Update README.md
Browse files
README.md
CHANGED
|
@@ -1,11 +1,19 @@
|
|
| 1 |
---
|
| 2 |
-
base_model:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
library_name: transformers
|
| 4 |
tags:
|
| 5 |
- mergekit
|
| 6 |
- merge
|
| 7 |
-
|
| 8 |
---
|
|
|
|
| 9 |
# Della
|
| 10 |
|
| 11 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
|
@@ -13,18 +21,18 @@ This is a merge of pre-trained language models created using [mergekit](https://
|
|
| 13 |
## Merge Details
|
| 14 |
### Merge Method
|
| 15 |
|
| 16 |
-
This model was merged using the [Linear DELLA](https://arxiv.org/abs/2406.11617) merge method using
|
| 17 |
|
| 18 |
### Models Merged
|
| 19 |
|
| 20 |
The following models were included in the merge:
|
| 21 |
-
*
|
| 22 |
-
*
|
| 23 |
-
*
|
| 24 |
-
*
|
| 25 |
-
*
|
| 26 |
-
*
|
| 27 |
-
*
|
| 28 |
|
| 29 |
### Configuration
|
| 30 |
|
|
@@ -32,16 +40,16 @@ The following YAML configuration was used to produce this model:
|
|
| 32 |
|
| 33 |
```yaml
|
| 34 |
models:
|
| 35 |
-
- model:
|
| 36 |
-
- model:
|
| 37 |
-
- model:
|
| 38 |
-
- model:
|
| 39 |
-
- model:
|
| 40 |
-
- model:
|
| 41 |
-
- model:
|
| 42 |
merge_method: della_linear
|
| 43 |
chat_template: llama3
|
| 44 |
-
base_model:
|
| 45 |
parameters:
|
| 46 |
weight: 0.14
|
| 47 |
density: 0.7
|
|
@@ -50,5 +58,5 @@ parameters:
|
|
| 50 |
normalize: true
|
| 51 |
dtype: bfloat16
|
| 52 |
tokenizer:
|
| 53 |
-
source:
|
| 54 |
```
|
|
|
|
| 1 |
---
|
| 2 |
+
base_model:
|
| 3 |
+
- ArliAI/Llama-3.3-70B-ArliAI-RPMax-v1.4
|
| 4 |
+
- SicariusSicariiStuff/Negative_LLAMA_70B
|
| 5 |
+
- LatitudeGames/Wayfarer-Large-70B-Llama-3.3
|
| 6 |
+
- TheDrummer/Anubis-70B-v1
|
| 7 |
+
- TheDrummer/Fallen-Llama-3.3-R1-70B-v1
|
| 8 |
+
- TareksLab/Genesis-R1-L3.3-70B
|
| 9 |
+
- EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
|
| 10 |
library_name: transformers
|
| 11 |
tags:
|
| 12 |
- mergekit
|
| 13 |
- merge
|
| 14 |
+
license: llama3.3
|
| 15 |
---
|
| 16 |
+
My ideal vision for Dungeonmaster were these 7 models. However I was concerned about that many models in one single. I decided to try both and see...
|
| 17 |
# Della
|
| 18 |
|
| 19 |
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
|
|
|
| 21 |
## Merge Details
|
| 22 |
### Merge Method
|
| 23 |
|
| 24 |
+
This model was merged using the [Linear DELLA](https://arxiv.org/abs/2406.11617) merge method using TareksLab/Genesis-R1-L3.3-70B as a base.
|
| 25 |
|
| 26 |
### Models Merged
|
| 27 |
|
| 28 |
The following models were included in the merge:
|
| 29 |
+
* ArliAI/Llama-3.3-70B-ArliAI-RPMax-v1.4
|
| 30 |
+
* SicariusSicariiStuff/Negative_LLAMA_70B
|
| 31 |
+
* LatitudeGames/Wayfarer-Large-70B-Llama-3.3
|
| 32 |
+
* TheDrummer/Anubis-70B-v1
|
| 33 |
+
* TheDrummer/Fallen-Llama-3.3-R1-70B-v1
|
| 34 |
+
* TareksLab/Genesis-R1-L3.3-70B
|
| 35 |
+
* EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
|
| 36 |
|
| 37 |
### Configuration
|
| 38 |
|
|
|
|
| 40 |
|
| 41 |
```yaml
|
| 42 |
models:
|
| 43 |
+
- model: LatitudeGames/Wayfarer-Large-70B-Llama-3.3
|
| 44 |
+
- model: ArliAI/Llama-3.3-70B-ArliAI-RPMax-v1.4
|
| 45 |
+
- model: Sao10K/70B-L3.3-mhnnn-x1
|
| 46 |
+
- model: TheDrummer/Anubis-70B-v1
|
| 47 |
+
- model: EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
|
| 48 |
+
- model: SicariusSicariiStuff/Negative_LLAMA_70B
|
| 49 |
+
- model: TheDrummer/Fallen-Llama-3.3-R1-70B-v1
|
| 50 |
merge_method: della_linear
|
| 51 |
chat_template: llama3
|
| 52 |
+
base_model: TareksLab/Genesis-R1-L3.3-70B
|
| 53 |
parameters:
|
| 54 |
weight: 0.14
|
| 55 |
density: 0.7
|
|
|
|
| 58 |
normalize: true
|
| 59 |
dtype: bfloat16
|
| 60 |
tokenizer:
|
| 61 |
+
source: TareksLab/Genesis-R1-L3.3-70B
|
| 62 |
```
|