Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: HuggingFaceH4/zephyr-7b-beta
|
| 9 |
+
model-index:
|
| 10 |
+
- name: zephyr_instruct_generation
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# zephyr_instruct_generation
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [HuggingFaceH4/zephyr-7b-beta](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.1686
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 8e-05
|
| 41 |
+
- train_batch_size: 4
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
+
- seed: 42
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: constant
|
| 46 |
+
- lr_scheduler_warmup_steps: 0.03
|
| 47 |
+
- training_steps: 200
|
| 48 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 53 |
+
| 3.048 | 0.77 | 10 | 2.4666 |
|
| 54 |
+
| 2.1245 | 1.54 | 20 | 1.6932 |
|
| 55 |
+
| 1.3178 | 2.31 | 30 | 1.1619 |
|
| 56 |
+
| 1.0341 | 3.08 | 40 | 0.9138 |
|
| 57 |
+
| 0.8234 | 3.85 | 50 | 0.8017 |
|
| 58 |
+
| 0.6618 | 4.62 | 60 | 0.7330 |
|
| 59 |
+
| 0.6123 | 5.38 | 70 | 0.7278 |
|
| 60 |
+
| 0.539 | 6.15 | 80 | 0.6836 |
|
| 61 |
+
| 0.4648 | 6.92 | 90 | 0.7171 |
|
| 62 |
+
| 0.4244 | 7.69 | 100 | 0.7712 |
|
| 63 |
+
| 0.3771 | 8.46 | 110 | 0.8586 |
|
| 64 |
+
| 0.351 | 9.23 | 120 | 0.8261 |
|
| 65 |
+
| 0.3233 | 10.0 | 130 | 0.8205 |
|
| 66 |
+
| 0.2946 | 10.77 | 140 | 0.8420 |
|
| 67 |
+
| 0.289 | 11.54 | 150 | 1.0008 |
|
| 68 |
+
| 0.2604 | 12.31 | 160 | 0.9676 |
|
| 69 |
+
| 0.2615 | 13.08 | 170 | 1.0381 |
|
| 70 |
+
| 0.2557 | 13.85 | 180 | 1.0656 |
|
| 71 |
+
| 0.2514 | 14.62 | 190 | 1.0422 |
|
| 72 |
+
| 0.2357 | 15.38 | 200 | 1.1686 |
|
| 73 |
+
|
| 74 |
+
|
| 75 |
+
### Framework versions
|
| 76 |
+
|
| 77 |
+
- PEFT 0.7.1
|
| 78 |
+
- Transformers 4.36.2
|
| 79 |
+
- Pytorch 2.1.2+cu121
|
| 80 |
+
- Datasets 2.16.0
|
| 81 |
+
- Tokenizers 0.15.0
|