mistral-finetune

This model is a fine-tuned version of sid321axn/Mistral-7B-text-to-sql-finetuned on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3199

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: 0.00025
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss
0.2637	25.0	25	0.2518
0.0062	50.0	50	0.2701
0.0001	75.0	75	0.3082
0.0152	100.0	100	0.3453
0.0362	125.0	125	0.3130
0.0745	150.0	150	0.2873
0.0064	175.0	175	0.2935
0.0996	200.0	200	0.3202
0.0279	225.0	225	0.3330
0.0046	250.0	250	0.3273
0.0012	275.0	275	0.3133
0.0	300.0	300	0.3142
0.0	325.0	325	0.3159
0.0	350.0	350	0.3173
0.0	375.0	375	0.3185
0.0	400.0	400	0.3191
0.0	425.0	425	0.3196
0.0	450.0	450	0.3197
0.0	475.0	475	0.3199
0.0	500.0	500	0.3199

Framework versions

• PEFT 0.10.0
• Transformers 4.40.2
• Pytorch 2.2.2+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1