llama-7b-SFT_ds_wiki_1024_full_r_64_alpha_16
This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.2375
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: 0.0002
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.275 | 0.3 | 153 | 1.2611 |
1.2629 | 0.6 | 306 | 1.2488 |
1.2662 | 0.9 | 459 | 1.2431 |
1.1963 | 1.2 | 612 | 1.2454 |
1.2011 | 1.5 | 765 | 1.2411 |
1.1941 | 1.8 | 918 | 1.2375 |
1.1101 | 2.1 | 1071 | 1.2509 |
1.098 | 2.4 | 1224 | 1.2506 |
1.1113 | 2.7 | 1377 | 1.2466 |
1.1321 | 3.0 | 1530 | 1.2449 |
Framework versions
- Transformers 4.32.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3
Model tree for dhmeltzer/llama-7b-SFT_ds_wiki65k_1024_r_64_alpha_16
Base model
meta-llama/Llama-2-7b-hf