Qwen2.5-7B-spin-2k-hhrlhf

This model is a fine-tuned version of Qwen/Qwen2.5-7B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1118
• Rewards/real: 7.4786
• Rewards/generated: -3.0273
• Rewards/accuracies: 1.0
• Rewards/margins: 10.5060
• Logps/generated: -309.5072
• Logps/real: -106.0837
• Logits/generated: -0.7103
• Logits/real: -1.2792

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: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• total_eval_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/real	Rewards/generated	Rewards/accuracies	Rewards/margins	Logps/generated	Logps/real	Logits/generated	Logits/real
0.1077	0.8	50	0.1118	7.4786	-3.0273	1.0	10.5060	-309.5072	-106.0837	-0.7103	-1.2792

Framework versions

• Transformers 4.43.3
• Pytorch 2.2.2+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1