Commit Graph

159 Commits (049fd2fd507877d538f408139b09fd9c4cdd3bd3)
 

Author SHA1 Message Date
Andriy Mulyar 049fd2fd50
Update README.md 2 years ago
Zach Nussbaum a3485c4b32 Merge: main into gptj 2 years ago
Zach Nussbaum 8a94a8c068 fix: multi-turn data breaks 2 years ago
Zach Nussbaum 15f7c5b68f chore: peft 2 years ago
Zach Nussbaum e550e4ed34 feat: commits for eval + generation 2 years ago
Zach Nussbaum cd6a054a6c chore: remove not needed 2 years ago
Zach Nussbaum 9056a46b55 chore: submodule ff 2 years ago
Zach Nussbaum bbbf007ed9 Merge branch 'gptj' of github.com:nomic-ai/gpt4all into gptj 2 years ago
Zach Nussbaum 9dfd8e1a7c fix: num training steps for lr decay 2 years ago
Zach 311c818934 feat: evals on new gptj models 2 years ago
Zach 195f8a7d4e fix: topic model for embeddings 2 years ago
Zach Nussbaum 7807a80bbb fix: bs try one more time? 2 years ago
Zach Nussbaum 2f0eba211d fix: smaller bs for 40gb 2 years ago
Zach Nussbaum 7f95ab3a06 fix: config for lora gptj 2 years ago
Zach Nussbaum 9efdf56e38 fix: saving name 2 years ago
Zach Nussbaum 633df8edb4 Merge remote-tracking branch 'origin/mosaic' into gptj 2 years ago
Zach Nussbaum 31195270cb fix: eos/pad token + wd 2 years ago
Zach Nussbaum c82ee7d882 fix: add wd + min lr to config 2 years ago
Zach Nussbaum be3f528810 fix: tokenization error 2 years ago
Zach Nussbaum b66f127ade fix: config + ignore pkl 2 years ago
Zach Nussbaum 0606ab46b9 feat: build map script 2 years ago
Zach 1c6d2d9622 fix: embeddings instead of logits!!! 2 years ago
zanussbaum 147c2fd7eb feat: lora gptj 2 years ago
zanussbaum 2b001e8932 fix: batch size 2 years ago
zanussbaum 7cfda6a21f feat: update for mosaic 2 years ago
Zach Nussbaum 4b51e6ef37 fix: pyarrow filter 2 years ago
Zach Nussbaum 7a9f6d1cdc fix: inference save shards 2 years ago
Zach 0bd6acb4dd fix: drop uneven batch size 2 years ago
Zach 985da51fbc fix: concat 2 years ago
Zach 1b14b1f723 fix: data for inference 2 years ago
Zach fb9ff9c40d feat: inference for embedding plots 2 years ago
Zach 809680d621 fix: grad accum loss calc 2 years ago
Zach 7751f39432 fix: data processing 2 years ago
Zach 5baead45be fix: configs 2 years ago
Zach a57adb0344 fix: try except push 2 years ago
Zach Nussbaum 399a65e779 feat: multinode setup 2 years ago
Zach Nussbaum 0a3834d086 fix: gptj multinode 2 years ago
Zach Nussbaum fde7d9506f fix: ignore env 2 years ago
Zach Nussbaum 97d4499d79 fix: only on first process, not once on every node 2 years ago
Zach Nussbaum d0402288bd fix: eval func 2 years ago
Zach 65ec606f21 fix: prompt len for larger 2 years ago
Zach Nussbaum df2d5f7e46 feat: gpt-j config 2 years ago
Zach Nussbaum 3efc19ebc5 feat: adamw, fix training, log gradients 2 years ago
Zach Nussbaum 5c5f41ba36 fix: clean up data, pad at end 2 years ago
Zach Nussbaum 2e2e9f4339 fix: clean where prompt is randomly 1 char 2 years ago
Zach Nussbaum 2e3e35c7a2 chore: gitignore ckpts 2 years ago
Andriy Mulyar 252676ff05
Merge pull request #3 from nomic-ai/train
log wandb multi-epoch
2 years ago
Andriy Mulyar aa4dd0eaef
Qualified number of epochs for LoRa weights 2 years ago
Andriy Mulyar b10890f8f9
Merge pull request #42 from tiendung/main
Change git clone url to https://github.com/nomic-ai/gpt4all.git to avoid `Permission denied`
2 years ago
Andriy Mulyar c197eee060
Added Torrent Magnet Link 2 years ago