metadata

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

org_aug_a

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: 0.8808
F1 Micro: 0.6615
F1 Macro: 0.6476
F1 Weighted: 0.6615

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
2.2195	0.0154	25	1.5207	0.5691	0.5521	0.5694
1.4371	0.0308	50	1.2747	0.6089	0.5857	0.6052
1.2556	0.0462	75	1.1545	0.6304	0.6036	0.6240
1.2415	0.0615	100	1.0691	0.6320	0.6132	0.6301
0.9864	0.0769	125	1.0264	0.6399	0.6278	0.6411
1.0647	0.0923	150	0.9918	0.6510	0.6266	0.6455
0.9849	0.1077	175	0.9679	0.6576	0.6317	0.6511
1.0067	0.1231	200	0.9383	0.6501	0.6384	0.6513
0.8928	0.1385	225	0.9243	0.6620	0.6405	0.6579
0.9858	0.1538	250	0.9132	0.6627	0.6405	0.6582
0.9085	0.1692	275	0.9011	0.6575	0.6446	0.6581
1.0059	0.1846	300	0.9018	0.6686	0.6436	0.6623
0.8939	0.2	325	0.8928	0.6682	0.6448	0.6629
0.864	0.2154	350	0.8833	0.6622	0.6478	0.6619
0.9499	0.2308	375	0.8837	0.6585	0.6463	0.6593
0.9721	0.2462	400	0.8808	0.6615	0.6476	0.6615

Framework versions

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