metadata
license: apache-2.0
library_name: peft
tags:
- alignment-handbook
- generated_from_trainer
base_model: h2oai/h2o-danube2-1.8b-base
datasets:
- Ritvik19/open-hermes-2_5-reformatted
model-index:
- name: openhermes-danube-sft-qlora
results: []
Note: This model card has been generated automatically according to the information the Trainer had access to. Visit the model card to see the full description
openhermes-danube-sft-qlora
This model is a fine-tuned version of h2oai/h2o-danube2-1.8b-base on the Ritvik19/open-hermes-2_5-reformatted dataset. It achieves the following results on the evaluation set:
- Loss: 1.1197
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 128
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.0838 | 0.9999 | 1704 | 1.1197 |
Framework versions
- PEFT 0.7.1
- Transformers 4.40.1
- Pytorch 2.1.2+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
Metric | Value |
---|---|
Avg. | 44.12 |
AI2 Reasoning Challenge (25-Shot) | 43.26 |
HellaSwag (10-Shot) | 73.12 |
MMLU (5-Shot) | 40.19 |
TruthfulQA (0-shot) | 38.93 |
Winogrande (5-shot) | 67.88 |
GSM8k (5-shot) | 1.36 |