Edit model card

mistral-sigoIAspirantes-Orca-oass-500-gpu

This model is a fine-tuned version of NickyNicky/Mistral-7B-OpenOrca-oasst_top1_2023-08-25-v2 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1371

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: 2.5e-05
  • train_batch_size: 2
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 1
  • training_steps: 500
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
1.6909 0.3521 25 1.2135
1.026 0.7042 50 0.8374
0.7146 1.0563 75 0.6276
0.5094 1.4085 100 0.4925
0.3916 1.7606 125 0.3939
0.3408 2.1127 150 0.3084
0.1724 2.4648 175 0.2717
0.2586 2.8169 200 0.2026
0.1434 3.1690 225 0.1940
0.1253 3.5211 250 0.1579
0.1197 3.8732 275 0.1526
0.0792 4.2254 300 0.1582
0.0937 4.5775 325 0.1579
0.0898 4.9296 350 0.1381
0.0717 5.2817 375 0.1386
0.0665 5.6338 400 0.1350
0.0773 5.9859 425 0.1325
0.0602 6.3380 450 0.1394
0.055 6.6901 475 0.1377
0.0602 7.0423 500 0.1371

Framework versions

  • PEFT 0.10.1.dev0
  • Transformers 4.41.0.dev0
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1
Downloads last month
0
Safetensors
Model size
3.86B params
Tensor type
F32
·
FP16
·
U8
·
Inference API
Unable to determine this model’s pipeline type. Check the docs .

Model tree for fergos80/mistral-sigoIAspirantes-Orca-oass-500-gpu