results
This model is a fine-tuned version of kingkim/kodialogpt_v1.1_SecurityManual on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9083
Model description
ν΄λΉ λͺ¨λΈμ λΉλ© 보μ 맀λ΄μΌμ νμ΅ν νμΈνλλ λͺ¨λΈλ‘, 보μ μν©μμμ λμ² λ°©λ²μ μ§λ¬Έκ³Ό λ΅λ³ νμμΌλ‘ νμ΅νμ΅λλ€. μ μΈ΅λΆ λ° κ³ μΈ΅λΆ νμ¬ λμ, μΉ¨μ μ λ°μ μ λμ λ°©λ² λ± λ€μν 보μ μν©μμμ 맀λ΄μΌμ λ°νμΌλ‘ νλ ¨λμμ΅λλ€.
Intended uses & limitations
μ©λ:
- 보μ μμ νλ ¨ μμ€ν
- λΉλ© 보μ κ΄λ ¨ μλ΅ μμ€ν
- 맀λ΄μΌ κΈ°λ° μ±λ΄ μμ© νλ‘κ·Έλ¨
μ ν μ¬ν:
- λͺ¨λΈμ λΉλ© 보μ 맀λ΄μΌ λ°μ΄ν°λ§μ κΈ°λ°μΌλ‘ νμ΅λμμΌλ©°, λ€λ₯Έ 보μ μλ리μ€μ λν μΌλ°ν λ₯λ ₯μ μ νμ μΌ μ μμ΅λλ€.
Training and evaluation data
λ³Έ λͺ¨λΈμ λΉλ© 보μ 맀λ΄μΌμ λ°νμΌλ‘ λ§λ€μ΄μ§ λ°μ΄ν°μ μ κΈ°λ°μΌλ‘ νμ΅λμμ΅λλ€. kingkim/DS_Building_SecurityManual kingkim/DS_Building_SecurityManual_V3 kingkim/DS_Building_SecurityManual_V5 λ°μ΄ν°μ μ μ€μ 맀λ΄μΌμ λ°νμΌλ‘ μμ±λμμΌλ©°, 200κ° μ΄μμ 보μ μλ리μ€λ₯Ό ν¬ν¨νκ³ μμ΅λλ€.
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
2.2114 | 0.9565 | 11 | 1.5838 |
0.8439 | 2.0 | 23 | 1.2342 |
0.6033 | 2.9565 | 34 | 0.9828 |
0.3294 | 4.0 | 46 | 0.9062 |
0.2423 | 4.7826 | 55 | 0.9083 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.19.1
- Downloads last month
- 6
Model tree for kingkim/kodialogpt_v3.0_SecurityManual
Base model
heegyu/kodialogpt-v1