gemma7b-summarize-gemini1_5flash-256k

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: 2.4690

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
0.9973	0.9988	414	2.5221
0.9036	2.0	829	2.4358
0.7651	2.9988	1243	2.3987
0.7192	4.0	1658	2.3970
0.6986	4.9988	2072	2.4163
0.6737	6.0	2487	2.4236
0.6633	6.9988	2901	2.4494
0.661	8.0	3316	2.4621
0.643	8.9988	3730	2.4791
0.6511	9.9879	4140	2.4690

Framework versions

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