peft-starcoder-lora-a100-adalflow

This model is a fine-tuned version of bigcode/starcoderbase-1b on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.1664

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0005
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• 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_steps: 30
• training_steps: 2000

Training results

Training Loss	Epoch	Step	Validation Loss
0.6214	0.05	100	1.2932
0.2305	0.1	200	1.5985
0.1187	0.15	300	1.7457
0.0805	0.2	400	1.8529
0.0674	0.25	500	1.9213
0.0568	0.3	600	1.9515
0.0513	0.35	700	1.9906
0.0482	0.4	800	2.0139
0.0442	0.45	900	2.0442
0.0429	0.5	1000	2.0661
0.039	0.55	1100	2.1006
0.0372	0.6	1200	2.1061
0.0362	0.65	1300	2.1219
0.0354	0.7	1400	2.1374
0.0331	0.75	1500	2.1450
0.0313	0.8	1600	2.1494
0.0321	0.85	1700	2.1612
0.0321	0.9	1800	2.1665
0.0327	0.95	1900	2.1652
0.0325	1.0	2000	2.1664

Framework versions

• PEFT 0.14.0
• Transformers 4.49.0
• Pytorch 2.6.0+cu124
• Datasets 3.3.1
• Tokenizers 0.21.0