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 Loss	Epoch	Step	Validation Loss
3.7300	1.0	152	0.9321
3.7185	2.0	304	0.9307
3.7013	3.0	456	0.9294
3.6962	4.0	608	0.9297
3.6019	5.0	760	0.9298
3.6561	6.0	912	0.9297
3.6522	7.0	1064	0.9291
3.6127	8.0	1216	0.9294
3.5965	9.0	1368	0.9279
3.5150	10.0	1520	0.9278
3.6231	11.0	1672	0.9280
3.5786	12.0	1824	0.9274
3.6104	13.0	1976	0.9281
3.5895	14.0	2128	0.9276
3.5172	15.0	2280	0.9274
3.6367	16.0	2432	0.9276
3.6046	17.0	2584	0.9274
3.5938	18.0	2736	0.9273
3.5756	19.0	2888	0.9273
3.5059	20.0	3040	0.9271
3.5772	21.0	3192	0.9273
3.5817	22.0	3344	0.9272
3.5915	23.0	3496	0.9272
3.5628	24.0	3648	0.9272
3.4553	25.0	3800	0.9272

SociauxLing/modernbert-CGEdit-AAE_sv3_d3_final

modernbert-CGEdit-AAE_sv3_d3_final