meta-math/MetaMath

Issues

• Issue with Fine-tuning Mistral 7B Model - Results Discrepancy

  #29 opened 10 months ago by lyf-00
  4

• OOM error

  #34 opened a month ago by Davido111200
  0

• Transformers and Tokenizers version conflict.

  #31 opened 8 months ago by anhledger12
  1

• The accuracy of MATH test set trained on 7B is not consistent with paper.

  #27 opened a year ago by caihaunqai
  1

• Are the MetaMath Eval Result valid? e.g., What is the standard way to evaluate on MATH?

  #33 opened 8 months ago by brando90
  0

• MetaMath-Mistral-7B gsm8k/math acc different from the reported values

  #20 opened 9 months ago by haoxiongliu
  2

• eval_math and eval_gsm8k

  #30 opened 9 months ago by zhentingqi
  0

• Metamath dataset for separated for different math subjects

  #28 opened 10 months ago by sobhan93hmt
  0

• Error in run_create_backward_questions.sh

  #25 opened a year ago by ustccyf
  2

• Questions about MetaMATH dataset

  #26 opened a year ago by caihaunqai
  0

• path/to/llama-2

  #24 opened a year ago by poojitharamachandra
  2

• Potential error in eval_gsm8k.py

  #23 opened a year ago by hbin0701
  0

• May I know how you load QLoRA+LLaMA 70B with vllm?

  #15 opened a year ago by SuperBruceJia
  1

• Dataset generation script

  #1 opened a year ago by imoneoi
  5

• the few-shot in-context learning issues.

  #21 opened a year ago by runzeer
  1

• Ask for adding a new baseline——MuggleMATH in the Comparing MetaMath with the LLM models

  #9 opened a year ago by ChengpengLi1003
  1

• Is the MetaMathQA dataset suitable for pre-training?

  #22 opened a year ago by JingyiWang3
  1

• How many tokens did MetaMath train on?

  #14 opened a year ago by brando90
  0

• Problem regarding the evaluation codes of the GSM8K - `eval_gsm8k.py`

  #11 opened a year ago by SuperBruceJia
  1

• License inconsistency

  #8 opened a year ago by muelletm
  1

• OOM

  #12 opened a year ago by hbin0701
  1

• will there be ablation studies?

  #13 opened a year ago by yucc-leon
  2

• Modified parameter name that is different from README.

  #10 opened a year ago by xukp20
  1

• eval_math script outputs 0 accuracy

  #4 opened a year ago by zhangir-azerbayev
  4

• RuntimeError: probability tensor contains either `inf`, `nan` or element < 0 when setting max_new_tokens

  #7 opened a year ago by AegeanYan
  1

• License

  #6 opened a year ago by muelletm
  1

• training_scripts

  #5 opened a year ago by choco9966
  1

• Dataset

  #2 opened a year ago by zhangir-azerbayev
  10