metadata
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
datasets:
- generator
base_model: teknium/OpenHermes-2.5-Mistral-7B
model-index:
- name: teknium/OpenHermes-2.5-Mistral-7B
results: []
teknium/OpenHermes-2.5-Mistral-7B
This model is a fine-tuned version of teknium/OpenHermes-2.5-Mistral-7B on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 0.4764
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: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 0.03
- training_steps: 220
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.6429 | 0.23 | 10 | 1.3400 |
| 1.2684 | 0.47 | 20 | 1.0591 |
| 1.0218 | 0.7 | 30 | 0.8600 |
| 0.8095 | 0.93 | 40 | 0.7113 |
| 0.6887 | 1.16 | 50 | 0.6353 |
| 0.6207 | 1.4 | 60 | 0.5958 |
| 0.5726 | 1.63 | 70 | 0.5717 |
| 0.5639 | 1.86 | 80 | 0.5548 |
| 0.5496 | 2.09 | 90 | 0.5409 |
| 0.5353 | 2.33 | 100 | 0.5297 |
| 0.4968 | 2.56 | 110 | 0.5204 |
| 0.5031 | 2.79 | 120 | 0.5118 |
| 0.4906 | 3.02 | 130 | 0.5053 |
| 0.4671 | 3.26 | 140 | 0.5015 |
| 0.4734 | 3.49 | 150 | 0.4960 |
| 0.4734 | 3.72 | 160 | 0.4890 |
| 0.4522 | 3.95 | 170 | 0.4857 |
| 0.44 | 4.19 | 180 | 0.4824 |
| 0.4376 | 4.42 | 190 | 0.4797 |
| 0.4361 | 4.65 | 200 | 0.4778 |
| 0.4428 | 4.88 | 210 | 0.4769 |
| 0.4591 | 5.12 | 220 | 0.4764 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0