metadata

license: apache-2.0
library_name: peft
tags:
  - generated_from_trainer
base_model: mistralai/Mistral-7B-v0.1
metrics:
  - accuracy
  - precision
  - recall
model-index:
  - name: Mistral_7B_MT
    results: []

Mistral_7B_MT

This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.8388
Accuracy: 0.8167
Precision: 0.8519
Recall: 0.7667
F1 score: 0.8070

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: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Training results

Training Loss	Epoch	Step	Accuracy	F1 score	Precision	Recall	Validation Loss
1.687	0.25	200	0.6233	0.4378	0.8627	0.2933	2.0030
0.9482	0.5	400	0.68	0.5616	0.8913	0.41	1.4557
0.9232	0.75	600	0.72	0.6471	0.875	0.5133	0.8805
0.7781	1.0	800	0.57	0.3246	0.7561	0.2067	1.4515
0.5468	1.25	1000	0.7233	0.6483	0.8895	0.51	0.8474
0.5549	1.5	1200	0.7767	0.7403	0.8843	0.6367	0.7168
0.4883	1.75	1400	0.8	0.7719	0.8982	0.6767	0.6943
0.4639	2.0	1600	0.7767	0.7276	0.9323	0.5967	0.7637
0.3804	2.25	1800	0.7617	0.7146	0.8905	0.5967	0.8467
0.3847	2.5	2000	0.81	0.7942	0.8661	0.7333	0.6699
0.346	2.75	2200	0.7833	0.7575	0.8602	0.6767	0.8569
0.3488	3.0	2400	0.7824	0.815	0.9238	0.6867	0.7878
0.2654	3.25	2600	1.0799	0.7683	0.9259	0.5833	0.7157
0.2506	3.5	2800	0.8567	0.8033	0.9062	0.6767	0.7748
0.2574	3.75	3000	0.7490	0.8083	0.7846	0.85	0.816
0.2137	4.0	3200	0.7665	0.8333	0.8546	0.8033	0.8282
0.1335	4.25	3400	0.8591	0.8133	0.8013	0.8333	0.8170
0.1486	4.5	3600	0.9781	0.83	0.9091	0.7333	0.8118
0.126	4.75	3800	0.8723	0.8217	0.8642	0.7633	0.8106
0.1474	5.0	4000	0.8388	0.8167	0.8519	0.7667	0.8070

Framework versions

PEFT 0.11.1
Transformers 4.44.2
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.19.1