RiverTest/unmerged7

mistralai/Mistral-7B-Instruct-v0.2

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: 0.5526

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: 2.5e-05
• train_batch_size: 32
• 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: 0.03
• training_steps: 600

Training results

Training Loss	Epoch	Step	Validation Loss
2.7925	0.22	10	2.0998
1.6897	0.43	20	1.3864
1.3495	0.65	30	1.2622
1.2144	0.87	40	1.1882
1.1546	1.09	50	1.1397
1.1002	1.3	60	1.0843
1.0023	1.52	70	0.9794
0.897	1.74	80	0.9370
0.8625	1.96	90	0.8557
0.7492	2.17	100	0.7909
0.7296	2.39	110	0.7455
0.6738	2.61	120	0.7239
0.656	2.83	130	0.7071
0.6289	3.04	140	0.6852
0.5835	3.26	150	0.6704
0.5647	3.48	160	0.6481
0.5416	3.7	170	0.6326
0.5159	3.91	180	0.6219
0.475	4.13	190	0.6091
0.4529	4.35	200	0.5903
0.4358	4.57	210	0.5769
0.4124	4.78	220	0.5574
0.3925	5.0	230	0.5433
0.3325	5.22	240	0.5396
0.3307	5.43	250	0.5241
0.3122	5.65	260	0.5185
0.2973	5.87	270	0.5042
0.2695	6.09	280	0.5082
0.2345	6.3	290	0.5020
0.2307	6.52	300	0.4859
0.2226	6.74	310	0.4771
0.2083	6.96	320	0.4717
0.1858	7.17	330	0.4881
0.1677	7.39	340	0.4791
0.1663	7.61	350	0.4774
0.1609	7.83	360	0.4780
0.1493	8.04	370	0.4820
0.1332	8.26	380	0.4940
0.1351	8.48	390	0.4898
0.1251	8.7	400	0.4894
0.1243	8.91	410	0.4836
0.1121	9.13	420	0.5108
0.1059	9.35	430	0.5055
0.1037	9.57	440	0.4974
0.102	9.78	450	0.4981
0.1032	10.0	460	0.5100
0.0887	10.22	470	0.5267
0.09	10.43	480	0.5231
0.084	10.65	490	0.5228
0.0865	10.87	500	0.5166
0.0838	11.09	510	0.5337
0.0762	11.3	520	0.5444
0.0792	11.52	530	0.5375
0.0765	11.74	540	0.5397
0.0747	11.96	550	0.5386
0.0684	12.17	560	0.5517
0.0697	12.39	570	0.5547
0.0701	12.61	580	0.5528
0.0702	12.83	590	0.5522
0.0693	13.04	600	0.5526

Framework versions

• PEFT 0.7.1
• Transformers 4.36.2
• Pytorch 2.1.2+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0