finetuned_models

This model is a fine-tuned version of Qwen/Qwen2.5-7B-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4359
• Accuracy: 0.9399
• F1: 0.9399
• Precision: 0.9400
• Recall: 0.9399

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
1.0992	0.0865	50	0.8783	0.6714	0.6681	0.6770	0.6714
0.5328	0.1730	100	0.6396	0.7875	0.7841	0.8010	0.7875
0.3164	0.2595	150	0.5246	0.8255	0.8237	0.8318	0.8255
0.6413	0.3460	200	0.3856	0.8792	0.8790	0.8813	0.8792
0.5142	0.4325	250	0.3306	0.8852	0.8856	0.8880	0.8852
0.4424	0.5190	300	0.3374	0.8848	0.8854	0.8881	0.8848
0.3363	0.6055	350	0.2875	0.9089	0.9088	0.9087	0.9089
0.0819	0.6920	400	0.3526	0.8999	0.8991	0.9027	0.8999
0.2416	0.7785	450	0.2647	0.9132	0.9135	0.9152	0.9132
0.5143	0.8651	500	0.3480	0.9042	0.9036	0.9071	0.9042
0.1396	0.9516	550	0.3365	0.8973	0.8972	0.9032	0.8973
0.101	1.0381	600	0.3081	0.9190	0.9189	0.9216	0.9190
0.1129	1.1246	650	0.3920	0.9147	0.9154	0.9206	0.9147
0.0622	1.2111	700	0.3411	0.9275	0.9276	0.9287	0.9275
0.0808	1.2976	750	0.3357	0.9346	0.9347	0.9348	0.9346
0.0043	1.3841	800	0.3568	0.9290	0.9292	0.9299	0.9290
0.1549	1.4706	850	0.2835	0.9328	0.9328	0.9329	0.9328
0.1464	1.5571	900	0.2998	0.9348	0.9348	0.9351	0.9348
0.193	1.6436	950	0.3660	0.9253	0.9258	0.9284	0.9253
0.2246	1.7301	1000	0.4104	0.9281	0.9281	0.9290	0.9281
0.2503	1.8166	1050	0.3155	0.9347	0.9348	0.9353	0.9347
0.0571	1.9031	1100	0.3476	0.9320	0.9318	0.9322	0.9320
0.034	1.9896	1150	0.3135	0.9390	0.9389	0.9389	0.9390
0.0002	2.0761	1200	0.3508	0.9381	0.9381	0.9381	0.9381
0.0004	2.1626	1250	0.3696	0.9381	0.9382	0.9384	0.9381
0.0022	2.2491	1300	0.3761	0.9392	0.9392	0.9394	0.9392
0.0023	2.3356	1350	0.4013	0.9378	0.9379	0.9385	0.9378
0.0602	2.4221	1400	0.4008	0.9384	0.9385	0.9389	0.9384
0.0095	2.5087	1450	0.4055	0.9387	0.9388	0.9390	0.9387
0.0	2.5952	1500	0.4149	0.9390	0.9390	0.9390	0.9390
0.0	2.6817	1550	0.4279	0.9388	0.9388	0.9388	0.9388
0.0014	2.7682	1600	0.4286	0.9397	0.9397	0.9397	0.9397
0.0002	2.8547	1650	0.4330	0.9407	0.9407	0.9408	0.9407
0.0046	2.9412	1700	0.4357	0.9391	0.9390	0.9391	0.9391
0.0	3.0277	1750	0.4364	0.9395	0.9395	0.9395	0.9395
0.0	3.1142	1800	0.4356	0.9398	0.9398	0.9398	0.9398
0.0	3.2007	1850	0.4367	0.9403	0.9402	0.9403	0.9403
0.0	3.2872	1900	0.4349	0.9400	0.9400	0.9400	0.9400
0.0	3.3737	1950	0.4353	0.9399	0.9399	0.9400	0.9399
0.0	3.4602	2000	0.4349	0.9403	0.9403	0.9403	0.9403
0.0	3.5467	2050	0.4350	0.9400	0.9400	0.9400	0.9400
0.0	3.6332	2100	0.4353	0.9397	0.9397	0.9397	0.9397
0.0	3.7197	2150	0.4358	0.9403	0.9402	0.9402	0.9403
0.0	3.8062	2200	0.4350	0.9400	0.9400	0.9401	0.9400
0.0	3.8927	2250	0.4343	0.9397	0.9397	0.9398	0.9397
0.0	3.9792	2300	0.4333	0.9399	0.9399	0.9400	0.9399
0.0	4.0657	2350	0.4336	0.9403	0.9403	0.9403	0.9403
0.0	4.1522	2400	0.4349	0.9403	0.9403	0.9403	0.9403
0.0	4.2388	2450	0.4348	0.9397	0.9397	0.9398	0.9397
0.0	4.3253	2500	0.4346	0.9398	0.9398	0.9399	0.9398
0.0	4.4118	2550	0.4342	0.9398	0.9398	0.9399	0.9398
0.0	4.4983	2600	0.4357	0.9397	0.9397	0.9398	0.9397
0.0	4.5848	2650	0.4357	0.9399	0.9399	0.9400	0.9399
0.0	4.6713	2700	0.4351	0.9397	0.9397	0.9398	0.9397
0.0	4.7578	2750	0.4350	0.9401	0.9402	0.9402	0.9401
0.0	4.8443	2800	0.4350	0.9395	0.9395	0.9395	0.9395
0.0	4.9308	2850	0.4359	0.9399	0.9399	0.9400	0.9399

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1