metadata

license: llama2
library_name: peft
tags:
  - generated_from_trainer
base_model: meta-llama/Llama-2-13b-hf
metrics:
  - accuracy
  - precision
  - recall
model-index:
  - name: llama2-13B_MT
    results: []

llama2-13B_MT

This model is a fine-tuned version of meta-llama/Llama-2-13b-hf on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.5717
Accuracy: 0.7967
Precision: 0.8296
Recall: 0.7467
F1 score: 0.7860

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: 3

Training results

Training Loss	Epoch	Step	Accuracy	F1 score	Precision	Recall	Validation Loss
0.6169	0.25	200	0.735	0.7145	0.7743	0.6633	0.5712
0.5218	0.5	400	0.765	0.7487	0.8046	0.7	0.5406
0.4976	0.75	600	0.7717	0.7486	0.8327	0.68	0.4991
0.4531	1.0	800	0.7567	0.7045	0.8969	0.58	0.5727
0.339	1.25	1000	0.6176	0.7617	0.9198	0.5733	0.7064
0.3046	1.5	1200	0.5150	0.7767	0.7964	0.7433	0.7690
0.3363	1.75	1400	0.5185	0.795	0.8554	0.71	0.7760
0.3074	2.0	1600	0.4635	0.79	0.7862	0.7967	0.7914
0.2052	2.25	1800	0.5411	0.8	0.8659	0.71	0.7802
0.1841	2.5	2000	0.5959	0.8033	0.8889	0.6933	0.7790
0.1629	2.75	2200	0.5510	0.7933	0.8143	0.76	0.7862
0.1559	3.0	2400	0.5717	0.7967	0.8296	0.7467	0.7860

Framework versions

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