support ORPO

Former-commit-id: f44a4c27e2461cdaa1b16865f597a31033c0e6d9

examples/lora_single_gpu/README.md

@@ -1,8 +1,9 @@
 Usage:
 
 - `pretrain.sh`: do pre-train (optional)
-- `sft.sh`: do supervised fine-tune
+- `sft.sh`: do supervised fine-tuning
 - `reward.sh`: do reward modeling (must after sft.sh)
 - `ppo.sh`: do PPO training (must after sft.sh and reward.sh)
 - `dpo.sh`: do DPO training (must after sft.sh)
+- `orpo.sh`: do ORPO training
 - `predict.sh`: do predict (must after sft.sh and dpo.sh)

examples/lora_single_gpu/orpo.sh

@@ -0,0 +1,32 @@
+#!/bin/bash
+
+CUDA_VISIBLE_DEVICES=0 python ../../src/train_bash.py \
+    --stage orpo \
+    --do_train \
+    --model_name_or_path meta-llama/Llama-2-7b-hf \
+    --dataset comparison_gpt4_en \
+    --dataset_dir ../../data \
+    --template default \
+    --finetuning_type lora \
+    --lora_target q_proj,v_proj \
+    --output_dir ../../saves/LLaMA2-7B/lora/orpo \
+    --overwrite_cache \
+    --overwrite_output_dir \
+    --cutoff_len 1024 \
+    --preprocessing_num_workers 16 \
+    --per_device_train_batch_size 1 \
+    --per_device_eval_batch_size 1 \
+    --gradient_accumulation_steps 8 \
+    --lr_scheduler_type cosine \
+    --logging_steps 10 \
+    --warmup_steps 20 \
+    --save_steps 100 \
+    --eval_steps 100 \
+    --evaluation_strategy steps \
+    --load_best_model_at_end \
+    --learning_rate 1e-5 \
+    --num_train_epochs 1.0 \
+    --max_samples 1000 \
+    --val_size 0.1 \
+    --plot_loss \
+    --fp16