Edit model card

gemma7b-summarize-11k

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

  • Loss: 2.6219

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
2.8367 0.9907 53 2.7327
1.0189 2.0 107 2.3246
0.8788 2.9907 160 2.2786
0.8104 4.0 214 2.2754
0.7626 4.9907 267 2.3205
0.6665 6.0 321 2.3903
0.6015 6.9907 374 2.4630
0.5699 8.0 428 2.5837
0.5017 8.9907 481 2.6300
0.4969 9.9065 530 2.6219

Framework versions

  • PEFT 0.10.0
  • Transformers 4.40.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
0
Unable to determine this model’s pipeline type. Check the docs .

Adapter for

Dataset used to train llama-duo/gemma7b-summarize-gemini1.0pro-11k