modicalSFT_ckpt15800

---

library_name: peft

license: other

base_model: /home/ma-user/modelarts/lss/models/qwen2.5_7b

tags:

• base_model:adapter:/home/ma-user/modelarts/lss/models/qwen2.5_7b

• llama-factory

• lora

• transformers

pipeline_tag: text-generation

model-index:

• name: output_med_7b_8x

  results: []

---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

output_med_7b_8x

This model is a fine-tuned version of /home/ma-user/modelarts/lss/models/qwen2.5_7b on the med_train_zh and the med_train_en datasets. It achieves the following results on the evaluation set:

• Loss: 1.8047

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: 2
• total_train_batch_size: 128
• total_eval_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 0.05
• num_epochs: 1.0

Training results

Framework versions

• PEFT 0.17.1
• Transformers 4.57.1
• Pytorch 2.9.1+cu128
• Datasets 4.0.0
• Tokenizers 0.22.1