llama-8b-instruct-simpo-full-label_smoothing-0.1

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the princeton-nlp/llama3-ultrafeedback-armorm dataset. It achieves the following results on the evaluation set:

• Loss: 1.0112
• Rewards/chosen: -5.2812
• Rewards/rejected: -6.8125
• Rewards/accuracies: 0.7764
• Rewards/margins: 1.5391
• Logps/rejected: -2.7188
• Logps/chosen: -2.1094
• Logits/rejected: -0.7656
• Logits/chosen: -0.7617

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: 1e-06
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 16
• total_train_batch_size: 256
• total_eval_batch_size: 16
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 1

Training results

Framework versions

• Transformers 4.57.1
• Pytorch 2.9.0+cu128
• Datasets 4.4.1
• Tokenizers 0.22.1