|
--- |
|
license: apache-2.0 |
|
base_model: mosaicml/mpt-7b-instruct |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
model-index: |
|
- name: mpt_1000_STEPS_1e5_SFT_SFT |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# mpt_1000_STEPS_1e5_SFT_SFT |
|
|
|
This model is a fine-tuned version of [mosaicml/mpt-7b-instruct](https://huggingface.co/mosaicml/mpt-7b-instruct) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3857 |
|
|
|
## 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: 2 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 4 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 100 |
|
- training_steps: 1000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 0.5003 | 0.05 | 50 | 0.4827 | |
|
| 0.4582 | 0.1 | 100 | 0.5370 | |
|
| 0.485 | 0.15 | 150 | 0.4682 | |
|
| 0.5202 | 0.2 | 200 | 0.4486 | |
|
| 0.4687 | 0.24 | 250 | 0.4494 | |
|
| 0.469 | 0.29 | 300 | 0.4485 | |
|
| 0.4331 | 0.34 | 350 | 0.4359 | |
|
| 0.4189 | 0.39 | 400 | 0.4268 | |
|
| 0.4444 | 0.44 | 450 | 0.4190 | |
|
| 0.4187 | 0.49 | 500 | 0.4140 | |
|
| 0.4045 | 0.54 | 550 | 0.4090 | |
|
| 0.4159 | 0.59 | 600 | 0.4021 | |
|
| 0.3862 | 0.64 | 650 | 0.3973 | |
|
| 0.3773 | 0.68 | 700 | 0.3930 | |
|
| 0.382 | 0.73 | 750 | 0.3893 | |
|
| 0.3892 | 0.78 | 800 | 0.3873 | |
|
| 0.4079 | 0.83 | 850 | 0.3862 | |
|
| 0.3667 | 0.88 | 900 | 0.3857 | |
|
| 0.3724 | 0.93 | 950 | 0.3857 | |
|
| 0.3908 | 0.98 | 1000 | 0.3857 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.39.3 |
|
- Pytorch 2.0.0+cu117 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
