Downloads · 30 days
54
1% of all-time downloads
RichardErkhov/Sao10K_-_Medusa-1.1-L2-7B-gguf
Sao10K_-_Medusa-1.1-L2-7B-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
54
1% of all-time downloads
All-time downloads
6.2K
Public
Repo size
76.6 GB
Likes
0
Public
Click a slice to open those files.
.gguf88.7 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
Medusa-1.1-L2-7B - GGUF
| Name | Quant method | Size |
|---|---|---|
| Medusa-1.1-L2-7B.Q2_K.gguf | Q2_K | 2.36GB |
| Medusa-1.1-L2-7B.IQ3_XS.gguf | IQ3_XS | 2.6GB |
| Medusa-1.1-L2-7B.IQ3_S.gguf | IQ3_S | 2.75GB |
| Medusa-1.1-L2-7B.Q3_K_S.gguf | Q3_K_S | 2.75GB |
| Medusa-1.1-L2-7B.IQ3_M.gguf | IQ3_M | 2.9GB |
| Medusa-1.1-L2-7B.Q3_K.gguf | Q3_K | 3.07GB |
| Medusa-1.1-L2-7B.Q3_K_M.gguf | Q3_K_M | 3.07GB |
| Medusa-1.1-L2-7B.Q3_K_L.gguf | Q3_K_L | 3.35GB |
| Medusa-1.1-L2-7B.IQ4_XS.gguf | IQ4_XS | 3.4GB |
| Medusa-1.1-L2-7B.Q4_0.gguf | Q4_0 | 3.56GB |
| Medusa-1.1-L2-7B.IQ4_NL.gguf | IQ4_NL | 3.58GB |
| Medusa-1.1-L2-7B.Q4_K_S.gguf | Q4_K_S | 3.59GB |
| Medusa-1.1-L2-7B.Q4_K.gguf | Q4_K | 3.8GB |
| Medusa-1.1-L2-7B.Q4_K_M.gguf | Q4_K_M | 3.8GB |
| Medusa-1.1-L2-7B.Q4_1.gguf | Q4_1 | 3.95GB |
| Medusa-1.1-L2-7B.Q5_0.gguf | Q5_0 | 4.33GB |
| Medusa-1.1-L2-7B.Q5_K_S.gguf | Q5_K_S | 4.33GB |
| Medusa-1.1-L2-7B.Q5_K.gguf | Q5_K | 4.45GB |
| Medusa-1.1-L2-7B.Q5_K_M.gguf | Q5_K_M | 4.45GB |
| Medusa-1.1-L2-7B.Q5_1.gguf | Q5_1 | 4.72GB |
| Medusa-1.1-L2-7B.Q6_K.gguf | Q6_K | 5.15GB |
| Medusa-1.1-L2-7B.Q8_0.gguf | Q8_0 | 6.67GB |
license: llama2 language:
Experimental Ties-Merge between 5 Models and 2 LORAs at varying weights and densities. <br> And trained with some dataset.
This is purely for my personal testing. Use if you want.
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 49.62 |
| ARC (25-shot) | 56.48 |
| HellaSwag (10-shot) | 78.57 |
| MMLU (5-shot) | 51.56 |
| TruthfulQA (0-shot) | 47.7 |
| Winogrande (5-shot) | 75.06 |
| GSM8K (5-shot) | 1.44 |
| DROP (3-shot) | 36.53 |