Implement the DPO loss given policy/reference logprobs for chosen/rejected $\beta$

Easy

Direct Alignment: DPO, Rejection Sampling & Distillation1 / 5

Chapter 131 / 5

Direct Alignment: DPO, Rejection Sampling & Distillation

Implement the DPO loss given policy/reference logprobs for chosen/rejected $\beta$

Easy

Target interface

dpo_loss(pi_logps_w, pi_logps_l, ref_logps_w, ref_logps_l, beta=0.1)

Implement the function/class skeleton in the editor. Any correct approach is accepted.

Hints0 / 2

Reference solution

Your own code stays in the editor.

solution.pyPython

Local draft

import torch
import torch.nn.functional as F

def dpo_loss(pi_logps_w, pi_logps_l, ref_logps_w, ref_logps_l, beta=0.1):
    raise NotImplementedError

⌘/Ctrl↵to submit

AI review

No review yet

Submit your solution for a verdict, a score out of 100 and concrete feedback. Any correct approach passes.

Chapter 131 / 5

Direct Alignment: DPO, Rejection Sampling & Distillation

Implement the DPO loss given policy/reference logprobs for chosen/rejected $\beta$

Easy

Target interface

dpo_loss(pi_logps_w, pi_logps_l, ref_logps_w, ref_logps_l, beta=0.1)

Implement the function/class skeleton in the editor. Any correct approach is accepted.

Hints0 / 2

Reference solution

Your own code stays in the editor.