Summary4500_M2_1000steps_1e8rate_SFT

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

• Loss: 1.9634

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: 1e-08
• train_batch_size: 2
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss
1.9563	0.0447	50	1.9699
1.9559	0.0895	100	1.9696
1.9636	0.1342	150	1.9675
1.9608	0.1790	200	1.9666
1.9525	0.2237	250	1.9654
1.9514	0.2685	300	1.9645
1.9704	0.3132	350	1.9644
1.9596	0.3579	400	1.9639
1.9558	0.4027	450	1.9641
1.9481	0.4474	500	1.9635
1.945	0.4922	550	1.9639
1.9532	0.5369	600	1.9634
1.955	0.5817	650	1.9642
1.9589	0.6264	700	1.9635
1.9638	0.6711	750	1.9632
1.9679	0.7159	800	1.9634
1.9484	0.7606	850	1.9634
1.9593	0.8054	900	1.9634
1.9598	0.8501	950	1.9634
1.9584	0.8949	1000	1.9634

Framework versions

• Transformers 4.42.3
• Pytorch 2.0.0+cu117
• Datasets 2.20.0
• Tokenizers 0.19.1