|
--- |
|
license: apache-2.0 |
|
base_model: mistralai/Mistral-7B-v0.1 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: qlora-out |
|
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. --> |
|
|
|
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl) |
|
# qlora-out |
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5477 |
|
|
|
## 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: 0.0004 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 8 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 300 |
|
- num_epochs: 3 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 0.8333 | 0.06 | 20 | 0.6411 | |
|
| 0.6715 | 0.12 | 40 | 0.5899 | |
|
| 0.5905 | 0.18 | 60 | 0.5573 | |
|
| 0.5845 | 0.24 | 80 | 0.5342 | |
|
| 0.5524 | 0.3 | 100 | 0.5260 | |
|
| 0.5516 | 0.36 | 120 | 0.5273 | |
|
| 0.5465 | 0.42 | 140 | 0.5132 | |
|
| 0.439 | 0.48 | 160 | 0.5085 | |
|
| 0.6857 | 0.54 | 180 | 0.4982 | |
|
| 0.7326 | 0.6 | 200 | 0.5096 | |
|
| 0.8225 | 0.66 | 220 | 0.5080 | |
|
| 0.6148 | 0.72 | 240 | 0.4883 | |
|
| 0.4524 | 0.78 | 260 | 0.4970 | |
|
| 0.6084 | 0.84 | 280 | 0.5425 | |
|
| 0.6737 | 0.9 | 300 | 0.5059 | |
|
| 0.459 | 0.96 | 320 | 0.4968 | |
|
| 0.6138 | 1.02 | 340 | 0.5111 | |
|
| 0.4023 | 1.08 | 360 | 0.5499 | |
|
| 0.4406 | 1.14 | 380 | 0.5657 | |
|
| 0.4054 | 1.2 | 400 | 0.5387 | |
|
| 0.4707 | 1.26 | 420 | 0.5698 | |
|
| 0.577 | 1.32 | 440 | 0.5181 | |
|
| 0.279 | 1.38 | 460 | 0.5243 | |
|
| 0.5576 | 1.44 | 480 | 0.5172 | |
|
| 0.382 | 1.5 | 500 | 0.5178 | |
|
| 0.4541 | 1.56 | 520 | 0.5166 | |
|
| 0.339 | 1.62 | 540 | 0.5087 | |
|
| 0.4609 | 1.68 | 560 | 0.5257 | |
|
| 0.4768 | 1.74 | 580 | 0.4990 | |
|
| 0.5313 | 1.8 | 600 | 0.4952 | |
|
| 0.347 | 1.86 | 620 | 0.4823 | |
|
| 0.4216 | 1.92 | 640 | 0.4832 | |
|
| 0.3905 | 1.98 | 660 | 0.4748 | |
|
| 0.1525 | 2.04 | 680 | 0.6280 | |
|
| 0.3269 | 2.1 | 700 | 0.5995 | |
|
| 0.1502 | 2.16 | 720 | 0.5412 | |
|
| 0.1845 | 2.22 | 740 | 0.5421 | |
|
| 0.2009 | 2.28 | 760 | 0.5564 | |
|
| 0.1896 | 2.34 | 780 | 0.5275 | |
|
| 0.1433 | 2.4 | 800 | 0.5569 | |
|
| 0.1758 | 2.46 | 820 | 0.5463 | |
|
| 0.1336 | 2.51 | 840 | 0.5564 | |
|
| 0.2063 | 2.57 | 860 | 0.5505 | |
|
| 0.1724 | 2.63 | 880 | 0.5392 | |
|
| 0.2444 | 2.69 | 900 | 0.5468 | |
|
| 0.2315 | 2.75 | 920 | 0.5484 | |
|
| 0.194 | 2.81 | 940 | 0.5492 | |
|
| 0.2251 | 2.87 | 960 | 0.5483 | |
|
| 0.1779 | 2.93 | 980 | 0.5484 | |
|
| 0.3551 | 2.99 | 1000 | 0.5477 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.1 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
