|
--- |
|
base_model: [] |
|
library_name: transformers |
|
tags: |
|
- mergekit |
|
- merge |
|
|
|
--- |
|
## Evolutionary model merge |
|
|
|
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit). |
|
|
|
104 evaluations |
|
|
|
## Merge Details |
|
### Merge Method |
|
|
|
This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using NeuralBeagle14-7B as a base. |
|
|
|
### Models Merged |
|
|
|
The following models were included in the merge: |
|
* Mistral-7B-v0.1-flashback-v2 |
|
* Mistral-7B-Merge-14-v0.2 |
|
* Starling-LM-7B-beta_581094980 |
|
|
|
|
|
|
|
|
|
### Configuration |
|
|
|
The following YAML configuration was used to produce this model: |
|
|
|
```yaml |
|
base_model: /content/evol_merge_storage/input_models/NeuralBeagle14-7B_2368216670 |
|
dtype: bfloat16 |
|
merge_method: dare_ties |
|
parameters: |
|
int8_mask: 1.0 |
|
normalize: 1.0 |
|
slices: |
|
- sources: |
|
- layer_range: [0, 8] |
|
model: /content/evol_merge_storage/input_models/Mistral-7B-v0.1-flashback-v2_2000655885 |
|
parameters: |
|
density: 0.9063003498824225 |
|
weight: 0.2716275746104375 |
|
- layer_range: [0, 8] |
|
model: /content/evol_merge_storage/input_models/Mistral-7B-Merge-14-v0.2_3453453312 |
|
parameters: |
|
density: 0.8605347663753816 |
|
weight: 0.7040535407789865 |
|
- layer_range: [0, 8] |
|
model: /content/evol_merge_storage/input_models/Starling-LM-7B-beta_581094980 |
|
parameters: |
|
density: 1.0 |
|
weight: 0.29417107478605065 |
|
- layer_range: [0, 8] |
|
model: /content/evol_merge_storage/input_models/NeuralBeagle14-7B_2368216670 |
|
- sources: |
|
- layer_range: [8, 16] |
|
model: /content/evol_merge_storage/input_models/Mistral-7B-v0.1-flashback-v2_2000655885 |
|
parameters: |
|
density: 0.9575970148743844 |
|
weight: 0.15956926996874868 |
|
- layer_range: [8, 16] |
|
model: /content/evol_merge_storage/input_models/Mistral-7B-Merge-14-v0.2_3453453312 |
|
parameters: |
|
density: 1.0 |
|
weight: 0.4071229613448434 |
|
- layer_range: [8, 16] |
|
model: /content/evol_merge_storage/input_models/Starling-LM-7B-beta_581094980 |
|
parameters: |
|
density: 1.0 |
|
weight: 0.29267434269480536 |
|
- layer_range: [8, 16] |
|
model: /content/evol_merge_storage/input_models/NeuralBeagle14-7B_2368216670 |
|
- sources: |
|
- layer_range: [16, 24] |
|
model: /content/evol_merge_storage/input_models/Mistral-7B-v0.1-flashback-v2_2000655885 |
|
parameters: |
|
density: 0.853521244265145 |
|
weight: 0.7268702601235844 |
|
- layer_range: [16, 24] |
|
model: /content/evol_merge_storage/input_models/Mistral-7B-Merge-14-v0.2_3453453312 |
|
parameters: |
|
density: 1.0 |
|
weight: 0.3526854709444127 |
|
- layer_range: [16, 24] |
|
model: /content/evol_merge_storage/input_models/Starling-LM-7B-beta_581094980 |
|
parameters: |
|
density: 0.8904104909249966 |
|
weight: 0.565939501390856 |
|
- layer_range: [16, 24] |
|
model: /content/evol_merge_storage/input_models/NeuralBeagle14-7B_2368216670 |
|
- sources: |
|
- layer_range: [24, 32] |
|
model: /content/evol_merge_storage/input_models/Mistral-7B-v0.1-flashback-v2_2000655885 |
|
parameters: |
|
density: 1.0 |
|
weight: 0.3075681562252658 |
|
- layer_range: [24, 32] |
|
model: /content/evol_merge_storage/input_models/Mistral-7B-Merge-14-v0.2_3453453312 |
|
parameters: |
|
density: 0.6564325638087776 |
|
weight: -0.24554943561719403 |
|
- layer_range: [24, 32] |
|
model: /content/evol_merge_storage/input_models/Starling-LM-7B-beta_581094980 |
|
parameters: |
|
density: 0.5678792182777617 |
|
weight: 0.218593901640624 |
|
- layer_range: [24, 32] |
|
model: /content/evol_merge_storage/input_models/NeuralBeagle14-7B_2368216670 |
|
``` |
|
|
