 02a5fe6aa2
    Bump accelerate from 0.20.3 to 0.21.0 dependabot[bot] 2023-07-17 20:18:31 +0000
  • 34d3e126cf
    Bump gradio-client from 0.2.5 to 0.2.10 dependabot[bot] 2023-07-17 20:18:18 +0000
  • 7e7c54df2f
    Bump bitsandbytes from 0.40.1.post1 to 0.40.2 dependabot[bot] 2023-07-17 20:18:09 +0000
  • 61504271a5 new min temp: 0.02, whitespace Matthew Ashton 2023-07-17 16:14:13 -0400
  • 60a3e70242 Update LLaMA links and info oobabooga 2023-07-17 12:51:01 -0700
  • f83fdb9270 Don't reset LoRA menu when loading a model oobabooga 2023-07-17 12:50:25 -0700
  • 7a9df40fcd fixup rwkv with temp < 0.01, missing context ok Matthew Ashton 2023-07-17 15:08:58 -0400
  • 4ce766414b Bump AutoGPTQ version oobabooga 2023-07-17 10:02:12 -0700
  • b1a6ea68dd Disable "autoload the model" by default oobabooga 2023-07-17 07:40:56 -0700
  • 656b457795 Add Airoboros-v1.2 template oobabooga 2023-07-17 07:27:42 -0700
  • 27a85e72b6 fixup! [GGML] Support for customizable RoPE randoentity 2023-07-17 13:03:45 +0200
  • 8b72e7e6d3 [GGML] Support for customizable RoPE randoentity 2023-07-09 20:11:25 +0200
  • 1a8752ff5a Add tensor_split parameter support Shouyi Wang 2023-07-07 22:06:00 +1000
  • d5a14dd315 readme netrunnereve 2023-07-17 00:36:01 -0400
  • ed5d582228 llama.cpp rope netrunnereve 2023-07-17 00:25:29 -0400
  • 1f66d56206 update name tc 2023-07-16 21:10:07 -0700
  • a199f21799 Optimize llamacpp_hf a bit oobabooga 2023-07-16 20:49:48 -0700
  • 5c9d2dae39
    Merge branch 'main' into xinference Bojun-Feng 2023-07-17 10:29:17 +0800
  • 01ff95690a Lot of enhancements in training - new scheduler - checks for gradient accumulation - creating graphs - new raw text parser FPHam 2023-07-16 15:41:51 -0400
  • 64a6091aa2 Resolved merge conflict. FPHam 2023-07-16 15:39:34 -0400
  • b30fbc5b76 Merge branch 'openai_update' of github.com:matatonic/text-generation-webui into openai_update Matthew Ashton 2023-07-16 14:54:43 -0400
  • 52b390413b logprob fix for streaming completions (unofficial) Matthew Ashton 2023-07-16 14:54:40 -0400
  • 846978011e
    Merge branch 'oobabooga:main' into openai_update matatonic 2023-07-16 11:59:32 -0400
  • fdc7483e30 Added OPENAI_PORT in Dockerfile and .env.example petcumihai2002@yahoo.com 2023-07-16 15:11:33 +0300
  • 8ecf72008e Dockerfile <- added openai requirements; docker-compose.yml <- replaced device_ids: [ '0' ] with count: all (to let all available GPUs to be used) petcumihai2002@yahoo.com 2023-07-16 14:44:59 +0300
  • dc28bf3ada Use 'torch.backends.mps.is_available' to check if mps is supported appe233 2023-07-16 14:28:44 +0800
  • 9f08038864
    Merge pull request #3163 from oobabooga/dev v1.2 oobabooga 2023-07-16 02:43:18 -0300
  • 6a3edb0542 Clean up llamacpp_hf.py oobabooga 2023-07-15 22:40:55 -0700
  • 2de0cedce3 Fix reload screen color oobabooga 2023-07-15 22:39:39 -0700
  • 13449aa44d Decrease download timeout oobabooga 2023-07-15 22:30:08 -0700
  • 27a84b4e04 Make AutoGPTQ the default again oobabooga 2023-07-15 22:29:23 -0700
  • 5e3f7e00a9
    Create llamacpp_HF loader (#3062) oobabooga 2023-07-16 02:21:13 -0300
  • 66ba79ed05 Unwanted changes oobabooga 2023-07-15 22:18:54 -0700
  • 17ddbf4dee Merge branch 'dev' into llamacpp_hf oobabooga 2023-07-15 22:14:52 -0700
  • 05d6afba68 Clean up oobabooga 2023-07-15 22:12:10 -0700
  • 7c4d4fc7d3
    Increase alpha value limit for NTK RoPE scaling for exllama/exllama_HF (#3149) Panchovix 2023-07-16 00:56:04 -0400
  • 780a2f2e16
    Bump llama cpp version (#3160) ofirkris 2023-07-16 07:54:56 +0300
  • 522809e14b
    Merge branch 'dev' into patch-5 oobabooga 2023-07-16 01:54:36 -0300
  • ed3ffd212d
    Bump bitsandbytes to 0.40.1.post1 (#3156) jllllll 2023-07-15 23:53:32 -0500
  • 94dfcec237
    Make it possible to evaluate exllama perplexity (#3138) oobabooga 2023-07-16 01:52:55 -0300
  • 5031c55a10 Minor change oobabooga 2023-07-15 21:46:42 -0700
  • 11a8fd1eb9
    Add cuBLAS llama-cpp-python wheel installation (#102) jllllll 2023-07-15 23:31:33 -0500
  • 0adb62dbfe update whisper tc 2023-07-15 18:33:55 -0700
  • c85b26dd1e update plugin to chinese tc 2023-07-15 18:32:20 -0700
  • 83739a04c3 rename tc 2023-07-15 17:42:42 -0700
  • 5ec2f0d1d7 test extention rename tc 2023-07-15 17:42:40 -0700
  • 86e11f65c5
    Bump llama cpp version ofirkris 2023-07-16 01:54:46 +0300
  • 93dcf9f835 experimental: logprobs & logit_bias Matthew Ashton 2023-07-15 18:16:16 -0400
  • 4300ae8316 fix typo and add truncation_length support Foxtr0t1337 2023-07-16 05:47:24 +0800
  • 1cd95a7512 Ensure markdown conversion still runs for messages without probability dropdowns SeanScripts 2023-07-15 16:17:40 -0500
  • f5c35feb3d Add temporary fix for generation speed issue SeanScripts 2023-07-15 16:07:42 -0500
  • 847707562d
    Lowering gradient accumulation to go around Transformers bug FartyPants 2023-07-15 15:33:42 -0400
  • ce4e811c31
    Bump bitsandbytes to 0.40.1.post1 jllllll 2023-07-15 10:04:06 -0500
  • f17db8f5c7
    Update script.py SodaPrettyCold 2023-07-15 13:28:53 +0800
  • 0fff5ec1a5 Decrease download timeout oobabooga 2023-07-14 19:09:48 -0700
  • cf1cdfcee4
    typo Panchovix 2023-07-14 21:59:47 -0400
  • 8a180c2ffe
    Increase alpha value limit Panchovix 2023-07-14 21:51:17 -0400
  • 797e97ee08 Fix most of the CSS issues and clean up code SeanScripts 2023-07-14 18:26:37 -0500
  • fa3f6a5eb5 some logit_bias/logprobs improvements. Matthew Ashton 2023-07-14 19:18:23 -0400
  • b284f2407d Make ExLlama_HF the new default for GPTQ oobabooga 2023-07-14 14:03:56 -0700
  • c9aba55fa1 Make evaluation functional oobabooga 2023-07-14 10:18:36 -0700
  • cb99f202ce
    Merge branch 'oobabooga:main' into openai_update matatonic 2023-07-14 12:15:38 -0400
  • d21c1e34f9 fix token count, better errors, fix tiktoken lists Matthew Ashton 2023-07-14 11:50:58 -0400
  • 6ccc066cd9
    Attempt to using latest versions of requirements ofirkris 2023-07-14 10:18:29 +0300
  • 16bd2eaf13 Add hover probability dropdown SeanScripts 2023-07-14 00:14:25 -0500
  • 7d4b743782 Merge branch 'dev' into llamacpp_hf oobabooga 2023-07-13 16:18:09 -0700
  • de328ee19e Make it possible to evaluate exllama oobabooga 2023-07-13 16:00:35 -0700
  • 5bfe55fc76 Merge branch 'dev' into IllogicalDesigns-main oobabooga 2023-07-13 14:15:57 -0700
  • 6e648ca494 Minor changes oobabooga 2023-07-13 14:15:17 -0700