You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am issue regarding the Anthropic/hh-rlhf dataset, in reward_dataset.py:
# Anthropic/hh-rlhf# tasksource/oasst1_pairwise_rlhf_rewardifexist_and_not_none(data, "chosen") andexist_and_not_none(data, "rejected"):
prompt=data["prompt"] ifexist_and_not_none(data, "prompt") else""ifprompt.startswith("prompter:"):
prompt= (
prompt.replace("prompter:", "\nHuman:\n").replace("assistant:", "\nAssistant:\n")
+"\nAssistant:\n"
)
chosen=data["chosen"]
reject=data["rejected"]
input_template=None# do not modified with input template again
This will result in prompt="" for every sample as the Anthropic/hh-rlhf = {'chosen', rejected}`. Will this cause an issue in accounting the logits of prompts in the following code:
I am issue regarding the
Anthropic/hh-rlhf
dataset, inreward_dataset.py
:This will result in
prompt=""
for every sample as theAnthropic/hh-rlhf
= {'chosen',rejected
}`. Will this cause an issue in accounting the logits of prompts in the following code:OpenRLHF/openrlhf/trainer/dpo_trainer.py
Line 301 in fbc1eab
The text was updated successfully, but these errors were encountered: