mistral_7b_v_Magiccoder_evol_10k

This model is a fine-tuned version of mistralai/Mistral-7B-v0.3 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.1309

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: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 0.02
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
1.1655	0.0261	4	1.1621
1.0912	0.0523	8	1.1461
1.1696	0.0784	12	1.1691
1.0845	0.1046	16	1.1686
1.1548	0.1307	20	1.1627
1.0495	0.1569	24	1.1600
1.192	0.1830	28	1.1692
1.1397	0.2092	32	1.1666
1.0956	0.2353	36	1.1564
1.1748	0.2614	40	1.1647
1.1924	0.2876	44	1.1664
1.1258	0.3137	48	1.1596
1.1319	0.3399	52	1.1619
1.1099	0.3660	56	1.1577
1.122	0.3922	60	1.1573
1.1749	0.4183	64	1.1538
1.0708	0.4444	68	1.1579
1.0763	0.4706	72	1.1419
1.0635	0.4967	76	1.1494
1.1717	0.5229	80	1.1519
1.0674	0.5490	84	1.1404
1.1492	0.5752	88	1.1620
1.2029	0.6013	92	1.1477
1.1744	0.6275	96	1.1346
1.104	0.6536	100	1.1438
1.1398	0.6797	104	1.1436
1.1296	0.7059	108	1.1409
1.0167	0.7320	112	1.1469
1.1048	0.7582	116	1.1396
1.1004	0.7843	120	1.1358
1.1283	0.8105	124	1.1333
1.1287	0.8366	128	1.1322
1.1421	0.8627	132	1.1315
1.0848	0.8889	136	1.1303
1.184	0.9150	140	1.1300
1.0453	0.9412	144	1.1304
1.0604	0.9673	148	1.1307
1.2116	0.9935	152	1.1309

Framework versions

• PEFT 0.7.1
• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1