ania3000/glot500-base-oss-morph

trainer_output

This model is a fine-tuned version of cis-lmu/glot500-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3215
• Accuracy: 95.0787
• Sentence accuracy: 55.7798

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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Sentence accuracy
3.0485	1.0	546	1.6035	71.6425	8.9908
1.2157	2.0	1092	0.7569	84.1410	20.5505
0.6796	3.0	1638	0.4906	89.3392	34.3119
0.4445	4.0	2184	0.3826	91.6174	40.3670
0.3298	5.0	2730	0.3129	93.0522	45.1376
0.2501	6.0	3276	0.3018	93.5809	47.8899
0.2004	7.0	3822	0.2719	93.8578	48.4404
0.1585	8.0	4368	0.2615	94.4493	53.9450
0.1335	9.0	4914	0.2645	94.7011	53.3945
0.1063	10.0	5460	0.2664	94.6885	53.5780
0.072	11.0	6006	0.2641	94.9654	55.5963
0.0606	12.0	6552	0.2749	94.9150	55.4128
0.0484	13.0	7098	0.2852	95.0283	54.6789
0.0415	14.0	7644	0.3005	94.9276	55.5963
0.0344	15.0	8190	0.2984	95.2297	55.7798
0.0303	16.0	8736	0.3103	94.9654	55.0459
0.0247	17.0	9282	0.3146	95.1542	56.1468
0.0201	18.0	9828	0.3197	95.0157	55.2294
0.0193	19.0	10374	0.3215	95.0787	55.7798

Framework versions

• Transformers 4.57.3
• Pytorch 2.11.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2