PlatYi-34B-LoRA

Model Details

Model Developers Kyujin Han (kyujinpy)

Input Models input text only.

Output Models generate text only.

Model Architecture
PlatYi-34B-LoRA is an auto-regressive language model based on the Yi-34B transformer architecture.

Blog Link
Blog: [Coming soon...]
Github: [Coming soon...]

Base Model
01-ai/Yi-34B

Training Dataset
garage-bAInd/Open-Platypus.

Notice
While training, I used LoRA.
The lora_r values is 16.

Model Benchmark

Open leaderboard

• Follow up as link.

Model	Average	ARC	HellaSwag	MMLU	TruthfulQA	Winogrande	GSM8K
PlatYi-34B-Q	69.86	66.89	85.14	77.66	53.03	82.48	53.98
PlatYi-34B-LoRA	68.1	67.15	85.37	78.46	53.32	83.66	40.64
01-ai/Yi-34B	69.42	64.59	85.69	76.35	56.23	83.03	50.64

Implementation Code

### KO-Platypus
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

repo = "kyujinpy/PlatYi-34B-LoRA"
OpenOrca = AutoModelForCausalLM.from_pretrained(
        repo,
        return_dict=True,
        torch_dtype=torch.float16,
        device_map='auto'
)
OpenOrca_tokenizer = AutoTokenizer.from_pretrained(repo)

---

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	68.10
AI2 Reasoning Challenge (25-Shot)	67.15
HellaSwag (10-Shot)	85.37
MMLU (5-Shot)	78.46
TruthfulQA (0-shot)	53.32
Winogrande (5-shot)	83.66
GSM8k (5-shot)	40.64