qwen_lora

This model is a fine-tuned version of Qwen/Qwen2.5-1.5B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0622
• Mse: 0.0622
• Mae: 0.1968
• R Squared: 0.3060

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: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Mse	Mae	R Squared
0.0875	0.3115	100	0.0854	0.0854	0.2351	0.0471
0.0786	0.6231	200	0.0741	0.0741	0.2186	0.1735
0.0709	0.9346	300	0.0716	0.0716	0.2193	0.2018
0.0675	1.2461	400	0.0735	0.0735	0.2106	0.1803
0.0681	1.5576	500	0.0710	0.0710	0.2076	0.2081
0.0627	1.8692	600	0.0675	0.0675	0.2059	0.2468
0.0628	2.1807	700	0.0657	0.0657	0.2031	0.2677
0.0591	2.4922	800	0.0646	0.0646	0.2033	0.2799
0.06	2.8037	900	0.0660	0.0660	0.2007	0.2638
0.0553	3.1153	1000	0.0633	0.0633	0.2012	0.2944
0.0612	3.4268	1100	0.0654	0.0654	0.2078	0.2711
0.0542	3.7383	1200	0.0627	0.0627	0.1987	0.3009
0.0529	4.0498	1300	0.0623	0.0623	0.1970	0.3049
0.0546	4.3614	1400	0.0624	0.0624	0.1962	0.3044
0.0535	4.6729	1500	0.0623	0.0623	0.1972	0.3055
0.0536	4.9844	1600	0.0622	0.0622	0.1968	0.3060

Framework versions

• PEFT 0.13.2
• Transformers 4.45.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3