Upload folder using huggingface_hub
Browse files- README.md +9 -9
- config.json +1 -1
- mergekit_config.yml +3 -3
- model-00001-of-00001.safetensors +1 -1
README.md
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
base_model:
|
| 3 |
- mllm-dev/gpt2_f_experiment_0_1000
|
| 4 |
- mllm-dev/gpt2_f_experiment_4_1000
|
| 5 |
-
- mllm-dev/gpt2_f_experiment_3_1000
|
| 6 |
-
- mllm-dev/gpt2_f_experiment_2_1000
|
| 7 |
- mllm-dev/gpt2_f_experiment_1_1000
|
|
|
|
|
|
|
| 8 |
library_name: transformers
|
| 9 |
tags:
|
| 10 |
- mergekit
|
|
@@ -18,15 +18,15 @@ 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 [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [mllm-dev/
|
| 22 |
|
| 23 |
### Models Merged
|
| 24 |
|
| 25 |
The following models were included in the merge:
|
| 26 |
-
* [mllm-dev/
|
| 27 |
-
* [mllm-dev/gpt2_f_experiment_3_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_3_1000)
|
| 28 |
-
* [mllm-dev/gpt2_f_experiment_2_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_2_1000)
|
| 29 |
* [mllm-dev/gpt2_f_experiment_1_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_1_1000)
|
|
|
|
|
|
|
| 30 |
|
| 31 |
### Configuration
|
| 32 |
|
|
@@ -35,7 +35,7 @@ The following YAML configuration was used to produce this model:
|
|
| 35 |
```yaml
|
| 36 |
base_model:
|
| 37 |
model:
|
| 38 |
-
path: mllm-dev/
|
| 39 |
dtype: float16
|
| 40 |
merge_method: dare_ties
|
| 41 |
parameters:
|
|
@@ -45,7 +45,7 @@ slices:
|
|
| 45 |
- layer_range: [0, 12]
|
| 46 |
model:
|
| 47 |
model:
|
| 48 |
-
path: mllm-dev/
|
| 49 |
- layer_range: [0, 12]
|
| 50 |
model:
|
| 51 |
model:
|
|
@@ -70,7 +70,7 @@ slices:
|
|
| 70 |
- layer_range: [0, 12]
|
| 71 |
model:
|
| 72 |
model:
|
| 73 |
-
path: mllm-dev/
|
| 74 |
parameters:
|
| 75 |
density: 0.8
|
| 76 |
weight: 0.2
|
|
|
|
| 2 |
base_model:
|
| 3 |
- mllm-dev/gpt2_f_experiment_0_1000
|
| 4 |
- mllm-dev/gpt2_f_experiment_4_1000
|
|
|
|
|
|
|
| 5 |
- mllm-dev/gpt2_f_experiment_1_1000
|
| 6 |
+
- mllm-dev/gpt2_f_experiment_2_1000
|
| 7 |
+
- mllm-dev/gpt2_f_experiment_3_1000
|
| 8 |
library_name: transformers
|
| 9 |
tags:
|
| 10 |
- mergekit
|
|
|
|
| 18 |
## Merge Details
|
| 19 |
### Merge Method
|
| 20 |
|
| 21 |
+
This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [mllm-dev/gpt2_f_experiment_0_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_0_1000) as a base.
|
| 22 |
|
| 23 |
### Models Merged
|
| 24 |
|
| 25 |
The following models were included in the merge:
|
| 26 |
+
* [mllm-dev/gpt2_f_experiment_4_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_4_1000)
|
|
|
|
|
|
|
| 27 |
* [mllm-dev/gpt2_f_experiment_1_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_1_1000)
|
| 28 |
+
* [mllm-dev/gpt2_f_experiment_2_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_2_1000)
|
| 29 |
+
* [mllm-dev/gpt2_f_experiment_3_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_3_1000)
|
| 30 |
|
| 31 |
### Configuration
|
| 32 |
|
|
|
|
| 35 |
```yaml
|
| 36 |
base_model:
|
| 37 |
model:
|
| 38 |
+
path: mllm-dev/gpt2_f_experiment_0_1000
|
| 39 |
dtype: float16
|
| 40 |
merge_method: dare_ties
|
| 41 |
parameters:
|
|
|
|
| 45 |
- layer_range: [0, 12]
|
| 46 |
model:
|
| 47 |
model:
|
| 48 |
+
path: mllm-dev/gpt2_f_experiment_0_1000
|
| 49 |
- layer_range: [0, 12]
|
| 50 |
model:
|
| 51 |
model:
|
|
|
|
| 70 |
- layer_range: [0, 12]
|
| 71 |
model:
|
| 72 |
model:
|
| 73 |
+
path: mllm-dev/gpt2_f_experiment_4_1000
|
| 74 |
parameters:
|
| 75 |
density: 0.8
|
| 76 |
weight: 0.2
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "mllm-dev/
|
| 3 |
"activation_function": "gelu_new",
|
| 4 |
"architectures": [
|
| 5 |
"GPT2ForSequenceClassification"
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "mllm-dev/gpt2_f_experiment_0_1000",
|
| 3 |
"activation_function": "gelu_new",
|
| 4 |
"architectures": [
|
| 5 |
"GPT2ForSequenceClassification"
|
mergekit_config.yml
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
base_model:
|
| 2 |
model:
|
| 3 |
-
path: mllm-dev/
|
| 4 |
dtype: float16
|
| 5 |
merge_method: dare_ties
|
| 6 |
parameters:
|
|
@@ -10,7 +10,7 @@ slices:
|
|
| 10 |
- layer_range: [0, 12]
|
| 11 |
model:
|
| 12 |
model:
|
| 13 |
-
path: mllm-dev/
|
| 14 |
- layer_range: [0, 12]
|
| 15 |
model:
|
| 16 |
model:
|
|
@@ -35,7 +35,7 @@ slices:
|
|
| 35 |
- layer_range: [0, 12]
|
| 36 |
model:
|
| 37 |
model:
|
| 38 |
-
path: mllm-dev/
|
| 39 |
parameters:
|
| 40 |
density: 0.8
|
| 41 |
weight: 0.2
|
|
|
|
| 1 |
base_model:
|
| 2 |
model:
|
| 3 |
+
path: mllm-dev/gpt2_f_experiment_0_1000
|
| 4 |
dtype: float16
|
| 5 |
merge_method: dare_ties
|
| 6 |
parameters:
|
|
|
|
| 10 |
- layer_range: [0, 12]
|
| 11 |
model:
|
| 12 |
model:
|
| 13 |
+
path: mllm-dev/gpt2_f_experiment_0_1000
|
| 14 |
- layer_range: [0, 12]
|
| 15 |
model:
|
| 16 |
model:
|
|
|
|
| 35 |
- layer_range: [0, 12]
|
| 36 |
model:
|
| 37 |
model:
|
| 38 |
+
path: mllm-dev/gpt2_f_experiment_4_1000
|
| 39 |
parameters:
|
| 40 |
density: 0.8
|
| 41 |
weight: 0.2
|
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 sha256:3d420ef69a28358b3734eac69f3840aa3558517eec8921307dad490d29ddf218
|
| 3 |
size 248902264
|