adiyghub/mistral_7b_openorca_small_qlora_finetuned

results

This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on 800 examples from OpenOrca. It achieves the following results on the evaluation set:

• Loss: 1.3820

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: 0.0001
• 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: 50
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
1.6764	0.25	50	1.5037
1.3936	0.5	100	1.4108
1.4016	0.75	150	1.3547
1.3152	1.0	200	1.3373
1.0919	1.25	250	1.3368
1.1336	1.5	300	1.3290
1.0541	1.75	350	1.3222
1.0844	2.0	400	1.3174
0.8657	2.25	450	1.3790
0.8606	2.5	500	1.3779
0.8483	2.75	550	1.3824
0.8258	3.0	600	1.3820

Framework versions

• PEFT 0.10.0
• Transformers 4.39.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2