We are excited to introduce the new v0.9.3 release. Many new exciting features and algorithms. The highlights are as follows:
RLOO Trainer: RLOO (Reinforce Leave-one-out) is a new online RL algorithm for RLHF, proposed by Ahmadian et al from Cohere. Check out our docs here to get started
PPOv2 Trainer: We are introducing a new experimental PPOv2 trainer which is more aligned with OpenAI's PPO implementation based on https://arxiv.org/abs/2403.17031. Check out our docs here to get started
Reward model visualization: the reward model training now includes visualization on the eval dataset, as shown below.
New losses in the DPO Trainer: DPOTrainer now includes losses / support for Self-play Preference Optimization, Robust DPO, TR-DPO, Iterative Reasoning Preference Optimization, and Pairwise Noise Contrastive Alignment
New losses in the KTO Trainer: KTOTrainer now includes the loss for Binary Classifier Optimization (BCO)
Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.
Dependabot commands and options
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
- `@dependabot show ignore conditions` will show all of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)
Bumps trl from 0.8.6 to 0.9.4.
Release notes
Sourced from trl's releases.
... (truncated)
Commits
974b0d3
0.9.4 release (#1708)39a7d1c
SFTTrainer: Fix backward Compatibility issue withTrainingArguments
(#1707)0bdc638
Fixed doc string and docs for the SFTConfig update (#1706)275d33b
0.9.3 release (#1699)c0819ee
Update sft_trainer.py (#1698)a03e7cc
Release 0.9.2 (#1697)a13cb89
Quick fix on GPT4-eval (#1696)84156f1
Fix typo in DPOTrainer's warnings (#1688)4eb0b90
Skip packing validation (#1673)6c203f9
Fix overriding optimize_device_cache with optimize_cuda_cache in PPOConfig (#...Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting
@dependabot rebase
.Dependabot commands and options
You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot merge` will merge this PR after your CI passes on it - `@dependabot squash and merge` will squash and merge this PR after your CI passes on it - `@dependabot cancel merge` will cancel a previously requested merge and block automerging - `@dependabot reopen` will reopen this PR if it is closed - `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually - `@dependabot show