gemma2b-summarize-gemini1_5flash-1k

This model is a fine-tuned version of google/gemma-2b on the llama-duo/synth_summarize_dataset_dedup dataset. It achieves the following results on the evaluation set:

• Loss: 2.7426

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: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• total_eval_batch_size: 32
• 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
2.9957	1.0	2	3.1984
2.9957	2.0	4	3.0673
2.8525	3.0	6	2.9478
2.8525	4.0	8	2.8806
2.3323	5.0	10	2.8382
2.3323	6.0	12	2.8050
2.3323	7.0	14	2.7636
2.0887	8.0	16	2.7495
2.0887	9.0	18	2.7433
1.9997	10.0	20	2.7426

Framework versions

• PEFT 0.11.1
• Transformers 4.40.1
• Pytorch 2.2.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1