Upload folder using huggingface_hub
Browse files- README.md +5 -13
- mergekit_config.yml +1 -9
- model-00001-of-00001.safetensors +1 -1
README.md
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
---
|
| 2 |
base_model:
|
| 3 |
-
- mllm-dev/gpt2_f_experiment_2
|
| 4 |
- mllm-dev/gpt2_f_experiment_4
|
| 5 |
- mllm-dev/gpt2_f_experiment_1
|
| 6 |
-
- mllm-dev/
|
| 7 |
- mllm-dev/gpt2_f_experiment_0
|
|
|
|
| 8 |
library_name: transformers
|
| 9 |
tags:
|
| 10 |
- mergekit
|
|
@@ -18,14 +18,14 @@ This is a merge of pre-trained language models created using [mergekit](https://
|
|
| 18 |
## Merge Details
|
| 19 |
### Merge Method
|
| 20 |
|
| 21 |
-
This model was merged using the [
|
| 22 |
|
| 23 |
### Models Merged
|
| 24 |
|
| 25 |
The following models were included in the merge:
|
| 26 |
-
* [mllm-dev/gpt2_f_experiment_2](https://huggingface.co/mllm-dev/gpt2_f_experiment_2)
|
| 27 |
* [mllm-dev/gpt2_f_experiment_4](https://huggingface.co/mllm-dev/gpt2_f_experiment_4)
|
| 28 |
* [mllm-dev/gpt2_f_experiment_1](https://huggingface.co/mllm-dev/gpt2_f_experiment_1)
|
|
|
|
| 29 |
* [mllm-dev/gpt2_f_experiment_3](https://huggingface.co/mllm-dev/gpt2_f_experiment_3)
|
| 30 |
|
| 31 |
### Configuration
|
|
@@ -37,10 +37,7 @@ base_model:
|
|
| 37 |
model:
|
| 38 |
path: mllm-dev/gpt2_f_experiment_0
|
| 39 |
dtype: float16
|
| 40 |
-
merge_method:
|
| 41 |
-
parameters:
|
| 42 |
-
int8_mask: 0.0
|
| 43 |
-
normalize: 1.0
|
| 44 |
slices:
|
| 45 |
- sources:
|
| 46 |
- layer_range: [0, 12]
|
|
@@ -48,34 +45,29 @@ slices:
|
|
| 48 |
model:
|
| 49 |
path: mllm-dev/gpt2_f_experiment_0
|
| 50 |
parameters:
|
| 51 |
-
density: 1.0
|
| 52 |
weight: 1.0
|
| 53 |
- layer_range: [0, 12]
|
| 54 |
model:
|
| 55 |
model:
|
| 56 |
path: mllm-dev/gpt2_f_experiment_1
|
| 57 |
parameters:
|
| 58 |
-
density: 1.0
|
| 59 |
weight: 1.0
|
| 60 |
- layer_range: [0, 12]
|
| 61 |
model:
|
| 62 |
model:
|
| 63 |
path: mllm-dev/gpt2_f_experiment_2
|
| 64 |
parameters:
|
| 65 |
-
density: 1.0
|
| 66 |
weight: 1.0
|
| 67 |
- layer_range: [0, 12]
|
| 68 |
model:
|
| 69 |
model:
|
| 70 |
path: mllm-dev/gpt2_f_experiment_3
|
| 71 |
parameters:
|
| 72 |
-
density: 1.0
|
| 73 |
weight: 1.0
|
| 74 |
- layer_range: [0, 12]
|
| 75 |
model:
|
| 76 |
model:
|
| 77 |
path: mllm-dev/gpt2_f_experiment_4
|
| 78 |
parameters:
|
| 79 |
-
density: 1.0
|
| 80 |
weight: 1.0
|
| 81 |
```
|
|
|
|
| 1 |
---
|
| 2 |
base_model:
|
|
|
|
| 3 |
- mllm-dev/gpt2_f_experiment_4
|
| 4 |
- mllm-dev/gpt2_f_experiment_1
|
| 5 |
+
- mllm-dev/gpt2_f_experiment_2
|
| 6 |
- mllm-dev/gpt2_f_experiment_0
|
| 7 |
+
- mllm-dev/gpt2_f_experiment_3
|
| 8 |
library_name: transformers
|
| 9 |
tags:
|
| 10 |
- mergekit
|
|
|
|
| 18 |
## Merge Details
|
| 19 |
### Merge Method
|
| 20 |
|
| 21 |
+
This model was merged using the [linear](https://arxiv.org/abs/2203.05482) merge method using [mllm-dev/gpt2_f_experiment_0](https://huggingface.co/mllm-dev/gpt2_f_experiment_0) as a base.
|
| 22 |
|
| 23 |
### Models Merged
|
| 24 |
|
| 25 |
The following models were included in the merge:
|
|
|
|
| 26 |
* [mllm-dev/gpt2_f_experiment_4](https://huggingface.co/mllm-dev/gpt2_f_experiment_4)
|
| 27 |
* [mllm-dev/gpt2_f_experiment_1](https://huggingface.co/mllm-dev/gpt2_f_experiment_1)
|
| 28 |
+
* [mllm-dev/gpt2_f_experiment_2](https://huggingface.co/mllm-dev/gpt2_f_experiment_2)
|
| 29 |
* [mllm-dev/gpt2_f_experiment_3](https://huggingface.co/mllm-dev/gpt2_f_experiment_3)
|
| 30 |
|
| 31 |
### Configuration
|
|
|
|
| 37 |
model:
|
| 38 |
path: mllm-dev/gpt2_f_experiment_0
|
| 39 |
dtype: float16
|
| 40 |
+
merge_method: linear
|
|
|
|
|
|
|
|
|
|
| 41 |
slices:
|
| 42 |
- sources:
|
| 43 |
- layer_range: [0, 12]
|
|
|
|
| 45 |
model:
|
| 46 |
path: mllm-dev/gpt2_f_experiment_0
|
| 47 |
parameters:
|
|
|
|
| 48 |
weight: 1.0
|
| 49 |
- layer_range: [0, 12]
|
| 50 |
model:
|
| 51 |
model:
|
| 52 |
path: mllm-dev/gpt2_f_experiment_1
|
| 53 |
parameters:
|
|
|
|
| 54 |
weight: 1.0
|
| 55 |
- layer_range: [0, 12]
|
| 56 |
model:
|
| 57 |
model:
|
| 58 |
path: mllm-dev/gpt2_f_experiment_2
|
| 59 |
parameters:
|
|
|
|
| 60 |
weight: 1.0
|
| 61 |
- layer_range: [0, 12]
|
| 62 |
model:
|
| 63 |
model:
|
| 64 |
path: mllm-dev/gpt2_f_experiment_3
|
| 65 |
parameters:
|
|
|
|
| 66 |
weight: 1.0
|
| 67 |
- layer_range: [0, 12]
|
| 68 |
model:
|
| 69 |
model:
|
| 70 |
path: mllm-dev/gpt2_f_experiment_4
|
| 71 |
parameters:
|
|
|
|
| 72 |
weight: 1.0
|
| 73 |
```
|
mergekit_config.yml
CHANGED
|
@@ -2,10 +2,7 @@ base_model:
|
|
| 2 |
model:
|
| 3 |
path: mllm-dev/gpt2_f_experiment_0
|
| 4 |
dtype: float16
|
| 5 |
-
merge_method:
|
| 6 |
-
parameters:
|
| 7 |
-
int8_mask: 0.0
|
| 8 |
-
normalize: 1.0
|
| 9 |
slices:
|
| 10 |
- sources:
|
| 11 |
- layer_range: [0, 12]
|
|
@@ -13,33 +10,28 @@ slices:
|
|
| 13 |
model:
|
| 14 |
path: mllm-dev/gpt2_f_experiment_0
|
| 15 |
parameters:
|
| 16 |
-
density: 1.0
|
| 17 |
weight: 1.0
|
| 18 |
- layer_range: [0, 12]
|
| 19 |
model:
|
| 20 |
model:
|
| 21 |
path: mllm-dev/gpt2_f_experiment_1
|
| 22 |
parameters:
|
| 23 |
-
density: 1.0
|
| 24 |
weight: 1.0
|
| 25 |
- layer_range: [0, 12]
|
| 26 |
model:
|
| 27 |
model:
|
| 28 |
path: mllm-dev/gpt2_f_experiment_2
|
| 29 |
parameters:
|
| 30 |
-
density: 1.0
|
| 31 |
weight: 1.0
|
| 32 |
- layer_range: [0, 12]
|
| 33 |
model:
|
| 34 |
model:
|
| 35 |
path: mllm-dev/gpt2_f_experiment_3
|
| 36 |
parameters:
|
| 37 |
-
density: 1.0
|
| 38 |
weight: 1.0
|
| 39 |
- layer_range: [0, 12]
|
| 40 |
model:
|
| 41 |
model:
|
| 42 |
path: mllm-dev/gpt2_f_experiment_4
|
| 43 |
parameters:
|
| 44 |
-
density: 1.0
|
| 45 |
weight: 1.0
|
|
|
|
| 2 |
model:
|
| 3 |
path: mllm-dev/gpt2_f_experiment_0
|
| 4 |
dtype: float16
|
| 5 |
+
merge_method: linear
|
|
|
|
|
|
|
|
|
|
| 6 |
slices:
|
| 7 |
- sources:
|
| 8 |
- layer_range: [0, 12]
|
|
|
|
| 10 |
model:
|
| 11 |
path: mllm-dev/gpt2_f_experiment_0
|
| 12 |
parameters:
|
|
|
|
| 13 |
weight: 1.0
|
| 14 |
- layer_range: [0, 12]
|
| 15 |
model:
|
| 16 |
model:
|
| 17 |
path: mllm-dev/gpt2_f_experiment_1
|
| 18 |
parameters:
|
|
|
|
| 19 |
weight: 1.0
|
| 20 |
- layer_range: [0, 12]
|
| 21 |
model:
|
| 22 |
model:
|
| 23 |
path: mllm-dev/gpt2_f_experiment_2
|
| 24 |
parameters:
|
|
|
|
| 25 |
weight: 1.0
|
| 26 |
- layer_range: [0, 12]
|
| 27 |
model:
|
| 28 |
model:
|
| 29 |
path: mllm-dev/gpt2_f_experiment_3
|
| 30 |
parameters:
|
|
|
|
| 31 |
weight: 1.0
|
| 32 |
- layer_range: [0, 12]
|
| 33 |
model:
|
| 34 |
model:
|
| 35 |
path: mllm-dev/gpt2_f_experiment_4
|
| 36 |
parameters:
|
|
|
|
| 37 |
weight: 1.0
|
model-00001-of-00001.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 248902264
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:109d22198c42220534f2b55ff9566334f14c2d3c6976f90d83b3d654b92dbc74
|
| 3 |
size 248902264
|