File size: 1,978 Bytes
174ebd7 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
---
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
base_model: TheBloke/Mistral-7B-Instruct-v0.2-GPTQ
model-index:
- name: labmate-mistral-7b-ft
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. -->
# labmate-mistral-7b-ft
This model is a fine-tuned version of [TheBloke/Mistral-7B-Instruct-v0.2-GPTQ](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GPTQ) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.4153
## 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: 0.0002
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2
- num_epochs: 10
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 0.7983 | 1.0 | 1 | 1.4848 |
| 0.7983 | 2.0 | 2 | 1.4791 |
| 0.7863 | 3.0 | 3 | 1.4653 |
| 0.7603 | 4.0 | 4 | 1.4518 |
| 0.7353 | 5.0 | 5 | 1.4395 |
| 0.7135 | 6.0 | 6 | 1.4299 |
| 0.6955 | 7.0 | 7 | 1.4232 |
| 0.6815 | 8.0 | 8 | 1.4188 |
| 0.6712 | 9.0 | 9 | 1.4165 |
| 0.6642 | 10.0 | 10 | 1.4153 |
### Framework versions
- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.1.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1 |