mistral_instruct_generation

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

  • Loss: 1.0225

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: 32
  • 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: 200

Training results

Training Loss Epoch Step Validation Loss
2.1172 1.54 20 1.9621
1.7921 3.08 40 1.6326
1.468 4.62 60 1.3290
1.2425 6.15 80 1.1762
1.1558 7.69 100 1.1167
1.0912 9.23 120 1.0837
1.0749 10.77 140 1.0632
1.0534 12.31 160 1.0472
1.0213 13.85 180 1.0332
0.9962 15.38 200 1.0225

Framework versions

  • PEFT 0.7.1
  • Transformers 4.37.1
  • Pytorch 2.1.0+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.1
Downloads last month
0
Safetensors
Model size
7.24B params
Tensor type
F32
·
Inference API
Unable to determine this model’s pipeline type. Check the docs .

Model tree for Crackerjax2/mistral_instruct_generation

Adapter
(350)
this model