metadata

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

org_model

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: 0.9527
F1 Micro: 0.8011
F1 Macro: 0.7779
F1 Weighted: 0.8108

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: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
training_steps: 400

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Micro	F1 Macro	F1 Weighted
1.5515	0.0064	25	1.3111	0.7801	0.7504	0.7890
1.2983	0.0127	50	1.2188	0.7748	0.7572	0.7891
1.2193	0.0191	75	1.1271	0.7855	0.7583	0.7937
1.1269	0.0255	100	1.0890	0.7952	0.7639	0.8015
1.0734	0.0318	125	1.0594	0.7949	0.7635	0.8008
1.0384	0.0382	150	1.0389	0.7857	0.7614	0.7937
1.0168	0.0446	175	1.0126	0.8045	0.7794	0.8133
1.0043	0.0510	200	0.9998	0.8034	0.7786	0.8123
1.0406	0.0573	225	0.9874	0.8074	0.7803	0.8153
1.0488	0.0637	250	0.9838	0.7922	0.7664	0.8000
0.9894	0.0701	275	0.9673	0.8034	0.7780	0.8122
0.9969	0.0764	300	0.9629	0.7992	0.7720	0.8069
1.0047	0.0828	325	0.9655	0.8000	0.7689	0.8058
0.9812	0.0892	350	0.9623	0.8049	0.7839	0.8159
0.9681	0.0955	375	0.9551	0.8016	0.7794	0.8118
1.0594	0.1019	400	0.9527	0.8011	0.7779	0.8108

Framework versions

PEFT 0.10.0
Transformers 4.40.2
Pytorch 2.3.0+cu118
Datasets 2.19.0
Tokenizers 0.19.1