learning_rate: 5e-05
train_batch_size: 64
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 256
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: cosine
lr_scheduler_warmup_ratio: 0.05
num_epochs: 5

Training Loss	Epoch	Step	Validation Loss	Spearman	Pearson	Mse
0.0301	0.3998	1055	0.0265	0.9008	0.9017	0.0265
0.0253	0.7997	2110	0.0238	0.9108	0.9126	0.0238
0.0222	1.1995	3165	0.0231	0.9164	0.9179	0.0231
0.0217	1.5994	4220	0.0209	0.9222	0.9232	0.0209
0.0195	1.9992	5275	0.0211	0.9234	0.9254	0.0211
0.0171	2.3991	6330	0.0204	0.9244	0.9275	0.0204
0.0159	2.7989	7385	0.0197	0.9265	0.9291	0.0197
0.0124	3.1988	8440	0.0200	0.9274	0.9282	0.0200
0.0135	3.5986	9495	0.0196	0.9285	0.9301	0.0196
0.014	3.9985	10550	0.0194	0.9285	0.9304	0.0194
0.0103	4.3983	11605	0.0197	0.9287	0.9302	0.0197
0.0119	4.7982	12660	0.0197	0.9289	0.9303	0.0197

gutsartificial/bge-small-en-v1.5-quality-weight-0.4

bge-small-en-v1.5-2024-12-07_18-39-49-quality-weight-0.4