This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.2 on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.7900

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: linear
training_steps: 30

Training results

Training Loss	Epoch	Step	Validation Loss
2.0565	0.03	1	1.9073
1.9762	0.06	2	1.8026
1.7179	0.1	3	1.7420
1.7005	0.13	4	1.6867
1.5088	0.16	5	1.6289
1.5717	0.19	6	1.5767
1.6219	0.23	7	1.5234
1.552	0.26	8	1.4657
1.3237	0.29	9	1.4097
1.2997	0.32	10	1.3543
1.1756	0.35	11	1.2965
1.2622	0.39	12	1.2412
1.148	0.42	13	1.1946
1.2081	0.45	14	1.1595
1.0854	0.48	15	1.1244
1.2107	0.52	16	1.0912
1.0053	0.55	17	1.0595
1.025	0.58	18	1.0267
1.1565	0.61	19	0.9961
0.6961	0.65	20	0.9649
0.9958	0.68	21	0.9365
0.7097	0.71	22	0.9085
0.8728	0.74	23	0.8838
0.4737	0.77	24	0.8610
0.5576	0.81	25	0.8418
0.9218	0.84	26	0.8261
0.8374	0.87	27	0.8123
0.5547	0.9	28	0.8016
0.4174	0.94	29	0.7938
0.8493	0.97	30	0.7900

Framework versions

PEFT 0.7.1
Transformers 4.37.0
Pytorch 2.0.0
Datasets 2.16.1
Tokenizers 0.15.1

dancingninjas/Mistral-7b-Javelin-med

Mistral-7b-Javelin-med

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

dancingninjas/
Mistral-7b-Javelin-med