tyzhu's picture
Model save
d3cec89 verified
|
raw
history blame
No virus
2.67 kB
metadata
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3_lora2
    results: []

lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3_lora2

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.4374
  • Accuracy: 0.6510

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: 2
  • eval_batch_size: 2
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 4
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • total_eval_batch_size: 8
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: constant
  • lr_scheduler_warmup_ratio: 0.05
  • num_epochs: 20.0

Training results

Training Loss Epoch Step Validation Loss Accuracy
1.168 1.0 187 1.1156 0.6787
1.0838 2.0 375 1.1152 0.6790
0.9936 3.0 562 1.1289 0.6781
0.8717 4.0 750 1.1731 0.6742
0.7555 5.0 937 1.2357 0.6711
0.6292 6.0 1125 1.3174 0.6674
0.5073 7.0 1312 1.4164 0.6638
0.4021 8.0 1500 1.4974 0.6621
0.3017 9.0 1687 1.6756 0.6576
0.2292 10.0 1875 1.8038 0.6548
0.1766 11.0 2062 1.9233 0.6539
0.1406 12.0 2250 2.0465 0.6519
0.1146 13.0 2437 2.1205 0.6518
0.0988 14.0 2625 2.1974 0.6523
0.0915 15.0 2812 2.2592 0.6519
0.0859 16.0 3000 2.3234 0.6515
0.0785 17.0 3187 2.3664 0.6515
0.0761 18.0 3375 2.3999 0.6507
0.0767 19.0 3562 2.4329 0.6517
0.0756 19.95 3740 2.4374 0.6510

Framework versions

  • Transformers 4.34.0
  • Pytorch 2.1.0+cu121
  • Datasets 2.18.0
  • Tokenizers 0.14.1