Downloads · 30 days
370
25% of all-time downloads
RichardErkhov/mergekit-community_-_L3.1-Pneuma-8B-v1-gguf
mergekit-community_-_L3.1-Pneuma-8B-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
370
25% of all-time downloads
All-time downloads
1.5K
Public
Repo size
92.9 GB
Likes
0
Public
Click a slice to open those files.
.gguf108 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
L3.1-Pneuma-8B-v1 - GGUF
| Name | Quant method | Size |
|---|---|---|
| L3.1-Pneuma-8B-v1.Q2_K.gguf | Q2_K | 2.96GB |
| L3.1-Pneuma-8B-v1.IQ3_XS.gguf | IQ3_XS | 3.28GB |
| L3.1-Pneuma-8B-v1.IQ3_S.gguf | IQ3_S | 3.43GB |
| L3.1-Pneuma-8B-v1.Q3_K_S.gguf | Q3_K_S | 3.41GB |
| L3.1-Pneuma-8B-v1.IQ3_M.gguf | IQ3_M | 3.52GB |
| L3.1-Pneuma-8B-v1.Q3_K.gguf | Q3_K | 3.74GB |
| L3.1-Pneuma-8B-v1.Q3_K_M.gguf | Q3_K_M | 3.74GB |
| L3.1-Pneuma-8B-v1.Q3_K_L.gguf | Q3_K_L | 4.03GB |
| L3.1-Pneuma-8B-v1.IQ4_XS.gguf | IQ4_XS | 4.18GB |
| L3.1-Pneuma-8B-v1.Q4_0.gguf | Q4_0 | 4.34GB |
| L3.1-Pneuma-8B-v1.IQ4_NL.gguf | IQ4_NL | 4.38GB |
| L3.1-Pneuma-8B-v1.Q4_K_S.gguf | Q4_K_S | 4.37GB |
| L3.1-Pneuma-8B-v1.Q4_K.gguf | Q4_K | 4.58GB |
| L3.1-Pneuma-8B-v1.Q4_K_M.gguf | Q4_K_M | 4.58GB |
| L3.1-Pneuma-8B-v1.Q4_1.gguf | Q4_1 | 4.78GB |
| L3.1-Pneuma-8B-v1.Q5_0.gguf | Q5_0 | 5.21GB |
| L3.1-Pneuma-8B-v1.Q5_K_S.gguf | Q5_K_S | 5.21GB |
| L3.1-Pneuma-8B-v1.Q5_K.gguf | Q5_K | 5.34GB |
| L3.1-Pneuma-8B-v1.Q5_K_M.gguf | Q5_K_M | 5.34GB |
| L3.1-Pneuma-8B-v1.Q5_1.gguf | Q5_1 | 5.65GB |
| L3.1-Pneuma-8B-v1.Q6_K.gguf | Q6_K | 6.14GB |
| L3.1-Pneuma-8B-v1.Q8_0.gguf | Q8_0 | 7.95GB |
base_model:
This is a merge of pre-trained language models created using mergekit.
This model was merged using the della_linear merge method using ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.2 as a base.
The following models were included in the merge:
The following YAML configuration was used to produce this model:
out_dtype: bfloat16
dtype: float32
tokenizer_source: base
merge_method: della_linear
parameters:
int8_mask: true
density: 0.5
epsilon: 0.04
lambda: 1.05
base_model: ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.2
models:
- model: ArliAI/Llama-3.1-8B-ArliAI-RPMax-v1.2
parameters:
weight:
- filter: v_proj
value: [1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1]
- filter: o_proj
value: [1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1]
- filter: up_proj
value: [1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1]
- filter: gate_proj
value: [1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1]
- filter: down_proj
value: [1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1]
- value: 1
- model: Replete-AI/L3-Pneuma-8B
parameters:
weight:
- filter: v_proj
value: [0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0]
- filter: o_proj
value: [0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0]
- filter: up_proj
value: [0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0]
- filter: gate_proj
value: [0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0]
- filter: down_proj
value: [0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0]
- value: 0