gemma7b-summarize

This model is a fine-tuned version of google/gemma-7b on the llama-duo/coverage_dataset dataset. It achieves the following results on the evaluation set:

• Loss: 3.3615

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: 4
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• total_eval_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
17.2572	1.0	5	11.2704
10.1138	2.0	10	7.8912
7.7504	3.0	15	7.1928
7.0505	4.0	20	6.8083
6.6144	5.0	25	6.4041
6.066	6.0	30	5.7488
5.1349	7.0	35	4.6655
3.9891	8.0	40	3.7537
3.293	9.0	45	3.3924
3.098	10.0	50	3.3615

Framework versions

• PEFT 0.10.0
• Transformers 4.40.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1