update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: alpaca-lora-7b-en-pt-es-ca-eu-gl-at
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# alpaca-lora-7b-en-pt-es-ca-eu-gl-at
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [decapoda-research/llama-7b-hf](https://huggingface.co/decapoda-research/llama-7b-hf) on the None dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 1.0667
|
| 18 |
+
|
| 19 |
+
## Model description
|
| 20 |
+
|
| 21 |
+
More information needed
|
| 22 |
+
|
| 23 |
+
## Intended uses & limitations
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Training and evaluation data
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training procedure
|
| 32 |
+
|
| 33 |
+
### Training hyperparameters
|
| 34 |
+
|
| 35 |
+
The following hyperparameters were used during training:
|
| 36 |
+
- learning_rate: 0.0003
|
| 37 |
+
- train_batch_size: 26
|
| 38 |
+
- eval_batch_size: 26
|
| 39 |
+
- seed: 42
|
| 40 |
+
- distributed_type: multi-GPU
|
| 41 |
+
- gradient_accumulation_steps: 5
|
| 42 |
+
- total_train_batch_size: 130
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: cosine
|
| 45 |
+
- lr_scheduler_warmup_ratio: 0.03
|
| 46 |
+
- num_epochs: 1
|
| 47 |
+
- mixed_precision_training: Native AMP
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 53 |
+
| 1.3772 | 0.04 | 100 | 1.3860 |
|
| 54 |
+
| 1.3043 | 0.07 | 200 | 1.2904 |
|
| 55 |
+
| 1.2307 | 0.11 | 300 | 1.2409 |
|
| 56 |
+
| 1.2132 | 0.15 | 400 | 1.2086 |
|
| 57 |
+
| 1.1987 | 0.19 | 500 | 1.1854 |
|
| 58 |
+
| 1.1551 | 0.22 | 600 | 1.1660 |
|
| 59 |
+
| 1.1613 | 0.26 | 700 | 1.1516 |
|
| 60 |
+
| 1.144 | 0.3 | 800 | 1.1407 |
|
| 61 |
+
| 1.1494 | 0.34 | 900 | 1.1297 |
|
| 62 |
+
| 1.1072 | 0.37 | 1000 | 1.1196 |
|
| 63 |
+
| 1.1302 | 0.41 | 1100 | 1.1117 |
|
| 64 |
+
| 1.1074 | 0.45 | 1200 | 1.1058 |
|
| 65 |
+
| 1.0846 | 0.48 | 1300 | 1.0995 |
|
| 66 |
+
| 1.086 | 0.52 | 1400 | 1.0935 |
|
| 67 |
+
| 1.0793 | 0.56 | 1500 | 1.0889 |
|
| 68 |
+
| 1.0931 | 0.6 | 1600 | 1.0847 |
|
| 69 |
+
| 1.0905 | 0.63 | 1700 | 1.0804 |
|
| 70 |
+
| 1.0793 | 0.67 | 1800 | 1.0775 |
|
| 71 |
+
| 1.0795 | 0.71 | 1900 | 1.0748 |
|
| 72 |
+
| 1.0861 | 0.74 | 2000 | 1.0725 |
|
| 73 |
+
| 1.0881 | 0.78 | 2100 | 1.0705 |
|
| 74 |
+
| 1.0673 | 0.82 | 2200 | 1.0691 |
|
| 75 |
+
| 1.0626 | 0.86 | 2300 | 1.0681 |
|
| 76 |
+
| 1.0633 | 0.89 | 2400 | 1.0674 |
|
| 77 |
+
| 1.0601 | 0.93 | 2500 | 1.0669 |
|
| 78 |
+
| 1.0849 | 0.97 | 2600 | 1.0667 |
|
| 79 |
+
|
| 80 |
+
|
| 81 |
+
### Framework versions
|
| 82 |
+
|
| 83 |
+
- Transformers 4.28.0.dev0
|
| 84 |
+
- Pytorch 2.0.0+cu117
|
| 85 |
+
- Datasets 2.10.1
|
| 86 |
+
- Tokenizers 0.13.2
|