2023-05-19 10:42:41 -04:00
|
|
|
.*(llama|alpac|vicuna|guanaco|koala|llava|wizardlm|metharme|pygmalion-7b|wizard-mega|openbuddy|vigogne|h2ogpt-research|manticore):
|
2023-04-14 10:07:28 -04:00
|
|
|
model_type: 'llama'
|
2023-05-10 01:07:22 -04:00
|
|
|
.*(opt-|opt_|opt1|opt3|optfor|galactica|galpaca|pygmalion-350m):
|
|
|
|
model_type: 'opt'
|
2023-05-19 10:42:41 -04:00
|
|
|
.*(gpt-j|gptj|gpt4all-j|malion-6b|pygway|pygmalion-6b|dolly-v1):
|
2023-05-10 01:07:22 -04:00
|
|
|
model_type: 'gptj'
|
2023-05-19 10:42:41 -04:00
|
|
|
.*(gpt-neox|koalpaca-polyglot|polyglot.*koalpaca|polyglot-ko|polyglot_ko|pythia|stablelm|incite|dolly-v2|polycoder|h2ogpt-oig|h2ogpt-oasst1|h2ogpt-gm):
|
|
|
|
model_type: 'gpt_neox'
|
|
|
|
.*llama:
|
|
|
|
model_type: 'llama'
|
|
|
|
.*bloom:
|
|
|
|
model_type: 'bloom'
|
|
|
|
llama-65b-gptq-3bit:
|
|
|
|
groupsize: 'None'
|
2023-05-09 19:37:31 -04:00
|
|
|
.*(4bit|int4):
|
2023-04-14 10:07:28 -04:00
|
|
|
wbits: 4
|
2023-05-09 19:37:31 -04:00
|
|
|
.*(3bit|int3):
|
2023-04-14 10:07:28 -04:00
|
|
|
wbits: 3
|
2023-05-09 19:37:31 -04:00
|
|
|
.*(-2bit|_2bit|int2-):
|
|
|
|
wbits: 2
|
|
|
|
.*(-1bit|_1bit|int1-):
|
|
|
|
wbits: 1
|
|
|
|
.*(8bit|int8):
|
|
|
|
wbits: 8
|
|
|
|
.*(-7bit|_7bit|int7-):
|
|
|
|
wbits: 7
|
|
|
|
.*(-6bit|_6bit|int6-):
|
|
|
|
wbits: 6
|
|
|
|
.*(-5bit|_5bit|int5-):
|
|
|
|
wbits: 5
|
2023-05-12 05:09:45 -04:00
|
|
|
.*(-gr32-|-32g-|groupsize32|-32g$):
|
2023-05-09 19:37:31 -04:00
|
|
|
groupsize: 32
|
2023-05-12 05:11:58 -04:00
|
|
|
.*(-gr64-|-64g-|groupsize64|-64g$):
|
2023-05-09 19:37:31 -04:00
|
|
|
groupsize: 64
|
|
|
|
.*(gr128|128g|groupsize128):
|
2023-04-14 10:07:28 -04:00
|
|
|
groupsize: 128
|
2023-05-09 19:37:31 -04:00
|
|
|
.*(gr1024|1024g|groupsize1024):
|
|
|
|
groupsize: 1024
|
2023-05-19 10:42:41 -04:00
|
|
|
.*(oasst|openassistant-|stablelm-7b-sft-v7-epoch-3):
|
2023-04-14 10:07:28 -04:00
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Open Assistant'
|
2023-05-09 19:37:31 -04:00
|
|
|
skip_special_tokens: false
|
2023-05-19 10:42:41 -04:00
|
|
|
(?!.*galactica)(?!.*reward).*openassistant:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Open Assistant'
|
|
|
|
skip_special_tokens: false
|
|
|
|
(?!.*v0)(?!.*1.1)(?!.*1_1)(?!.*stable)(?!.*chinese).*vicuna:
|
2023-04-14 10:07:28 -04:00
|
|
|
mode: 'instruct'
|
2023-04-26 15:20:27 -04:00
|
|
|
instruction_template: 'Vicuna-v0'
|
2023-05-09 19:37:31 -04:00
|
|
|
.*vicuna.*v0:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Vicuna-v0'
|
|
|
|
.*vicuna.*(1.1|1_1):
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Vicuna-v1.1'
|
2023-05-11 10:06:39 -04:00
|
|
|
.*wizard.*vicuna:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Vicuna-v1.1'
|
2023-05-09 19:37:31 -04:00
|
|
|
.*stable.*vicuna:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'StableVicuna'
|
2023-05-19 10:42:41 -04:00
|
|
|
(?!.*chat).*chinese-vicuna:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Alpaca'
|
|
|
|
.*chinese-vicuna.*chat:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Chinese-Vicuna-Chat'
|
2023-04-14 10:07:28 -04:00
|
|
|
.*alpaca:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Alpaca'
|
2023-04-14 10:15:59 -04:00
|
|
|
.*alpaca-native-4bit:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Alpaca'
|
|
|
|
wbits: 4
|
|
|
|
groupsize: 128
|
2023-05-09 19:37:31 -04:00
|
|
|
.*galactica:
|
2023-04-16 13:24:49 -04:00
|
|
|
skip_special_tokens: false
|
|
|
|
.*dolly-v[0-9]-[0-9]*b:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Alpaca'
|
|
|
|
skip_special_tokens: false
|
|
|
|
custom_stopping_strings: '"### End"'
|
2023-04-16 13:40:45 -04:00
|
|
|
.*koala:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Koala'
|
2023-04-16 18:15:03 -04:00
|
|
|
.*chatglm:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'ChatGLM'
|
2023-05-09 21:29:22 -04:00
|
|
|
.*metharme:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Metharme'
|
2023-04-23 19:32:22 -04:00
|
|
|
.*llava:
|
|
|
|
mode: 'instruct'
|
|
|
|
model_type: 'llama'
|
|
|
|
instruction_template: 'LLaVA'
|
2023-04-26 02:47:34 -04:00
|
|
|
custom_stopping_strings: '"\n###"'
|
2023-04-26 02:21:53 -04:00
|
|
|
.*raven:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'RWKV-Raven'
|
2023-05-04 22:19:23 -04:00
|
|
|
.*moss-moon.*sft:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'MOSS'
|
2023-05-09 19:37:31 -04:00
|
|
|
.*stablelm-tuned:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'StableLM'
|
|
|
|
truncation_length: 4096
|
|
|
|
chat_prompt_size: 4096
|
|
|
|
chat_prompt_size_max: 4096
|
|
|
|
.*stablelm-base:
|
|
|
|
truncation_length: 4096
|
|
|
|
chat_prompt_size: 4096
|
|
|
|
chat_prompt_size_max: 4096
|
|
|
|
.*wizardlm:
|
|
|
|
mode: 'instruct'
|
|
|
|
model_type: 'llama'
|
|
|
|
instruction_template: 'WizardLM'
|
|
|
|
.*galactica.*finetuned:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Galactica Finetuned'
|
|
|
|
.*galactica.*-v2:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Galactica v2'
|
|
|
|
(?!.*finetuned)(?!.*-v2).*galactica:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Galactica'
|
|
|
|
.*guanaco:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Guanaco non-chat'
|
|
|
|
.*baize:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Baize'
|
|
|
|
.*mpt-.*instruct:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Alpaca'
|
|
|
|
.*mpt-.*chat:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'MPT-Chat'
|
|
|
|
(?!.*-flan-)(?!.*-t5-).*lamini-:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Alpaca'
|
|
|
|
.*incite.*chat:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'INCITE-Chat'
|
|
|
|
.*incite.*instruct:
|
|
|
|
mode: 'instruct'
|
2023-05-11 10:06:39 -04:00
|
|
|
instruction_template: 'INCITE-Instruct'
|
2023-05-19 10:42:41 -04:00
|
|
|
.*wizard.*mega:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Wizard-Mega'
|
|
|
|
.*ziya-:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Ziya'
|
|
|
|
.*koalpaca:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'KoAlpaca'
|
|
|
|
.*openbuddy:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'OpenBuddy'
|
|
|
|
(?!.*chat).*vigogne:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Vigogne-Instruct'
|
|
|
|
.*vigogne.*chat:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Vigogne-Chat'
|
|
|
|
.*(llama-deus|supercot|llama-natural-instructions|open-llama-0.3t-7b-instruct-dolly-hhrlhf|open-llama-0.3t-7b-open-instruct):
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Alpaca'
|
|
|
|
.*bactrian:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Bactrian'
|
|
|
|
.*(h2ogpt-oig-|h2ogpt-oasst1-|h2ogpt-research-oasst1-):
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'H2O-human_bot'
|
|
|
|
.*h2ogpt-gm-:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'H2O-prompt_answer'
|
|
|
|
.*manticore:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Manticore Chat'
|
2023-06-01 13:42:12 -04:00
|
|
|
.*bluemoonrp-(30|13)b:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Bluemoon'
|
2023-06-05 11:07:52 -04:00
|
|
|
truncation_length: 4096
|
2023-06-05 10:52:13 -04:00
|
|
|
.*Nous-Hermes-13b:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Alpaca'
|
|
|
|
.*airoboros-13b-gpt4:
|
|
|
|
mode: 'instruct'
|
|
|
|
instruction_template: 'Vicuna-v1.1'
|
2023-06-05 11:07:52 -04:00
|
|
|
truncation_length: 4096
|