Feedback and support: TensorBlock's Twitter/X, Telegram Group and Discord server
Sharathhebbar24/math_gpt2_sft - GGUF
This repo contains GGUF format model files for Sharathhebbar24/math_gpt2_sft.
The files were quantized using machines provided by TensorBlock, and they are compatible with llama.cpp as of commit b4242.
Prompt template
Model file specification
Filename | Quant type | File Size | Description |
---|---|---|---|
math_gpt2_sft-Q2_K.gguf | Q2_K | 0.081 GB | smallest, significant quality loss - not recommended for most purposes |
math_gpt2_sft-Q3_K_S.gguf | Q3_K_S | 0.090 GB | very small, high quality loss |
math_gpt2_sft-Q3_K_M.gguf | Q3_K_M | 0.098 GB | very small, high quality loss |
math_gpt2_sft-Q3_K_L.gguf | Q3_K_L | 0.102 GB | small, substantial quality loss |
math_gpt2_sft-Q4_0.gguf | Q4_0 | 0.107 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
math_gpt2_sft-Q4_K_S.gguf | Q4_K_S | 0.107 GB | small, greater quality loss |
math_gpt2_sft-Q4_K_M.gguf | Q4_K_M | 0.113 GB | medium, balanced quality - recommended |
math_gpt2_sft-Q5_0.gguf | Q5_0 | 0.122 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
math_gpt2_sft-Q5_K_S.gguf | Q5_K_S | 0.122 GB | large, low quality loss - recommended |
math_gpt2_sft-Q5_K_M.gguf | Q5_K_M | 0.127 GB | large, very low quality loss - recommended |
math_gpt2_sft-Q6_K.gguf | Q6_K | 0.138 GB | very large, extremely low quality loss |
math_gpt2_sft-Q8_0.gguf | Q8_0 | 0.178 GB | very large, extremely low quality loss - not recommended |
Downloading instruction
Command line
Firstly, install Huggingface Client
pip install -U "huggingface_hub[cli]"
Then, downoad the individual model file the a local directory
huggingface-cli download tensorblock/math_gpt2_sft-GGUF --include "math_gpt2_sft-Q2_K.gguf" --local-dir MY_LOCAL_DIR
If you wanna download multiple model files with a pattern (e.g., *Q4_K*gguf
), you can try:
huggingface-cli download tensorblock/math_gpt2_sft-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
- Downloads last month
- 134
Model tree for tensorblock/math_gpt2_sft-GGUF
Base model
Sharathhebbar24/math_gpt2_sftDataset used to train tensorblock/math_gpt2_sft-GGUF
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard22.870
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard30.410
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard25.060
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard37.620
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard51.540
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard0.680