top | item 42352107 (no title) eggie5 | 1 year ago What's _online_ RL for an LLM? Saw this on the llama 3.3 reports too... discuss order hn newest whimsicalism|1 year ago Online RL for LLMs means you are sampling from the model, scoring immediately, and passing gradients back to the model.As opposed to, sampling from the model a bunch, getting scores offline, and then fine tuning the model on those offline scored generations.
whimsicalism|1 year ago Online RL for LLMs means you are sampling from the model, scoring immediately, and passing gradients back to the model.As opposed to, sampling from the model a bunch, getting scores offline, and then fine tuning the model on those offline scored generations.
whimsicalism|1 year ago
As opposed to, sampling from the model a bunch, getting scores offline, and then fine tuning the model on those offline scored generations.