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
- tokenizer.json +6 -1
- tokenizer_config.json +4 -0
README.md
CHANGED
@@ -1,10 +1,10 @@
|
|
1 |
---
|
2 |
base_model:
|
3 |
-
- mllm-dev/gpt2_f_experiment_0_1000
|
4 |
-
- mllm-dev/gpt2_f_experiment_3_1000
|
5 |
-
- mllm-dev/gpt2_f_experiment_2_1000
|
6 |
- mllm-dev/gpt2_f_experiment_1_1000
|
7 |
- mllm-dev/gpt2_f_experiment_4_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/gpt2_f_experiment_3_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_3_1000)
|
27 |
* [mllm-dev/gpt2_f_experiment_2_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_2_1000)
|
28 |
-
* [mllm-dev/
|
29 |
-
* [mllm-dev/gpt2_f_experiment_4_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_4_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:
|
@@ -45,11 +45,11 @@ slices:
|
|
45 |
- layer_range: [0, 12]
|
46 |
model:
|
47 |
model:
|
48 |
-
path: mllm-dev/
|
49 |
- layer_range: [0, 12]
|
50 |
model:
|
51 |
model:
|
52 |
-
path: mllm-dev/
|
53 |
parameters:
|
54 |
density: 0.8
|
55 |
weight: 0.2
|
|
|
1 |
---
|
2 |
base_model:
|
|
|
|
|
|
|
3 |
- mllm-dev/gpt2_f_experiment_1_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_0_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_1_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_1_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_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_0_1000](https://huggingface.co/mllm-dev/gpt2_f_experiment_0_1000)
|
|
|
30 |
|
31 |
### Configuration
|
32 |
|
|
|
35 |
```yaml
|
36 |
base_model:
|
37 |
model:
|
38 |
+
path: mllm-dev/gpt2_f_experiment_1_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_1_1000
|
49 |
- layer_range: [0, 12]
|
50 |
model:
|
51 |
model:
|
52 |
+
path: mllm-dev/gpt2_f_experiment_0_1000
|
53 |
parameters:
|
54 |
density: 0.8
|
55 |
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_1_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:
|
@@ -10,11 +10,11 @@ slices:
|
|
10 |
- layer_range: [0, 12]
|
11 |
model:
|
12 |
model:
|
13 |
-
path: mllm-dev/
|
14 |
- layer_range: [0, 12]
|
15 |
model:
|
16 |
model:
|
17 |
-
path: mllm-dev/
|
18 |
parameters:
|
19 |
density: 0.8
|
20 |
weight: 0.2
|
|
|
1 |
base_model:
|
2 |
model:
|
3 |
+
path: mllm-dev/gpt2_f_experiment_1_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_1_1000
|
14 |
- layer_range: [0, 12]
|
15 |
model:
|
16 |
model:
|
17 |
+
path: mllm-dev/gpt2_f_experiment_0_1000
|
18 |
parameters:
|
19 |
density: 0.8
|
20 |
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:f0c4e976b3588d122d65ca3d4732e087356109eaa0e662609b3738547512252b
|
3 |
size 248902264
|
tokenizer.json
CHANGED
@@ -1,6 +1,11 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
-
"truncation":
|
|
|
|
|
|
|
|
|
|
|
4 |
"padding": null,
|
5 |
"added_tokens": [
|
6 |
{
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
+
"truncation": {
|
4 |
+
"direction": "Right",
|
5 |
+
"max_length": 1024,
|
6 |
+
"strategy": "LongestFirst",
|
7 |
+
"stride": 0
|
8 |
+
},
|
9 |
"padding": null,
|
10 |
"added_tokens": [
|
11 |
{
|
tokenizer_config.json
CHANGED
@@ -13,8 +13,12 @@
|
|
13 |
"bos_token": "<|endoftext|>",
|
14 |
"clean_up_tokenization_spaces": true,
|
15 |
"eos_token": "<|endoftext|>",
|
|
|
16 |
"model_max_length": 1024,
|
17 |
"pad_token": "<|endoftext|>",
|
|
|
18 |
"tokenizer_class": "GPT2Tokenizer",
|
|
|
|
|
19 |
"unk_token": "<|endoftext|>"
|
20 |
}
|
|
|
13 |
"bos_token": "<|endoftext|>",
|
14 |
"clean_up_tokenization_spaces": true,
|
15 |
"eos_token": "<|endoftext|>",
|
16 |
+
"max_length": 1024,
|
17 |
"model_max_length": 1024,
|
18 |
"pad_token": "<|endoftext|>",
|
19 |
+
"stride": 0,
|
20 |
"tokenizer_class": "GPT2Tokenizer",
|
21 |
+
"truncation_side": "right",
|
22 |
+
"truncation_strategy": "longest_first",
|
23 |
"unk_token": "<|endoftext|>"
|
24 |
}
|