ROE_Patent_Breeze7B_V1

This model is a fine-tuned version of MediaTek-Research/Breeze-7B-Instruct-v1_0 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7530

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: 1
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 11
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.9896	0.9697	12	0.9971
0.9038	1.9394	24	0.9124
0.812	2.9899	37	0.8596
0.8009	3.9596	49	0.8248
0.7401	4.9293	61	0.8017
0.6997	5.9798	74	0.7835
0.6831	6.9495	86	0.7715
0.656	8.0	99	0.7635
0.6389	8.9697	111	0.7558
0.6094	9.9394	123	0.7550
0.5967	10.6667	132	0.7530

Framework versions

• PEFT 0.13.3.dev0
• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.1