Llama3-8B-lora-finetuned-epoch-5 – AI Model by Siqi-Hu | AlphaNeural AI

You can deploy this model and start earning money today!

Siqi-Hu/
Llama3-8B-lora-finetuned-epoch-5

like

0

generated_from_trainer

meta-llama/Meta-Llama-3-8B

Views

No views yet

Llama3-8B-lora-finetuned-epoch-5

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B on the None dataset. It achieves the following results on the evaluation set:

Loss: 2.3390

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: 5e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 100
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
3.2669	0.4598	20	2.9447
3.0258	0.9195	40	2.6991
2.7679	1.3793	60	2.5032
2.6345	1.8391	80	2.4008
2.4524	2.2989	100	2.3298
2.4312	2.7586	120	2.3044
2.2706	3.2184	140	2.3276
2.2474	3.6782	160	2.3182
2.1656	4.1379	180	2.3390

Framework versions

PEFT 0.15.2
Transformers 4.45.2
Pytorch 2.5.0+cu121
Datasets 3.2.0
Tokenizers 0.20.3