| ### Model Sources | |
| Paper: LLaMAX: Scaling Linguistic Horizons of LLM by Enhancing Translation Capabilities Beyond 100 Languages | |
| Link: https://arxiv.org/pdf/2407 | |
| ### Model Description | |
| 🔥 LLaMAX-7B-MetaMath is fully fine-tuned on the MetaMathQA dataset based on the powerful multilingual model LLaMAX-7B. | |
| 🔥 Compared with the [MetaMath-7B](https://huggingface.co/meta-math/MetaMath-7B-V1.0), LLaMAX-7B-MetaMath performs significantly better in mathematical reasoning in low-resource languages, improving the average accuracy of low-resource languages on MGSM dataset by up to 18.8%. | |
| 🔥 LLaMAX-7B-MetaMath demonstrates good multilingual math reasoning capability in all languages, improving the average accuracy by 6.2% across all languages in MGSM dataset. | |
| ### Model Usage | |
| Prompt template: | |
| ```angular2html | |
| def Prompt_template(query): | |
| prompt = ( | |
| "Below is an instruction that describes a task. " | |
| "Write a response that appropriately completes the request.\n\n" | |
| f"### Instruction:\n{query}\n\n### Response: Let's think step by step." | |
| ) | |
| return prompt | |
| ``` | |
| Code Example: | |
| ```angular2html | |
| from transformers import AutoTokenizer, LlamaForCausalLM | |
| model = LlamaForCausalLM.from_pretrained(PATH_TO_CONVERTED_WEIGHTS) | |
| tokenizer = AutoTokenizer.from_pretrained(PATH_TO_CONVERTED_TOKENIZER) | |
| query = "Bert fills out the daily crossword puzzle in the newspaper every day. He uses a pencil to fill out the puzzles every two weeks. On average, it takes him 1050 words to use up a pencil. How many words are in each crossword puzzle on average?" | |
| prompt = Prompt_template(query) | |
| inputs = tokenizer(prompt, return_tensors="pt") | |
| generate_ids = model.generate(inputs.input_ids, max_length=30) | |
| tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0] | |
| # => "If Bert uses up a pencil to fill out the puzzles every two weeks and it takes him 1050 | |
| words to use up a pencil, then he must be filling out 1050 words of crossword puzzles every | |
| two weeks. To find out how many words are in each daily crossword puzzle, we need to divide | |
| the total number of words (1050) by the number of days in two weeks (14). So, there are | |
| 1050/14 = 75 words in each daily crossword puzzle on average. #### The answer is: 75“ | |
| ``` | |
| ### Experiments | |
| We evaluated LLaMAX-7B-MetaMath on the MGSM dataset. Compared with MetaMath-7B, LLaMAX-7B-MetaMath achieves a leading on both high-resource languages (Hrl.) and low-resource languages (Lrl.). | |
| | MGSM | Bn | Th | Sw | Ja | Zh | De | Fr | Ru | Es | En | Lrl. | Hrl. | Avg. | | |
| |-----------------------------|-------|------|----|-------|------|----|----|------|----|----|------|------|--------| | |
| | MetaMath-7B (official) | 6.8 | 7.2 |6.8| 36.4 | 38.4 | 55.2|54.4| 52.0 |57.2|68.8| 6.9 | 51.8 | 38.32 | | |
| | MetaMath-7B (Reproduced) | 6.0 | 10.0 |4.4|36.4|42.8|52.8|56.0|48.8|58.8|64.8| 6.8 | 51.5 | 38.08 | | |
| | LLaMAX-7B-MetaMath |26.8| 24.0 |26.0|35.6|42.4|56.8|55.2|53.6|56.8|65.6| 25.6 | 52.3 | 44.28 | | |
| ### Citation | |
| if our model helps your work, please cite this paper: | |
| ``` | |
| @inproceedings{Huang2024MindMergerEB, | |
| title={XLLaMA2: Scaling Linguistic Horizons of LLM by Enhancing Translation Capabilities Beyond 100 Languages}, | |
| year={2024}, | |
| } | |
| ``` | |