|
--- |
|
license: apache-2.0 |
|
library_name: peft |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
base_model: mistralai/Mixtral-8x7B-v0.1 |
|
model-index: |
|
- name: mixtral_mix_2 |
|
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. --> |
|
|
|
# mixtral_mix_2 |
|
|
|
This model is a fine-tuned version of [mistralai/Mixtral-8x7B-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1) on the generator dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2832 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 32 |
|
- total_train_batch_size: 64 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_ratio: 0.03 |
|
- training_steps: 200 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.686 | 0.36 | 10 | 1.1952 | |
|
| 1.0961 | 0.71 | 20 | 0.9786 | |
|
| 0.9359 | 1.07 | 30 | 0.8352 | |
|
| 0.7603 | 1.42 | 40 | 0.6454 | |
|
| 0.5832 | 1.78 | 50 | 0.4951 | |
|
| 0.4764 | 2.14 | 60 | 0.4361 | |
|
| 0.4172 | 2.49 | 70 | 0.3998 | |
|
| 0.3911 | 2.85 | 80 | 0.3735 | |
|
| 0.37 | 3.2 | 90 | 0.3564 | |
|
| 0.3408 | 3.56 | 100 | 0.3407 | |
|
| 0.3454 | 3.92 | 110 | 0.3270 | |
|
| 0.3153 | 4.27 | 120 | 0.3142 | |
|
| 0.3025 | 4.63 | 130 | 0.3046 | |
|
| 0.3076 | 4.98 | 140 | 0.2963 | |
|
| 0.2791 | 5.34 | 150 | 0.2903 | |
|
| 0.2907 | 5.7 | 160 | 0.2863 | |
|
| 0.2787 | 6.05 | 170 | 0.2840 | |
|
| 0.284 | 6.41 | 180 | 0.2832 | |
|
| 0.264 | 6.76 | 190 | 0.2832 | |
|
| 0.2734 | 7.12 | 200 | 0.2832 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.7.2.dev0 |
|
- Transformers 4.38.1 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.16.1 |
|
- Tokenizers 0.15.0 |
