mistral_instruct_v4_KQL

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.2 on the generator dataset. It achieves the following results on the evaluation set:

• Loss: 0.3304

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: 0.0002
• 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: constant
• lr_scheduler_warmup_steps: 0.03
• training_steps: 200

Training results

Training Loss	Epoch	Step	Validation Loss
0.981	0.1709	20	0.8715
0.6642	0.3419	40	0.5926
0.5325	0.5128	60	0.4796
0.4483	0.6838	80	0.4293
0.4118	0.8547	100	0.3956
0.3709	1.0256	120	0.3746
0.3597	1.1966	140	0.3595
0.3584	1.3675	160	0.3490
0.3235	1.5385	180	0.3378
0.3314	1.7094	200	0.3304

Framework versions

• PEFT 0.11.1
• Transformers 4.41.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1