Edit model card

gemma7b-summarize-gpt4o-80k

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

  • Loss: 5.1111

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: 15

Training results

Training Loss Epoch Step Validation Loss
1.2272 1.0 111 2.3900
0.9374 2.0 222 2.1928
0.8471 3.0 333 2.1682
0.7873 4.0 444 2.2036
0.685 5.0 555 2.2977
0.6223 6.0 666 2.4441
0.5378 7.0 777 2.6715
0.458 8.0 888 2.9555
0.3843 9.0 999 3.4365
0.3241 10.0 1110 3.8823
0.2825 11.0 1221 4.4044
0.2549 12.0 1332 4.8382
0.2408 13.0 1443 5.0611
0.2361 14.0 1554 5.1061
0.2319 15.0 1665 5.1111

Framework versions

  • PEFT 0.11.1
  • Transformers 4.41.1
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
6
Unable to determine this model’s pipeline type. Check the docs .

Adapter for

Dataset used to train llama-duo/gemma7b-summarize-gpt4o-80k