gemma7b-summarize-gpt4o-16k

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

• Loss: 2.6528

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: 8
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• total_eval_batch_size: 16
• 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
19.7993	0.9818	27	7.7915
2.9592	2.0	55	3.8534
1.6171	2.9818	82	2.8793
1.3574	4.0	110	2.7211
1.2908	4.9818	137	2.6808
1.2355	6.0	165	2.6636
1.2095	6.9818	192	2.6539
1.1869	8.0	220	2.6529
1.1877	8.9818	247	2.6520
1.1898	9.8182	270	2.6528

Framework versions

• PEFT 0.10.0
• Transformers 4.40.0
• Pytorch 2.1.2+cu121
• Datasets 2.18.0
• Tokenizers 0.19.1