metadata
license: gemma
library_name: peft
tags:
- alignment-handbook
- trl
- sft
- generated_from_trainer
base_model: google/gemma-7b
datasets:
- chansung/merged_ds_coding
model-index:
- name: coding_llamaduo_result1
results: []
coding_llamaduo_result1
This model is a fine-tuned version of google/gemma-7b on the chansung/merged_ds_coding dataset. It achieves the following results on the evaluation set:
- Loss: 1.1871
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: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- total_eval_batch_size: 4
- 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
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.5404 | 0.99 | 36 | 1.5048 |
| 0.9147 | 2.0 | 73 | 1.2327 |
| 0.7658 | 2.99 | 109 | 1.1766 |
| 0.6657 | 4.0 | 146 | 1.1664 |
| 0.5601 | 4.93 | 180 | 1.1871 |
Framework versions
- PEFT 0.7.1
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2