metadata
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
base_model: ericzzz/falcon-rw-1b-instruct-openorca
model-index:
- name: falcon-rw-1b-instruct-openorca-retrained_network_corpus
results: []
falcon-rw-1b-instruct-openorca-retrained_network_corpus
This model is a fine-tuned version of ericzzz/falcon-rw-1b-instruct-openorca on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.5152
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- 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 |
|---|---|---|---|
| 3.7165 | 0.5714 | 1 | 2.1857 |
| 1.7689 | 1.7143 | 3 | 1.9718 |
| 1.588 | 2.8571 | 5 | 1.7478 |
| 1.4636 | 4.0 | 7 | 1.6172 |
| 2.7378 | 4.5714 | 8 | 1.5710 |
| 0.9276 | 5.7143 | 10 | 1.5152 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.1.2
- Datasets 2.19.2
- Tokenizers 0.19.1