metadata
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: >-
lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3__home_aiops_zhuty_lm_indexer_data_tyzhu_
results: []
library_name: peft
lmind_nq_train6000_eval6489_v1_reciteonly_qa_v3__home_aiops_zhuty_lm_indexer_data_tyzhu_
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0955
- Accuracy: 0.7196
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: 1
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 8
- 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 |
---|---|---|---|---|
0.7054 | 0.9973 | 187 | 0.5535 | 0.7686 |
0.4975 | 2.0 | 375 | 0.5416 | 0.7693 |
0.422 | 2.9973 | 562 | 0.5611 | 0.7645 |
0.3527 | 4.0 | 750 | 0.6100 | 0.7573 |
0.2941 | 4.9973 | 937 | 0.6599 | 0.7522 |
0.2518 | 6.0 | 1125 | 0.7200 | 0.7458 |
0.2138 | 6.9973 | 1312 | 0.7651 | 0.7421 |
0.1824 | 8.0 | 1500 | 0.8280 | 0.7379 |
0.1481 | 8.9973 | 1687 | 0.8700 | 0.7355 |
0.1298 | 10.0 | 1875 | 0.9146 | 0.7329 |
0.1167 | 10.9973 | 2062 | 0.9337 | 0.7309 |
0.1094 | 12.0 | 2250 | 0.9733 | 0.7281 |
0.1052 | 12.9973 | 2437 | 0.9980 | 0.7266 |
0.1007 | 14.0 | 2625 | 1.0022 | 0.7256 |
0.0971 | 14.9973 | 2812 | 1.0422 | 0.7234 |
0.0954 | 16.0 | 3000 | 1.0441 | 0.7236 |
0.0888 | 16.9973 | 3187 | 1.0574 | 0.7223 |
0.0879 | 18.0 | 3375 | 1.0728 | 0.7216 |
0.0879 | 18.9973 | 3562 | 1.0768 | 0.7200 |
0.0883 | 19.9467 | 3740 | 1.0955 | 0.7196 |
Framework versions
- PEFT 0.5.0
- Transformers 4.40.2
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.19.1