metadata
base_model: NousResearch/Llama-2-7b-hf
tags:
- generated_from_trainer
model-index:
- name: llama2_instruct_generation
results: []
llama2_instruct_generation
This model is a fine-tuned version of NousResearch/Llama-2-7b-hf on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.6733
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: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_steps: 0.03
- training_steps: 500
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.9478 | 0.0 | 20 | 1.8129 |
| 1.7992 | 0.0 | 40 | 1.7797 |
| 1.8765 | 0.0 | 60 | 1.7642 |
| 1.8488 | 0.01 | 80 | 1.7527 |
| 1.8512 | 0.01 | 100 | 1.7384 |
| 1.8856 | 0.01 | 120 | 1.7136 |
| 1.8429 | 0.01 | 140 | 1.7035 |
| 1.8258 | 0.01 | 160 | 1.6970 |
| 1.8125 | 0.01 | 180 | 1.6923 |
| 1.7902 | 0.01 | 200 | 1.6898 |
| 1.8622 | 0.02 | 220 | 1.6895 |
| 1.8823 | 0.02 | 240 | 1.6867 |
| 1.7728 | 0.02 | 260 | 1.6837 |
| 1.6331 | 0.02 | 280 | 1.6820 |
| 1.8399 | 0.02 | 300 | 1.6821 |
| 1.735 | 0.02 | 320 | 1.6810 |
| 1.8192 | 0.02 | 340 | 1.6804 |
| 1.7609 | 0.03 | 360 | 1.6771 |
| 1.6517 | 0.03 | 380 | 1.6768 |
| 1.6272 | 0.03 | 400 | 1.6738 |
| 1.7947 | 0.03 | 420 | 1.6751 |
| 1.7794 | 0.03 | 440 | 1.6740 |
| 1.6371 | 0.03 | 460 | 1.6726 |
| 1.7126 | 0.03 | 480 | 1.6730 |
| 1.7321 | 0.04 | 500 | 1.6733 |
Framework versions
- Transformers 4.35.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1