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_arb_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: 10045
training_steps: 100456
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
5.3332	0.9954	10000	5.2991
4.631	1.9909	20000	4.6242
4.3257	2.9863	30000	4.4051
4.0724	3.9817	40000	4.3142
3.8752	4.9771	50000	4.2919
3.6474	5.9726	60000	4.3248
3.3972	6.9680	70000	4.3988
3.1615	7.9634	80000	4.5020
2.9207	8.9588	90000	4.6064
2.7139	9.9542	100000	4.6790

Safetensors

Model size

0.2B params

Tensor type

F32