mistral7b_instruct_generation

This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on the generator dataset. It achieves the following results on the evaluation set:

  • Loss: 1.7350

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: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: constant
  • lr_scheduler_warmup_steps: 0.03
  • training_steps: 500

Training results

Training Loss Epoch Step Validation Loss
1.8197 0.0 20 1.8244
1.8114 0.01 40 1.7856
1.8041 0.01 60 1.7743
1.6704 0.01 80 1.7668
1.7122 0.01 100 1.7625
1.7686 0.02 120 1.7556
1.7956 0.02 140 1.7551
1.8042 0.02 160 1.7523
1.7853 0.03 180 1.7500
1.7753 0.03 200 1.7508
1.7498 0.03 220 1.7474
1.8012 0.03 240 1.7459
1.6975 0.04 260 1.7435
1.8533 0.04 280 1.7436
1.8107 0.04 300 1.7426
1.8557 0.04 320 1.7425
1.9098 0.05 340 1.7413
1.8075 0.05 360 1.7426
1.7687 0.05 380 1.7413
1.852 0.06 400 1.7392
1.7799 0.06 420 1.7388
1.822 0.06 440 1.7367
1.804 0.06 460 1.7368
1.8185 0.07 480 1.7365
1.7878 0.07 500 1.7350

Framework versions

  • PEFT 0.10.0
  • Transformers 4.39.1
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2
Downloads last month
4
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 ericrisco/mistral7b_instruct_generation

Adapter
(1457)
this model