ossetic-encoders/ossbert-morph

trainer_output

This model is a fine-tuned version of AlexeySorokin/ossbert-onc-unlab-from_multilingual-bs64-5epochs on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2729
• Accuracy: 95.5104
• Sentence accuracy: 60.7339

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Sentence accuracy
1.0799	1.0	546	0.3960	90.8605	37.6147
0.3583	2.0	1092	0.2930	93.3725	51.9266
0.2307	3.0	1638	0.2578	94.1742	54.3119
0.1588	4.0	2184	0.2583	94.2945	52.8440
0.1141	5.0	2730	0.2439	94.8557	56.5138
0.0831	6.0	3276	0.2520	95.2031	59.2661
0.0614	7.0	3822	0.2659	95.2699	58.7156
0.0433	8.0	4368	0.2624	95.3234	58.8991
0.0315	9.0	4914	0.2714	95.5772	61.4679
0.0245	10.0	5460	0.2729	95.5104	60.7339

Framework versions

• Transformers 4.57.3
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2