learning_rate: 2e-05
train_batch_size: 64
eval_batch_size: 64
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 128
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: linear
lr_scheduler_warmup_ratio: 0.3
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
1.945	0.7463	50	0.6971	0.7127	0.7107	0.7127	0.7115
1.2749	1.4925	100	0.5650	0.7892	0.7928	0.7892	0.7898
1.1245	2.2388	150	0.6353	0.7491	0.8034	0.7491	0.7572
1.024	2.9851	200	0.4373	0.8358	0.8350	0.8358	0.8354
0.8063	3.7313	250	0.4471	0.8246	0.8413	0.8246	0.8294
0.6552	4.4776	300	0.4742	0.8293	0.8424	0.8293	0.8203
0.5445	5.2239	350	0.4599	0.8535	0.8552	0.8535	0.8515

Qanadil/answerdotai-ModernBERT-large-arabic-fp16-allagree

answerdotai-ModernBERT-large-arabic-fp16-allagree