sparql-code-llama

This model is a fine-tuned version of codellama/CodeLlama-7b-hf on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4088

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: 0.0003
• train_batch_size: 32
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• training_steps: 400
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.695	0.0838	50	0.7324
0.4445	0.1676	100	0.5458
0.4184	0.2515	150	0.4852
0.4081	0.3353	200	0.4599
0.3996	0.4191	250	0.4389
0.3944	0.5029	300	0.4256
0.3925	0.5868	350	0.4150
0.3957	0.6706	400	0.4088

Framework versions

• PEFT 0.13.2
• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0