---
license: apache-2.0
datasets:
- mlabonne/orpo-dpo-mix-40k
language:
- en
library_name: transformers
base_model: h2oai/h2o-danube2-1.8b-base
---
# danube2-1.8b-ORPO

ChatML tokens are added and first fine-tuned with BAdam and then QLoRA+ on mlabonne/orpo-dpo-mix-40k, but as SFT and not DPO, and using LLama-Factory.

## Template

```jinja
<|im_start>user
{{instruction}}<|im_end|>
<|im_start>assistant
{{response}}<|im_end>
```

## BAdam

```yaml
### model
model_name_or_path: danube2-base-chatml

### method
stage: sft
do_train: true
finetuning_type: full
use_badam: true
badam_switch_mode: ascending
badam_switch_interval: 50
badam_verbose: 1
badam_start_block: 12
badam_mask_mode: scatter
seed: 314

### dataset
dataset: orpo_sft_mix_40k
template: ninja_chatml
cutoff_len: 8192
overwrite_cache: false
preprocessing_num_workers: 12

### output
output_dir: orpo-chatml-badam
logging_steps: 5
save_steps: 1
save_strategy: epoch
plot_loss: true
overwrite_output_dir: false

### train
per_device_train_batch_size: 2
gradient_accumulation_steps: 8
learning_rate: 0.00001
num_train_epochs: 2
lr_scheduler_type: cosine
warmup_ratio: 0.01
pure_bf16: true
flash_attn: fa2

### eval
val_size: 0.01
per_device_eval_batch_size: 1
eval_strategy: steps
eval_steps: 1000
```


### QLoRA+

```yaml
### model
model_name_or_path: orpo-chatml-badam

### method
stage: sft
do_train: true
finetuning_type: lora
lora_target: all
loraplus_lr_ratio: 16.0
lora_rank: 8
lora_alpha: 16
use_unsloth: true
quantization_bit: 4
upcast_layernorm: true
seed: 31415

### dataset
dataset: orpo_sft_mix_40k
template: hermes_chatml
cutoff_len: 8192
overwrite_cache: false
preprocessing_num_workers: 12

### output
output_dir: orpo-chatml-badam/loraplus
logging_steps: 1
save_steps: 1
save_strategy: epoch
plot_loss: true
overwrite_output_dir: false

### train
per_device_train_batch_size: 4
gradient_accumulation_steps: 4
learning_rate: 0.0001
num_train_epochs: 2.0
lr_scheduler_type: cosine
warmup_ratio: 0.01
bf16: true
flash_attn: fa2

### eval
val_size: 0.02
per_device_eval_batch_size: 1
eval_strategy: steps
eval_steps: 1000
```