license: llama2
model-index:
- name: LLaMA-Pro-8B-Instruct
results:
- task:
type: text-generation
name: Text Generation
dataset:
name: AI2 Reasoning Challenge (25-Shot)
type: ai2_arc
config: ARC-Challenge
split: test
args:
num_few_shot: 25
metrics:
- type: acc_norm
value: 52.99
name: normalized accuracy
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TencentARC/LLaMA-Pro-8B-Instruct
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: HellaSwag (10-Shot)
type: hellaswag
split: validation
args:
num_few_shot: 10
metrics:
- type: acc_norm
value: 76.98
name: normalized accuracy
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TencentARC/LLaMA-Pro-8B-Instruct
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MMLU (5-Shot)
type: cais/mmlu
config: all
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 52.58
name: accuracy
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TencentARC/LLaMA-Pro-8B-Instruct
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: TruthfulQA (0-shot)
type: truthful_qa
config: multiple_choice
split: validation
args:
num_few_shot: 0
metrics:
- type: mc2
value: 49.43
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TencentARC/LLaMA-Pro-8B-Instruct
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: Winogrande (5-shot)
type: winogrande
config: winogrande_xl
split: validation
args:
num_few_shot: 5
metrics:
- type: acc
value: 72.22
name: accuracy
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TencentARC/LLaMA-Pro-8B-Instruct
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: GSM8k (5-shot)
type: gsm8k
config: main
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 44.2
name: accuracy
source:
url: >-
https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=TencentARC/LLaMA-Pro-8B-Instruct
name: Open LLM Leaderboard
LLaMA-PRO-Instruct Model Card
Model Description
LLaMA-PRO-Instruct is a transformative expansion of the LLaMA2-7B model, now boasting 8.3 billion parameters. It uniquely specializes in programming, coding, and mathematical reasoning, maintaining versatility in general language tasks.
Development and Training
This model, developed by Tencent ARC team, extends LLaMA2-7B using innovative block expansion techniques. It's meticulously trained on a diverse blend of coding and mathematical data, encompassing over 80 billion tokens.
Intended Use
LLaMA-PRO-Instruct is ideal for complex NLP challenges, excelling in programming, mathematical reasoning, and general language processing, suitable for both specialized and broad applications.
Performance
It surpasses its predecessors in the LLaMA series, especially in code domains, demonstrating exceptional competence as a comprehensive language model.
Limitations
Despite advancements, it may encounter difficulties in highly niche or nuanced tasks.
Ethical Considerations
Users are advised to consider inherent biases and responsibly manage its application across various fields.
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 58.06 |
| AI2 Reasoning Challenge (25-Shot) | 52.99 |
| HellaSwag (10-Shot) | 76.98 |
| MMLU (5-Shot) | 52.58 |
| TruthfulQA (0-shot) | 49.43 |
| Winogrande (5-shot) | 72.22 |
| GSM8k (5-shot) | 44.20 |