adl-hw3-finetune-gemma-2-chinese-kyara-4

This model is a fine-tuned version of zake7749/gemma-2-2b-it-chinese-kyara-dpo on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.6104

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss
1.6765	0.3556	200	2.0620
2.0073	0.7111	400	1.9642
1.8574	1.0667	600	1.9876
1.4832	1.4222	800	2.0036
1.5055	1.7778	1000	1.9848
1.2995	2.1333	1200	2.2373
1.0272	2.4889	1400	2.2501
1.0427	2.8444	1600	2.2371
0.8523	3.2	1800	2.6061
0.6738	3.5556	2000	2.6141
0.6733	3.9111	2200	2.6104

Framework versions

• PEFT 0.13.2
• Transformers 4.45.1
• Pytorch 2.5.0+cu121
• Datasets 3.1.0
• Tokenizers 0.20.2