Inicio Explorar Axuda
Rexistro Iniciar sesión
AIForce
/
petals
réplica de https://github.com/bigscience-workshop/petals.git
5
0
Fork 0
Ficheiros Incidencias 0 Wiki
Rama: declare_adapters
Ramas Etiquetas

Commit History

Autor SHA1 Mensaxe Data
  Your Name 6db11c6483 add minimalistic peft test %!s(int64=2) %!d(string=hai) anos
  artek0chumak 03d2b05166 Remove server for tests %!s(int64=2) %!d(string=hai) anos
  artek0chumak a162e9d48e style %!s(int64=2) %!d(string=hai) anos
  artek0chumak 30e3f4a6b4 Add adapters loading %!s(int64=2) %!d(string=hai) anos
  artek0chumak 784168e8f5 fix bug: send adapter name last instead of first %!s(int64=2) %!d(string=hai) anos
  Your Name 04cf3183ee fix bug: do not raise "no adapter found" if ther is no adapter to be found %!s(int64=2) %!d(string=hai) anos
  artek0chumak 1e227240e5 select adapter by name in handler.py %!s(int64=2) %!d(string=hai) anos
  artek0chumak e452df25cc Add skeleton for peft init %!s(int64=2) %!d(string=hai) anos
  artek0chumak da204f1285 Add loading into device directly %!s(int64=2) %!d(string=hai) anos
  artek0chumak 92612aebed style %!s(int64=2) %!d(string=hai) anos
  artek0chumak ddd770000b Add layer specific loading %!s(int64=2) %!d(string=hai) anos
  artek0chumak c02ae4c9ba Fix long downloading %!s(int64=2) %!d(string=hai) anos
  artek0chumak 9ea77b0911 Safer way to download pefts %!s(int64=2) %!d(string=hai) anos
  artek0chumak 6563bf1750 Add first functional code %!s(int64=2) %!d(string=hai) anos
  artek0chumak 02e1c95415 Add possible tests %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov b28f5016ea Delete deprecated petals.cli scripts (#336) %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov fa095f6461 Use 4-bit for llama by default, use bitsandbytes 0.40.0.post3 (#340) %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov 158013a671 Implement direct server-to-server communication (#331) %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov 4d9c26fe5c Allow free_disk_space_for() remove arbitrary files from Petals cache (#339) %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov de930918a0 Support loading blocks in 4-bit (QLoRA NF4 format, disabled by default) (#333) %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov 66a47c763e Require pydantic < 2.0 (2.0 is incompatible with hivemind 1.1.8) (#337) %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov 10c72acdf4 Fix warmup steps and minor issues in benchmarks (#334) %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov d126ee3053 Add benchmark scripts (#319) %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov fecee8c4dc Show license links when loading models (#332) %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov 47a2b1ee65 Fix llama's lm_head.weight.requires_grad (#330) %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov 7a37513f77 Add AutoDistributed{Model, ModelForCausalLM, ModelForSequenceClassification} (#329) %!s(int64=2) %!d(string=hai) anos
  Alexander Borzunov cb3f018f9f Add LLaMA support (#323) %!s(int64=2) %!d(string=hai) anos
  Max Ryabinin 5c0733711a Use number of tokens for attn_cache_size (#286) %!s(int64=2) %!d(string=hai) anos
  Max Ryabinin c839173e57 Determine block dtype in a unified manner (#325) %!s(int64=2) %!d(string=hai) anos
  Max Ryabinin 3e7ae5116d Remove unused imports and attributes (#324) %!s(int64=2) %!d(string=hai) anos

Posterior Anterior