6b07a6e3-70f3-4d98-8dd5-5b05ea3a13c2

This model is a fine-tuned version of bigcode/starcoder2-3b on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.4154

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: 0.000202
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_BNB 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: 50
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0018	1	2.9028
4.6639	0.0877	50	2.4381
4.0149	0.1754	100	2.2864
3.6677	0.2632	150	2.2556
3.6976	0.3509	200	2.1955
3.4438	0.4386	250	1.8527
2.948	0.5263	300	1.6935
3.2375	0.6140	350	1.5908
3.1726	0.7018	400	1.4535
2.9842	0.7895	450	1.4186
2.6552	0.8772	500	1.4154

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1