Feedback and support: TensorBlock's Twitter/X, Telegram Group and Discord server
Felladrin/Smol-Llama-101M-Chat-v1 - GGUF
This repo contains GGUF format model files for Felladrin/Smol-Llama-101M-Chat-v1.
The files were quantized using machines provided by TensorBlock, and they are compatible with llama.cpp as of commit b4242.
Prompt template
<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
Model file specification
Filename | Quant type | File Size | Description |
---|---|---|---|
Smol-Llama-101M-Chat-v1-Q2_K.gguf | Q2_K | 0.048 GB | smallest, significant quality loss - not recommended for most purposes |
Smol-Llama-101M-Chat-v1-Q3_K_S.gguf | Q3_K_S | 0.054 GB | very small, high quality loss |
Smol-Llama-101M-Chat-v1-Q3_K_M.gguf | Q3_K_M | 0.056 GB | very small, high quality loss |
Smol-Llama-101M-Chat-v1-Q3_K_L.gguf | Q3_K_L | 0.059 GB | small, substantial quality loss |
Smol-Llama-101M-Chat-v1-Q4_0.gguf | Q4_0 | 0.064 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
Smol-Llama-101M-Chat-v1-Q4_K_S.gguf | Q4_K_S | 0.064 GB | small, greater quality loss |
Smol-Llama-101M-Chat-v1-Q4_K_M.gguf | Q4_K_M | 0.065 GB | medium, balanced quality - recommended |
Smol-Llama-101M-Chat-v1-Q5_0.gguf | Q5_0 | 0.074 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
Smol-Llama-101M-Chat-v1-Q5_K_S.gguf | Q5_K_S | 0.074 GB | large, low quality loss - recommended |
Smol-Llama-101M-Chat-v1-Q5_K_M.gguf | Q5_K_M | 0.074 GB | large, very low quality loss - recommended |
Smol-Llama-101M-Chat-v1-Q6_K.gguf | Q6_K | 0.084 GB | very large, extremely low quality loss |
Smol-Llama-101M-Chat-v1-Q8_0.gguf | Q8_0 | 0.108 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/Smol-Llama-101M-Chat-v1-GGUF --include "Smol-Llama-101M-Chat-v1-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/Smol-Llama-101M-Chat-v1-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
- Downloads last month
- 164
Model tree for tensorblock/Smol-Llama-101M-Chat-v1-GGUF
Base model
BEE-spoke-data/smol_llama-101M-GQA
Finetuned
Felladrin/Smol-Llama-101M-Chat-v1
Datasets used to train tensorblock/Smol-Llama-101M-Chat-v1-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 Leaderboard28.690
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard24.930
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard45.760
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard50.040
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard0.080