modernbert-CGEdit-AAEens3_final

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

• Loss: 0.9248

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use 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_steps: 0.1
• num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss
3.7277	1.0	121	0.9311
3.7164	2.0	242	0.9295
3.7158	3.0	363	0.9286
3.7201	4.0	484	0.9284
3.6646	5.0	605	0.9274
3.6613	6.0	726	0.9280
3.6436	7.0	847	0.9282
3.6323	8.0	968	0.9269
3.6019	9.0	1089	0.9260
3.4985	10.0	1210	0.9261
3.6265	11.0	1331	0.9251
3.5818	12.0	1452	0.9251
3.6067	13.0	1573	0.9252
3.6003	14.0	1694	0.9252
3.6067	15.0	1815	0.9251
3.6046	16.0	1936	0.9247
3.5877	17.0	2057	0.9248
3.5848	18.0	2178	0.9247
3.6011	19.0	2299	0.9247
3.5199	20.0	2420	0.9248
3.6095	21.0	2541	0.9248
3.6102	22.0	2662	0.9247
3.6265	23.0	2783	0.9248
3.6060	24.0	2904	0.9247
3.5859	25.0	3025	0.9247
3.5986	26.0	3146	0.9247
3.5762	27.0	3267	0.9248
3.6020	28.0	3388	0.9247
3.5988	29.0	3509	0.9248

Framework versions

• Transformers 5.0.0
• Pytorch 2.5.1+cu121
• Tokenizers 0.22.1