widthIrplus_msg_qm9s_mb_msd

This model is a fine-tuned version of /data/group/project1/models/Qwen3-32B on the QM9S_ir_train, the QM9S_uv_train, the QM9S_raman_train, the QM9S_all_train, the MB_train, the MSG_train, the MSD_ms_train, the MSD_nmr_train, the MSD_ir_train and the MSD_all_train datasets.

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-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• total_eval_batch_size: 64
• 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
• num_epochs: 1.0

Training results

Framework versions

• PEFT 0.15.1
• Transformers 4.51.3
• Pytorch 2.7.0+cu126
• Datasets 3.5.0
• Tokenizers 0.21.1