metadata
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
datasets:
- generator
base_model: princeton-nlp/Sheared-LLaMA-1.3B
model-index:
- name: fluent-capybara-27-replication
results: []
fluent-capybara-27-replication
This model is a fine-tuned version of princeton-nlp/Sheared-LLaMA-1.3B on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 2.2483
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: 2.5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 100
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 200
- num_epochs: 5.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.7039 | 1.0 | 750 | 2.6816 |
| 2.4411 | 2.0 | 1500 | 2.4269 |
| 2.3104 | 3.0 | 2250 | 2.3173 |
| 2.2149 | 4.0 | 3001 | 2.2593 |
| 2.0649 | 5.0 | 3750 | 2.2483 |
Framework versions
- PEFT 0.10.0
- Transformers 4.39.1
- Pytorch 2.1.2+cu118
- Datasets 2.18.0
- Tokenizers 0.15.1