Llama3.1-8b-instruct-SFT-2024-09-20_LoRAs_r64
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: 1.0712
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 |
---|---|---|---|
1.5247 | 0.0793 | 1000 | 1.3914 |
1.341 | 0.1586 | 2000 | 1.3288 |
1.314 | 0.2380 | 3000 | 1.2961 |
1.261 | 0.3173 | 4000 | 1.2635 |
1.2397 | 0.3966 | 5000 | 1.2353 |
1.212 | 0.4759 | 6000 | 1.2240 |
1.2051 | 0.5552 | 7000 | 1.2019 |
1.1845 | 0.6346 | 8000 | 1.1861 |
1.157 | 0.7139 | 9000 | 1.1684 |
1.1418 | 0.7932 | 10000 | 1.1580 |
1.1374 | 0.8725 | 11000 | 1.1385 |
1.1187 | 0.9519 | 12000 | 1.1243 |
1.0841 | 1.0312 | 13000 | 1.1180 |
1.0365 | 1.1105 | 14000 | 1.1062 |
1.0266 | 1.1898 | 15000 | 1.0974 |
1.0216 | 1.2691 | 16000 | 1.0935 |
1.0217 | 1.3485 | 17000 | 1.0818 |
1.0048 | 1.4278 | 18000 | 1.0712 |
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
- 0
Model tree for ccibeekeoc42/Llama3.1-8b-instruct-SFT-2024-09-20_LoRAs_r64
Base model
meta-llama/Meta-Llama-3.1-8B
Finetuned
meta-llama/Meta-Llama-3.1-8B-Instruct
Adapter
this model