qwen2.5-7b-dpo-list_mle

This model is a fine-tuned version of Qwen/Qwen2.5-7B-Instruct-1M on the yangzhao02/ListUltraFeedback dataset. It achieves the following results on the evaluation set:

• Loss: 6.8770
• Logps: -731.7206
• Logits: -0.5560
• Rank Correct Batch: 16.5650
• Rank Pair Batch: 28.0
• Rank Accuracy Batch: 0.5916

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-07
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 16
• total_train_batch_size: 128
• total_eval_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Logps	Logits	Rank Correct Batch	Rank Pair Batch	Rank Accuracy Batch
7.253	0.2672	125	7.2435	-631.0031	-0.5630	16.0366	28.0	0.5727
7.0458	0.5344	250	6.9577	-711.2150	-0.5474	16.4350	28.0	0.5870
6.879	0.8016	375	6.8770	-731.7206	-0.5560	16.5650	28.0	0.5916

Framework versions

• Transformers 4.45.1
• Pytorch 2.6.0+cu126
• Datasets 2.19.1
• Tokenizers 0.20.3