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_swe_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: 10870
training_steps: 108705
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
5.9258	0.9199	10000	5.8786
5.0954	1.8399	20000	5.0799
4.7106	2.7598	30000	4.8407
4.4249	3.6798	40000	4.7402
4.1411	4.5998	50000	4.7186
3.849	5.5197	60000	4.7632
3.5485	6.4397	70000	4.8468
3.2413	7.3597	80000	4.9499
2.9692	8.2796	90000	5.0563
2.7408	9.1996	100000	5.1414

Safetensors

Model size

0.2B params

Tensor type

F32