Hugging Face's logo

thrunlab
/
Mistral_Sparse_refined_web_50p_graceful_True

Text Generation
Transformers
Safetensors
sparse_mistral
Generated from Trainer
custom_code
Model card Files Community
Mistral_Sparse_refined_web_50p_graceful_True / README.md
lukeleeai's picture
lukeleeai
End of training
c7c20c6 verified
preview code
|
raw
history blame
4 kB
metadata
license: apache-2.0
base_model: mistralai/Mistral-7B-v0.1
tags:
  - generated_from_trainer
model-index:
  - name: Mistral_Sparse_refined_web_50p_graceful_True
    results: []

Mistral_Sparse_refined_web_50p_graceful_True

This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 2.3402

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: 1e-05
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 0
  • distributed_type: multi-GPU
  • num_devices: 4
  • gradient_accumulation_steps: 4
  • 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: linear
  • training_steps: 2500

Training results

Training Loss Epoch Step Validation Loss
3.7252 0.01 50 2.3893
2.2531 0.02 100 2.4723
2.32 0.02 150 2.4385
2.2363 0.03 200 2.4210
2.3078 0.04 250 2.4118
2.2389 0.05 300 2.4025
2.0902 0.06 350 2.3984
2.2878 0.06 400 2.3965
2.2485 0.07 450 2.3924
2.2375 0.08 500 2.3895
2.1901 0.09 550 2.3909
2.1128 0.1 600 2.3886
2.2983 0.1 650 2.3892
2.2547 0.11 700 2.3873
2.1322 0.12 750 2.3861
2.2715 0.13 800 2.3827
2.263 0.14 850 2.3845
2.2066 0.14 900 2.3836
2.2781 0.15 950 2.3837
2.2597 0.16 1000 2.3778
2.2642 0.17 1050 2.3764
2.2296 0.18 1100 2.3805
2.2289 0.18 1150 2.3784
2.1372 0.19 1200 2.3773
2.2059 0.2 1250 2.3732
2.2847 0.21 1300 2.3719
2.1404 0.22 1350 2.3739
2.2261 0.22 1400 2.3752
2.1713 0.23 1450 2.3750
2.1787 0.24 1500 2.3732
2.1866 0.25 1550 2.3759
2.2471 0.26 1600 2.3760
2.307 0.26 1650 2.3745
2.2457 0.27 1700 2.3746
2.2265 0.28 1750 2.3775
2.163 0.29 1800 2.3797
2.2411 0.3 1850 2.3760
2.247 0.3 1900 2.3770
2.2449 0.31 1950 2.3749
2.1884 0.32 2000 2.3728
2.1909 0.33 2050 2.3770
2.2813 0.34 2100 2.3773
2.2306 0.34 2150 2.3755
2.2158 0.35 2200 2.3777
2.1557 0.36 2250 2.3783
2.2715 0.37 2300 2.3704
2.2053 0.38 2350 2.3729
2.2541 0.38 2400 2.3715
2.0971 0.39 2450 2.3747
2.2791 0.4 2500 2.3727

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.1.1+cu121
  • Datasets 2.15.0
  • Tokenizers 0.15.0