gss1147 commited on
Commit
77fe136
·
verified ·
1 Parent(s): 2ddb882

Upload folder using huggingface_hub

Browse files
README.md ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model:
4
+ - google/flan-t5-small
5
+ - google/flan-t5-small
6
+ - google/flan-t5-small
7
+ - google/flan-t5-small
8
+ - google/flan-t5-small
9
+ - google/flan-t5-small
10
+ - google/flan-t5-small
11
+ tags:
12
+ - moe
13
+ - frankenmoe
14
+ - merge
15
+ - mergekit
16
+ - lazymergekit
17
+ - google/flan-t5-small
18
+ ---
19
+
20
+ # flan-t5-small-moe
21
+
22
+ flan-t5-small-moe is a Mixture of Experts (MoE) made with the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
23
+ * [google/flan-t5-small](https://huggingface.co/google/flan-t5-small)
24
+ * [google/flan-t5-small](https://huggingface.co/google/flan-t5-small)
25
+ * [google/flan-t5-small](https://huggingface.co/google/flan-t5-small)
26
+ * [google/flan-t5-small](https://huggingface.co/google/flan-t5-small)
27
+ * [google/flan-t5-small](https://huggingface.co/google/flan-t5-small)
28
+ * [google/flan-t5-small](https://huggingface.co/google/flan-t5-small)
29
+ * [google/flan-t5-small](https://huggingface.co/google/flan-t5-small)
30
+
31
+ ## ✨ Configuration
32
+
33
+ ```yaml
34
+ base_model: google/flan-t5-small
35
+ merge_method: moe
36
+ gate_mode: hidden
37
+ experts:
38
+ - source_model: google/flan-t5-small
39
+ positive_prompts: ["instruction", "task"]
40
+ - source_model: google/flan-t5-small
41
+ positive_prompts: ["reasoning", "logic"]
42
+ - source_model: google/flan-t5-small
43
+ positive_prompts: ["creative", "writing"]
44
+ - source_model: google/flan-t5-small
45
+ positive_prompts: ["code", "programming"]
46
+ - source_model: google/flan-t5-small
47
+ positive_prompts: ["science", "facts"]
48
+ - source_model: google/flan-t5-small
49
+ positive_prompts: ["math", "calculation"]
50
+ - source_model: google/flan-t5-small
51
+ positive_prompts: ["summary", "dialogue"]
52
+ ```
config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "T5ForConditionalGeneration"
4
+ ],
5
+ "classifier_dropout": 0.0,
6
+ "d_ff": 1024,
7
+ "d_kv": 64,
8
+ "d_model": 512,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "gelu_new",
11
+ "dropout_rate": 0.1,
12
+ "dtype": "float32",
13
+ "eos_token_id": 1,
14
+ "feed_forward_proj": "gated-gelu",
15
+ "initializer_factor": 1.0,
16
+ "is_decoder": false,
17
+ "is_encoder_decoder": true,
18
+ "is_gated_act": true,
19
+ "layer_norm_epsilon": 1e-06,
20
+ "model_type": "t5",
21
+ "n_positions": 512,
22
+ "num_decoder_layers": 8,
23
+ "num_heads": 6,
24
+ "num_layers": 8,
25
+ "output_past": true,
26
+ "pad_token_id": 0,
27
+ "relative_attention_max_distance": 128,
28
+ "relative_attention_num_buckets": 32,
29
+ "scale_decoder_outputs": false,
30
+ "task_specific_params": {
31
+ "summarization": {
32
+ "early_stopping": true,
33
+ "length_penalty": 2.0,
34
+ "max_length": 200,
35
+ "min_length": 30,
36
+ "no_repeat_ngram_size": 3,
37
+ "num_beams": 4,
38
+ "prefix": "summarize: "
39
+ },
40
+ "translation_en_to_de": {
41
+ "early_stopping": true,
42
+ "max_length": 300,
43
+ "num_beams": 4,
44
+ "prefix": "translate English to German: "
45
+ },
46
+ "translation_en_to_fr": {
47
+ "early_stopping": true,
48
+ "max_length": 300,
49
+ "num_beams": 4,
50
+ "prefix": "translate English to French: "
51
+ },
52
+ "translation_en_to_ro": {
53
+ "early_stopping": true,
54
+ "max_length": 300,
55
+ "num_beams": 4,
56
+ "prefix": "translate English to Romanian: "
57
+ }
58
+ },
59
+ "tie_word_embeddings": false,
60
+ "transformers_version": "5.2.0",
61
+ "use_cache": true,
62
+ "vocab_size": 32128
63
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "decoder_start_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "5.2.0"
7
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:495fa51e204676f1a857a9fc13c4c89f3f5ba9f480b898cebca02add25e6d749
3
+ size 307867048
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "eos_token": "</s>",
4
+ "extra_ids": 100,
5
+ "extra_special_tokens": [
6
+ "<extra_id_0>",
7
+ "<extra_id_1>",
8
+ "<extra_id_2>",
9
+ "<extra_id_3>",
10
+ "<extra_id_4>",
11
+ "<extra_id_5>",
12
+ "<extra_id_6>",
13
+ "<extra_id_7>",
14
+ "<extra_id_8>",
15
+ "<extra_id_9>",
16
+ "<extra_id_10>",
17
+ "<extra_id_11>",
18
+ "<extra_id_12>",
19
+ "<extra_id_13>",
20
+ "<extra_id_14>",
21
+ "<extra_id_15>",
22
+ "<extra_id_16>",
23
+ "<extra_id_17>",
24
+ "<extra_id_18>",
25
+ "<extra_id_19>",
26
+ "<extra_id_20>",
27
+ "<extra_id_21>",
28
+ "<extra_id_22>",
29
+ "<extra_id_23>",
30
+ "<extra_id_24>",
31
+ "<extra_id_25>",
32
+ "<extra_id_26>",
33
+ "<extra_id_27>",
34
+ "<extra_id_28>",
35
+ "<extra_id_29>",
36
+ "<extra_id_30>",
37
+ "<extra_id_31>",
38
+ "<extra_id_32>",
39
+ "<extra_id_33>",
40
+ "<extra_id_34>",
41
+ "<extra_id_35>",
42
+ "<extra_id_36>",
43
+ "<extra_id_37>",
44
+ "<extra_id_38>",
45
+ "<extra_id_39>",
46
+ "<extra_id_40>",
47
+ "<extra_id_41>",
48
+ "<extra_id_42>",
49
+ "<extra_id_43>",
50
+ "<extra_id_44>",
51
+ "<extra_id_45>",
52
+ "<extra_id_46>",
53
+ "<extra_id_47>",
54
+ "<extra_id_48>",
55
+ "<extra_id_49>",
56
+ "<extra_id_50>",
57
+ "<extra_id_51>",
58
+ "<extra_id_52>",
59
+ "<extra_id_53>",
60
+ "<extra_id_54>",
61
+ "<extra_id_55>",
62
+ "<extra_id_56>",
63
+ "<extra_id_57>",
64
+ "<extra_id_58>",
65
+ "<extra_id_59>",
66
+ "<extra_id_60>",
67
+ "<extra_id_61>",
68
+ "<extra_id_62>",
69
+ "<extra_id_63>",
70
+ "<extra_id_64>",
71
+ "<extra_id_65>",
72
+ "<extra_id_66>",
73
+ "<extra_id_67>",
74
+ "<extra_id_68>",
75
+ "<extra_id_69>",
76
+ "<extra_id_70>",
77
+ "<extra_id_71>",
78
+ "<extra_id_72>",
79
+ "<extra_id_73>",
80
+ "<extra_id_74>",
81
+ "<extra_id_75>",
82
+ "<extra_id_76>",
83
+ "<extra_id_77>",
84
+ "<extra_id_78>",
85
+ "<extra_id_79>",
86
+ "<extra_id_80>",
87
+ "<extra_id_81>",
88
+ "<extra_id_82>",
89
+ "<extra_id_83>",
90
+ "<extra_id_84>",
91
+ "<extra_id_85>",
92
+ "<extra_id_86>",
93
+ "<extra_id_87>",
94
+ "<extra_id_88>",
95
+ "<extra_id_89>",
96
+ "<extra_id_90>",
97
+ "<extra_id_91>",
98
+ "<extra_id_92>",
99
+ "<extra_id_93>",
100
+ "<extra_id_94>",
101
+ "<extra_id_95>",
102
+ "<extra_id_96>",
103
+ "<extra_id_97>",
104
+ "<extra_id_98>",
105
+ "<extra_id_99>"
106
+ ],
107
+ "is_local": false,
108
+ "model_max_length": 512,
109
+ "pad_token": "<pad>",
110
+ "sp_model_kwargs": {},
111
+ "tokenizer_class": "T5Tokenizer",
112
+ "unk_token": "<unk>"
113
+ }