rouge / nanbeige2-16b-chat

The Nanbeige2-16B-Chat is the latest 16B model developed by the Nanbeige Lab, which utilized 4.5T tokens of high-quality training data during the training phase.

47 Pulls Updated 3 months ago

3 months ago

7c913b7bcd68 · 9.6GB

model
llama
·
15.8B
·
Q4_K_M
template
{{ if .System }}<|start_header_id|>system<|end_header_id|> {{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|> {{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|> {{ .Response }}<|eot_id|>
params
{"stop":["<|start_header_id|>","<|end_header_id|>","<|eot_id|>","<|reserved_special_token"],"temperature":0.6}
system
You are a friendly, knowledgeable, and capable AI assistant. Your purpose is to help the user with a wide variety of tasks to the best of your abilities.

Readme

Introduction

The Nanbeige2-16B-Chat is the latest 16B model developed by the Nanbeige Lab, which utilized 4.5T tokens of high-quality training data during the training phase.
During the alignment phase, we initially trained our model using 1 million samples through Supervised Fine-Tuning (SFT). We then engaged in curriculum learning with 400,000 high-quality samples that presented a greater level of difficulty. Subsequently, we incorporated human feedback through the Direct Preference Optimization (DPO), culminating in the development of Nanbeige2-16B-Chat. Nanbeige2-16B-Chat has achieved superior performance across various authoritative benchmark datasets.

Evaluation

We have evaluated Nanbeige2-8B-Chat’s general question-answering capabilities and human preference alignments on several popular benchmark datasets. The model has achieved notable results in single-turn English QA (AlpacaEval 2.0), single-turn Chinese QA (AlignBench), and multi-turn English QA (MT-Bench).

AlpacaEval 2.0 AlignBench MT-Bench
43.0%/40.4% 7.62 8.60