eo_train1-20_eval1-10_lr1e-5

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2541
• Accuracy: 0.85

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: 128
• eval_batch_size: 128
• seed: 7658372
• 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: cosine
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 6000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0	0	2.7136	0.0
0.706	100.0	100	0.7050	0.5
0.6611	200.0	200	0.6612	0.55
0.6492	300.0	300	0.6500	0.575
0.6439	400.0	400	0.6421	0.55
0.6361	500.0	500	0.6336	0.55
0.627	600.0	600	0.6189	0.575
0.6182	700.0	700	0.6307	0.575
0.5771	800.0	800	0.5775	0.6
0.5633	900.0	900	0.5664	0.625
0.5517	1000.0	1000	0.5497	0.625
0.5317	1100.0	1100	0.5323	0.65
0.5331	1200.0	1200	0.5204	0.65
0.6811	1300.0	1300	0.5536	0.6
0.523	1400.0	1400	0.5144	0.65
0.4899	1500.0	1500	0.4920	0.65
0.4893	1600.0	1600	0.4854	0.675
0.5072	1700.0	1700	0.4797	0.675
0.4647	1800.0	1800	0.4675	0.675
0.6787	1900.0	1900	0.5977	0.6
0.4529	2000.0	2000	0.4521	0.7
0.4423	2100.0	2100	0.4567	0.7
0.4773	2200.0	2200	0.4749	0.675
0.4376	2300.0	2300	0.4358	0.7
0.4268	2400.0	2400	0.4237	0.7
0.4209	2500.0	2500	0.4199	0.7
0.4186	2600.0	2600	0.4201	0.725
0.4003	2700.0	2700	0.3993	0.7
0.3971	2800.0	2800	0.3942	0.725
0.4315	2900.0	2900	0.4076	0.725
0.3946	3000.0	3000	0.3889	0.7
0.4415	3100.0	3100	0.4902	0.675
0.3844	3200.0	3200	0.3857	0.75
0.368	3300.0	3300	0.3683	0.75
0.3581	3400.0	3400	0.3578	0.775
0.3529	3500.0	3500	0.3477	0.775
0.4454	3600.0	3600	0.3698	0.75
0.3518	3700.0	3700	0.3645	0.75
0.3441	3800.0	3800	0.3424	0.75
0.4046	3900.0	3900	0.3657	0.75
0.3285	4000.0	4000	0.3271	0.775
0.3245	4100.0	4100	0.3212	0.775
0.3265	4200.0	4200	0.3187	0.8
0.3302	4300.0	4300	0.3496	0.775
0.3266	4400.0	4400	0.3087	0.825
0.3803	4500.0	4500	0.4303	0.775
0.2938	4600.0	4600	0.2937	0.825
0.2908	4700.0	4700	0.2899	0.85
0.343	4800.0	4800	0.3310	0.775
0.2851	4900.0	4900	0.2857	0.85
0.2808	5000.0	5000	0.2803	0.85
0.2748	5100.0	5100	0.2767	0.85
0.271	5200.0	5200	0.2708	0.85
0.2683	5300.0	5300	0.2680	0.85
0.2654	5400.0	5400	0.2652	0.85
0.2619	5500.0	5500	0.2619	0.85
0.258	5600.0	5600	0.2579	0.85
0.2556	5700.0	5700	0.2556	0.85
0.2545	5800.0	5800	0.2545	0.85
0.2541	5900.0	5900	0.2541	0.85
0.2541	6000.0	6000	0.2541	0.85

Framework versions

• Transformers 4.46.0
• Pytorch 2.5.1
• Datasets 3.1.0
• Tokenizers 0.20.1