rishavranaut/Qwen2_Final_MT

llama2-7B_final_MT

This model is a fine-tuned version of Qwen/Qwen2-7B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5523
• Accuracy: 0.8117
• Precision: 0.7913
• Recall: 0.8467
• F1 score: 0.8180

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1 score
0.8396	0.5	200	0.8733	0.645	0.6064	0.8267	0.6996
0.6541	1.0	400	0.6882	0.695	0.7127	0.6533	0.6817
0.4601	1.5	600	0.6691	0.7067	0.6505	0.8933	0.7528
0.4437	2.0	800	0.5010	0.7833	0.7690	0.81	0.7890
0.3406	2.5	1000	0.5010	0.7767	0.7823	0.7667	0.7744
0.2919	3.0	1200	0.4927	0.8117	0.8127	0.81	0.8114
0.2219	3.5	1400	0.4971	0.8217	0.8044	0.85	0.8266
0.2154	4.0	1600	0.6404	0.7633	0.7147	0.8767	0.7874
0.1381	4.5	1800	0.5391	0.815	0.8	0.84	0.8195
0.1531	5.0	2000	0.5523	0.8117	0.7913	0.8467	0.8180

Framework versions

• PEFT 0.11.1
• Transformers 4.44.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1