llama-1b-sst-5

This model is a fine-tuned version of meta-llama/Llama-3.2-1B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.2811
• Accuracy: 0.4423
• Precision: 0.4403
• Recall: 0.4105
• F1: 0.4139

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: 0.0002
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
No log	1.4944	100	1.5805	0.3451	0.3330	0.3102	0.3019
No log	2.9888	200	1.3920	0.4205	0.4214	0.3813	0.3823
No log	4.4794	300	1.3328	0.4287	0.4343	0.3898	0.3924
No log	5.9738	400	1.3083	0.4278	0.4275	0.4091	0.4084
5.817	7.4644	500	1.2855	0.4414	0.4415	0.4096	0.4150
5.817	8.9588	600	1.2811	0.4423	0.4403	0.4105	0.4139

Framework versions

• PEFT 0.14.0
• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0