Downloads · 30 days
27
6% of all-time downloads
Orca-Llama-3-8B-Instruct-DPO
RDson/Orca-Llama-3-8B-Instruct-DPO
Orca-Llama-3-8B-Instruct-DPO is a text generation model from RDson. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as other.
Finetuned Llama 3 8B Instruct on Intel/orcadpopairs using a single 3090 24GB. Data formated using the ChatML template.
All-time downloads
437
Public
Parameters
8B
32.1 GB on disk
Likes
3
Public
Repo makeup
Click a slice to open those files.
.safetensors16.1 GB · 100%
At a glance
- Task
- Text Generation
- Library
- transformers
- License
- other
- Model type
- llama
- Access
- Public
- Created
- Apr 20, 2024
- Updated
- Apr 30, 2024
- SHA
- ae4029e4
Try a prompt
Datasets
Model card
From the Hugging Face model README
Orca-Llama-3-8B-Instruct-DPO
Finetuned Llama 3 8B Instruct on Intel/orca_dpo_pairs using a single 3090 24GB. Data formated using the ChatML template.
GGUF can be found here RDson/Orca-Llama-3-8B-Instruct-DPO-GGUF
ORPOConfig:
learning_rate=1e-6,
lr_scheduler_type="linear",
max_length=1024,
max_prompt_length=512,
overwrite_output_dir=True,
beta=0.1,
per_device_train_batch_size=2,
per_device_eval_batch_size=2,
gradient_accumulation_steps=4,
optim="paged_adamw_8bit",
num_train_epochs=1,
evaluation_strategy="steps",
eval_steps=0.2,
logging_steps=1,
warmup_steps=35,
report_to="wandb",
output_dir="./results/",
fp16=True,
save_steps=50
- Task
- Text Generation
- Library
- transformers
- Type
- llama
- License
- other
- Created
- Apr 20, 2024
- Updated
- Apr 30, 2024