Downloads · 30 days
158
2% of all-time downloads
RichardErkhov/Sao10K_-_L3-8B-Lunaris-v1-gguf
Sao10K_-_L3-8B-Lunaris-v1-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
158
2% of all-time downloads
All-time downloads
6.5K
Public
Repo size
92.9 GB
Likes
1
Public
Click a slice to open those files.
.gguf108 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
L3-8B-Lunaris-v1 - GGUF
| Name | Quant method | Size |
|---|---|---|
| L3-8B-Lunaris-v1.Q2_K.gguf | Q2_K | 2.96GB |
| L3-8B-Lunaris-v1.IQ3_XS.gguf | IQ3_XS | 3.28GB |
| L3-8B-Lunaris-v1.IQ3_S.gguf | IQ3_S | 3.43GB |
| L3-8B-Lunaris-v1.Q3_K_S.gguf | Q3_K_S | 3.41GB |
| L3-8B-Lunaris-v1.IQ3_M.gguf | IQ3_M | 3.52GB |
| L3-8B-Lunaris-v1.Q3_K.gguf | Q3_K | 3.74GB |
| L3-8B-Lunaris-v1.Q3_K_M.gguf | Q3_K_M | 3.74GB |
| L3-8B-Lunaris-v1.Q3_K_L.gguf | Q3_K_L | 4.03GB |
| L3-8B-Lunaris-v1.IQ4_XS.gguf | IQ4_XS | 4.18GB |
| L3-8B-Lunaris-v1.Q4_0.gguf | Q4_0 | 4.34GB |
| L3-8B-Lunaris-v1.IQ4_NL.gguf | IQ4_NL | 4.38GB |
| L3-8B-Lunaris-v1.Q4_K_S.gguf | Q4_K_S | 4.37GB |
| L3-8B-Lunaris-v1.Q4_K.gguf | Q4_K | 4.58GB |
| L3-8B-Lunaris-v1.Q4_K_M.gguf | Q4_K_M | 4.58GB |
| L3-8B-Lunaris-v1.Q4_1.gguf | Q4_1 | 4.78GB |
| L3-8B-Lunaris-v1.Q5_0.gguf | Q5_0 | 5.21GB |
| L3-8B-Lunaris-v1.Q5_K_S.gguf | Q5_K_S | 5.21GB |
| L3-8B-Lunaris-v1.Q5_K.gguf | Q5_K | 5.34GB |
| L3-8B-Lunaris-v1.Q5_K_M.gguf | Q5_K_M | 5.34GB |
| L3-8B-Lunaris-v1.Q5_1.gguf | Q5_1 | 5.65GB |
| L3-8B-Lunaris-v1.Q6_K.gguf | Q6_K | 6.14GB |
| L3-8B-Lunaris-v1.Q8_0.gguf | Q8_0 | 7.95GB |
license: llama3 language:
A generalist / roleplaying model merge based on Llama 3. Models are selected from my personal experience while using them.
I personally think this is an improvement over Stheno v3.2, considering the other models helped balance out its creativity and at the same time improving its logic.
Settings:
Instruct // Context Template: Llama-3-Instruct
Temperature: 1.4
min_p: 0.1
Merging seems to be a black box magic though? In my personal experience merging multiple models from different datasets / data works better than combining them all in one.
Values chosen are from long-running personal experimentation since Llama-2 Merging Era. I have tweaked them to fit this recipe.
Mergekit Config
models:
- model: meta-llama/Meta-Llama-3-8B-Instruct
- model: crestf411/L3-8B-sunfall-v0.1 # Another RP Model trained on... stuff
parameters:
density: 0.4
weight: 0.25
- model: Hastagaras/Jamet-8B-L3-MK1 - # Another RP / Storytelling Model
parameters:
density: 0.5
weight: 0.3
- model: maldv/badger-iota-llama-3-8b #Megamerge - Helps with General Knowledge
parameters:
density: 0.6
weight: 0.35
- model: Sao10K/Stheno-3.2-Beta # This is Stheno v3.2's Initial Name
parameters:
density: 0.7
weight: 0.4
merge_method: ties
base_model: meta-llama/Meta-Llama-3-8B-Instruct
parameters:
int8_mask: true
rescale: true
normalize: false
dtype: bfloat16