Llama3.1-8b-instruct-SFT-2024-09-20_LoRAs_r128
This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9433
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: 6
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 1.5
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.4799 | 0.0793 | 1000 | 1.3683 |
1.335 | 0.1586 | 2000 | 1.2939 |
1.2731 | 0.2380 | 3000 | 1.2520 |
1.2285 | 0.3173 | 4000 | 1.2185 |
1.201 | 0.3966 | 5000 | 1.1961 |
1.1713 | 0.4759 | 6000 | 1.1671 |
1.1423 | 0.5552 | 7000 | 1.1484 |
1.1153 | 0.6346 | 8000 | 1.1268 |
1.0999 | 0.7139 | 9000 | 1.1054 |
1.1006 | 0.7932 | 10000 | 1.0916 |
1.0708 | 0.8725 | 11000 | 1.0637 |
1.0577 | 0.9519 | 12000 | 1.0340 |
1.0059 | 1.0312 | 13000 | 1.0188 |
0.9419 | 1.1105 | 14000 | 1.0015 |
0.9204 | 1.1898 | 15000 | 0.9916 |
0.9274 | 1.2691 | 16000 | 0.9803 |
0.9167 | 1.3485 | 17000 | 0.9540 |
0.9004 | 1.4278 | 18000 | 0.9433 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.0.1+cu118
- Datasets 3.0.0
- Tokenizers 0.19.1
- Downloads last month
- 12
Model tree for ccibeekeoc42/Llama3.1-8b-instruct-SFT-2024-09-20_LoRAs_r128
Base model
meta-llama/Llama-3.1-8B
Finetuned
meta-llama/Llama-3.1-8B-Instruct