lucky9-cyou/qwen-32b-finetune-stage-1

train_2025-05-25-12-02-28

This model is a fine-tuned version of /workspace/qwen3-32b on the roleplay_multi_turn dataset. It achieves the following results on the evaluation set:

• Loss: 1.5254
• Num Input Tokens Seen: 111133312

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-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• total_eval_batch_size: 64
• optimizer: Use OptimizerNames.PAGED_LION_8BIT and the args are: No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 2
• num_epochs: 1.0

Training results

Framework versions

• Transformers 4.52.1
• Pytorch 2.6.0+cu124
• Datasets 3.3.2
• Tokenizers 0.21.1