File size: 1,837 Bytes
e64b907 | 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 | ---
base_model: /cpfs01/user/zhangyiqun/models/gemma-2-2b-it
library_name: peft
license: other
tags:
- llama-factory
- lora
- generated_from_trainer
model-index:
- name: cot
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. -->
# cot
This model is a fine-tuned version of [/cpfs01/user/zhangyiqun/models/gemma-2-2b-it](https://huggingface.co//cpfs01/user/zhangyiqun/models/gemma-2-2b-it) on the cot dataset.
It achieves the following results on the evaluation set:
- Loss: 1.1492
## 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: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 32
- total_eval_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5.0
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.7901 | 0.6770 | 1000 | 0.8920 |
| 0.7599 | 1.3541 | 2000 | 0.8573 |
| 0.5325 | 2.0311 | 3000 | 0.8718 |
| 0.5875 | 2.7082 | 4000 | 0.8556 |
| 0.372 | 3.3852 | 5000 | 0.9813 |
| 0.2649 | 4.0623 | 6000 | 1.1342 |
| 0.2722 | 4.7393 | 7000 | 1.1488 |
### Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.3.0+cu121
- Datasets 3.0.1
- Tokenizers 0.19.1 |