You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

dense_nor_100m_mult

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

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 0.0001
train_batch_size: 8
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 10372
training_steps: 103729
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
6.0481	0.9640	10000	6.0037
5.1835	1.9281	20000	5.1884
4.8254	2.8921	30000	4.9447
4.5387	3.8562	40000	4.8481
4.2689	4.8202	50000	4.8316
3.9664	5.7842	60000	4.8763
3.6776	6.7483	70000	4.9663
3.3734	7.7123	80000	5.0859
3.1048	8.6764	90000	5.1995
2.8348	9.6404	100000	5.2764

Safetensors

Model size

0.2B params

Tensor type

F32