Mixtral_7Bx4_MOE_24B/README.md

license, model-index

license

model-index

cc-by-nc-4.0

name results mixtral_7bx4_moe task dataset metrics source type name text-generation Text Generation name type config split args AI2 Reasoning Challenge (25-Shot) ai2_arc ARC-Challenge test num_few_shot 25 type value name acc_norm 65.27 normalized accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cloudyu/mixtral_7bx4_moe Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type split args HellaSwag (10-Shot) hellaswag validation num_few_shot 10 type value name acc_norm 85.28 normalized accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cloudyu/mixtral_7bx4_moe Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args MMLU (5-Shot) cais/mmlu all test num_few_shot 5 type value name acc 62.84 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cloudyu/mixtral_7bx4_moe Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args TruthfulQA (0-shot) truthful_qa multiple_choice validation num_few_shot 0 type value mc2 59.85 url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cloudyu/mixtral_7bx4_moe Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args Winogrande (5-shot) winogrande winogrande_xl validation num_few_shot 5 type value name acc 77.66 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cloudyu/mixtral_7bx4_moe Open LLM Leaderboard task dataset metrics source type name text-generation Text Generation name type config split args GSM8k (5-shot) gsm8k main test num_few_shot 5 type value name acc 62.09 accuracy url name https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=cloudyu/mixtral_7bx4_moe Open LLM Leaderboard

I don't know why so many downloads about this model. Please share your cases, thanks.

Now this model is improved by DPO to cloudyu/Pluto_24B_DPO_200

• Metrics improved by DPO

Mixtral MOE 4x7B

MOE the following models by mergekit:

• Q-bert/MetaMath-Cybertron-Starling
• mistralai/Mistral-7B-Instruct-v0.2
• teknium/Mistral-Trismegistus-7B
• meta-math/MetaMath-Mistral-7B
• openchat/openchat-3.5-1210

Metrics

• Average : 68.85
• ARC:65.36
• HellaSwag:85.23
• more details: https://huggingface.co/datasets/open-llm-leaderboard/results/blob/main/cloudyu/Mixtral_7Bx4_MOE_24B/results_2023-12-23T18-05-51.243288.json

gpu code example

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
import math

## v2 models
model_path = "cloudyu/Mixtral_7Bx4_MOE_24B"

tokenizer = AutoTokenizer.from_pretrained(model_path, use_default_system_prompt=False)
model = AutoModelForCausalLM.from_pretrained(
    model_path, torch_dtype=torch.float32, device_map='auto',local_files_only=False, load_in_4bit=True
)
print(model)
prompt = input("please input prompt:")
while len(prompt) > 0:
  input_ids = tokenizer(prompt, return_tensors="pt").input_ids.to("cuda")

  generation_output = model.generate(
    input_ids=input_ids, max_new_tokens=500,repetition_penalty=1.2
  )
  print(tokenizer.decode(generation_output[0]))
  prompt = input("please input prompt:")

CPU example

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
import math

## v2 models
model_path = "cloudyu/Mixtral_7Bx4_MOE_24B"

tokenizer = AutoTokenizer.from_pretrained(model_path, use_default_system_prompt=False)
model = AutoModelForCausalLM.from_pretrained(
    model_path, torch_dtype=torch.float32, device_map='cpu',local_files_only=False
)
print(model)
prompt = input("please input prompt:")
while len(prompt) > 0:
  input_ids = tokenizer(prompt, return_tensors="pt").input_ids

  generation_output = model.generate(
    input_ids=input_ids, max_new_tokens=500,repetition_penalty=1.2
  )
  print(tokenizer.decode(generation_output[0]))
  prompt = input("please input prompt:")

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	68.83
AI2 Reasoning Challenge (25-Shot)	65.27
HellaSwag (10-Shot)	85.28
MMLU (5-Shot)	62.84
TruthfulQA (0-shot)	59.85
Winogrande (5-shot)	77.66
GSM8k (5-shot)	62.09