metadata
license: other
library_name: peft
tags:
- llama-factory
- lora
- generated_from_trainer
base_model: /data1/model/llama2/meta-llama/Llama2-13b
model-index:
- name: topical_chat_no_sys
results: []
topical_chat_no_sys
This model is a fine-tuned version of /data1/model/llama2/meta-llama/Llama2-13b on the topical_chat_no_sys dataset. It achieves the following results on the evaluation set:
- Loss: 1.8941
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: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 8
- total_eval_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 20
- num_epochs: 5.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.1904 | 0.0472 | 100 | 2.1137 |
| 1.9627 | 0.0944 | 200 | 2.0589 |
| 2.0172 | 0.1416 | 300 | 2.0221 |
| 1.8965 | 0.1889 | 400 | 1.9968 |
| 1.9534 | 0.2361 | 500 | 1.9823 |
| 1.8621 | 0.2833 | 600 | 1.9679 |
| 1.9777 | 0.3305 | 700 | 1.9611 |
| 2.0865 | 0.3777 | 800 | 1.9544 |
| 1.9662 | 0.4249 | 900 | 1.9461 |
| 1.8352 | 0.4721 | 1000 | 1.9376 |
| 1.8973 | 0.5194 | 1100 | 1.9329 |
| 1.9688 | 0.5666 | 1200 | 1.9264 |
| 1.8383 | 0.6138 | 1300 | 1.9192 |
| 1.9032 | 0.6610 | 1400 | 1.9146 |
| 1.9295 | 0.7082 | 1500 | 1.9109 |
| 1.8207 | 0.7554 | 1600 | 1.9061 |
| 1.9119 | 0.8026 | 1700 | 1.9032 |
| 1.8392 | 0.8499 | 1800 | 1.9019 |
| 1.961 | 0.8971 | 1900 | 1.8994 |
| 1.8913 | 0.9443 | 2000 | 1.8945 |
| 1.8187 | 0.9915 | 2100 | 1.8941 |
| 1.7296 | 1.0387 | 2200 | 1.9006 |
| 1.6184 | 1.0859 | 2300 | 1.9040 |
| 1.6973 | 1.1331 | 2400 | 1.9056 |
Framework versions
- PEFT 0.10.0
- Transformers 4.40.0
- Pytorch 2.2.1
- Datasets 2.18.0
- Tokenizers 0.19.1