Gemma-2-2B_task-2_180-samples_config-1_auto

This model is a fine-tuned version of google/gemma-2-2b-it on the GaetanMichelet/chat-60_ft_task-2_auto, the GaetanMichelet/chat-120_ft_task-2_auto and the GaetanMichelet/chat-180_ft_task-2_auto datasets. It achieves the following results on the evaluation set:

  • Loss: 0.6153

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: 0.0001
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • distributed_type: multi-GPU
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 8
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 50

Training results

Training Loss Epoch Step Validation Loss
0.962 1.0 17 0.9375
0.7695 2.0 34 0.7492
0.6479 3.0 51 0.6654
0.5369 4.0 68 0.6236
0.4817 5.0 85 0.6153
0.4256 6.0 102 0.6260
0.3157 7.0 119 0.7120
0.2073 8.0 136 0.8673
0.1196 9.0 153 1.0612
0.0925 10.0 170 1.1791
0.0527 11.0 187 1.3142
0.031 12.0 204 1.3306

Framework versions

  • PEFT 0.12.0
  • Transformers 4.44.0
  • Pytorch 2.1.2+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1
Downloads last month
3
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and HF Inference API was unable to determine this model’s pipeline type.

Model tree for GaetanMichelet/Gemma-2-2B_task-2_180-samples_config-1_auto

Base model

google/gemma-2-2b
Adapter
(182)
this model

Collection including GaetanMichelet/Gemma-2-2B_task-2_180-samples_config-1_auto