Hugging Face's logo

virtuous7373
/
Mira-Demeter

Safetensors
GGUF
gemma3
mergekit
Merge
Lambent
Model card Files
xet
Community
virtuous7373 commited on
Commit
6b31a7a
·
verified ·
1 Parent(s): 5b90f85

Upload folder using huggingface_hub

Browse files
Files changed (25) hide show
  1. .gitattributes +1 -0
  2. added_tokens.json +3 -0
  3. chat_template.jinja +40 -0
  4. config.json +123 -0
  5. generation_config.json +9 -0
  6. mergekit_config.yml +92 -0
  7. model-00001-of-00012.safetensors +3 -0
  8. model-00002-of-00012.safetensors +3 -0
  9. model-00003-of-00012.safetensors +3 -0
  10. model-00004-of-00012.safetensors +3 -0
  11. model-00005-of-00012.safetensors +3 -0
  12. model-00006-of-00012.safetensors +3 -0
  13. model-00007-of-00012.safetensors +3 -0
  14. model-00008-of-00012.safetensors +3 -0
  15. model-00009-of-00012.safetensors +3 -0
  16. model-00010-of-00012.safetensors +3 -0
  17. model-00011-of-00012.safetensors +3 -0
  18. model-00012-of-00012.safetensors +3 -0
  19. model.safetensors.index.json +0 -0
  20. preprocessor_config.json +29 -0
  21. processor_config.json +4 -0
  22. special_tokens_map.json +33 -0
  23. tokenizer.json +3 -0
  24. tokenizer.model +3 -0
  25. tokenizer_config.json +0 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<image_soft_token>": 262144
3
+ }
chat_template.jinja ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {{ bos_token }}
2
+ {%- if messages[0]['role'] == 'system' -%}
3
+ {%- if messages[0]['content'] is string -%}
4
+ {%- set first_user_prefix = messages[0]['content'] + '\n\n' -%}
5
+ {%- else -%}
6
+ {%- set first_user_prefix = messages[0]['content'][0]['text'] + '\n\n' -%}
7
+ {%- endif -%}
8
+ {%- set loop_messages = messages[1:] -%}
9
+ {%- else -%}
10
+ {%- set first_user_prefix = "" -%}
11
+ {%- set loop_messages = messages -%}
12
+ {%- endif -%}
13
+ {%- for message in loop_messages -%}
14
+ {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
15
+ {{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
16
+ {%- endif -%}
17
+ {%- if (message['role'] == 'assistant') -%}
18
+ {%- set role = "model" -%}
19
+ {%- else -%}
20
+ {%- set role = message['role'] -%}
21
+ {%- endif -%}
22
+ {{ '<start_of_turn>' + role + '\n' + (first_user_prefix if loop.first else "") }}
23
+ {%- if message['content'] is string -%}
24
+ {{ message['content'] | trim }}
25
+ {%- elif message['content'] is iterable -%}
26
+ {%- for item in message['content'] -%}
27
+ {%- if item['type'] == 'image' -%}
28
+ {{ '<start_of_image>' }}
29
+ {%- elif item['type'] == 'text' -%}
30
+ {{ item['text'] | trim }}
31
+ {%- endif -%}
32
+ {%- endfor -%}
33
+ {%- else -%}
34
+ {{ raise_exception("Invalid content type") }}
35
+ {%- endif -%}
36
+ {{ '<end_of_turn>\n' }}
37
+ {%- endfor -%}
38
+ {%- if add_generation_prompt -%}
39
+ {{'<start_of_turn>model\n'}}
40
+ {%- endif -%}
config.json ADDED
@@ -0,0 +1,123 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForConditionalGeneration"
4
+ ],
5
+ "boi_token_index": 255999,
6
+ "dtype": "bfloat16",
7
+ "eoi_token_index": 256000,
8
+ "image_token_index": 262144,
9
+ "initializer_range": 0.02,
10
+ "mm_tokens_per_image": 256,
11
+ "model_type": "gemma3",
12
+ "text_config": {
13
+ "_sliding_window_pattern": 6,
14
+ "attention_bias": false,
15
+ "attention_dropout": 0.0,
16
+ "attn_logit_softcapping": null,
17
+ "dtype": "bfloat16",
18
+ "final_logit_softcapping": null,
19
+ "head_dim": 128,
20
+ "hidden_activation": "gelu_pytorch_tanh",
21
+ "hidden_size": 5376,
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 21504,
24
+ "layer_types": [
25
+ "sliding_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "full_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "full_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention",
55
+ "sliding_attention",
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "sliding_attention",
59
+ "sliding_attention",
60
+ "full_attention",
61
+ "sliding_attention",
62
+ "sliding_attention",
63
+ "sliding_attention",
64
+ "sliding_attention",
65
+ "sliding_attention",
66
+ "full_attention",
67
+ "sliding_attention",
68
+ "sliding_attention",
69
+ "sliding_attention",
70
+ "sliding_attention",
71
+ "sliding_attention",
72
+ "full_attention",
73
+ "sliding_attention",
74
+ "sliding_attention",
75
+ "sliding_attention",
76
+ "sliding_attention",
77
+ "sliding_attention",
78
+ "full_attention",
79
+ "sliding_attention",
80
+ "sliding_attention",
81
+ "sliding_attention",
82
+ "sliding_attention",
83
+ "sliding_attention",
84
+ "full_attention",
85
+ "sliding_attention",
86
+ "sliding_attention"
87
+ ],
88
+ "max_position_embeddings": 131072,
89
+ "model_type": "gemma3_text",
90
+ "num_attention_heads": 32,
91
+ "num_hidden_layers": 62,
92
+ "num_key_value_heads": 16,
93
+ "query_pre_attn_scalar": 168,
94
+ "rms_norm_eps": 1e-06,
95
+ "rope_local_base_freq": 10000.0,
96
+ "rope_scaling": {
97
+ "factor": 8.0,
98
+ "rope_type": "linear"
99
+ },
100
+ "rope_theta": 1000000.0,
101
+ "sliding_window": 1024,
102
+ "use_bidirectional_attention": false,
103
+ "use_cache": false,
104
+ "vocab_size": 262145
105
+ },
106
+ "transformers_version": "4.56.1",
107
+ "use_cache": true,
108
+ "vision_config": {
109
+ "attention_dropout": 0.0,
110
+ "dtype": "bfloat16",
111
+ "hidden_act": "gelu_pytorch_tanh",
112
+ "hidden_size": 1152,
113
+ "image_size": 896,
114
+ "intermediate_size": 4304,
115
+ "layer_norm_eps": 1e-06,
116
+ "model_type": "siglip_vision_model",
117
+ "num_attention_heads": 16,
118
+ "num_channels": 3,
119
+ "num_hidden_layers": 27,
120
+ "patch_size": 14,
121
+ "vision_use_head": false
122
+ }
123
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 2,
4
+ "do_sample": true,
5
+ "eos_token_id": 1,
6
+ "pad_token_id": 0,
7
+ "transformers_version": "4.56.1",
8
+ "use_cache": false
9
+ }
mergekit_config.yml ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ name: Mira-Aphrodite-v19-freckles
2
+
3
+ merge_method: magstock
4
+
5
+ models:
6
+ - model: ./Ingredients/Lambent_Mira-v1.8.1a-27B
7
+ parameters:
8
+ weight: 0.5
9
+ density: 0.5
10
+ epsilon: 0.4
11
+ - model: ./Ingredients/Mira-v14 # karcher
12
+ parameters:
13
+ weight: 0.5
14
+ density: 0.5
15
+
16
+ base_model: ./Ingredients/Lambent_Mira-v1.8.1a-27B
17
+
18
+ dtype: bfloat16
19
+
20
+ parameters:
21
+ rescale: true
22
+ t_factor: 1.0
23
+
24
+ tokenizer:
25
+ source: union
26
+
27
+ ---
28
+ name: Mira-Aphrodite-v31
29
+
30
+ merge_method: magvar
31
+
32
+ base_model: ./Ingredients/Lambent_Mira-v1.21-27B-rlvr
33
+ models:
34
+ - model: ./Ingredients/Lambent_Mira-v1.21-27B-rlvr
35
+ parameters:
36
+ weight: 0.40
37
+ - model: ./Ingredients/Lambent_Mira-v1.20-27B-dpo
38
+ parameters:
39
+ weight: 0.20
40
+ - model: ./Ingredients/Lambent_Mira-v1.19.3c-27B-dpo
41
+ parameters:
42
+ weight: 0.20
43
+ - model: ./Ingredients/Lambent_Mira-v1.19-27B-dpo
44
+ parameters:
45
+ weight: 0.20
46
+
47
+ parameters:
48
+ top_k: 0.45
49
+ eps: 1e-6
50
+ bond: 0.2
51
+
52
+ tokenizer:
53
+ source: base
54
+
55
+ dtype: float32
56
+ out_dtype: bfloat16
57
+
58
+ ---
59
+ name: Mira-Demeter
60
+
61
+ merge_method: cabs
62
+
63
+ base_model: ./Ingredients/Lambent_Mira-v1.21-27B-rlvr/
64
+
65
+ models:
66
+ - model: ./Ingredients/Lambent_Mira-v1.21-27B-rlvr/
67
+ - model: ./Ingredients/Mira-Aphrodite-v31/
68
+ parameters:
69
+ weight: 0.6
70
+ n_val: 16
71
+ m_val: 32
72
+ - model: ./Ingredients/Mira-Aphrodite-v19-freckles/
73
+ parameters:
74
+ weight: 0.4
75
+ n_val: 11
76
+ m_val: 33
77
+
78
+
79
+ default_n_val: 8
80
+ default_m_val: 32
81
+
82
+ pruning_order:
83
+ - ./Ingredients/Mira-Aphrodite-v31/
84
+ - ./Ingredients/Mira-Aphrodite-v19-freckles/
85
+
86
+ # base_model: base
87
+
88
+ dtype: float32
89
+ out_dtype: bfloat16
90
+
91
+ tokenizer:
92
+ source: union
model-00001-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b3c8ee9198858d54a5f11f911bc74422d91db4c04c02a200561385dfa2644fb
3
+ size 4932614000
model-00002-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7faab4341c9be91ba09e731bff498379ab264f6d33eba8d9eef38c47816fe22
3
+ size 4954793016
model-00003-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3765d66952d026556956021082ae05457c7dc8d2afa6a1d7678cd0ae03cfca44
3
+ size 4954793008
model-00004-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7adfbfb8a1c48a8b0069614af233e8af3c558975b868667477e4ebddbcec6fdd
3
+ size 4954793016
model-00005-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80877b590b4d922b51b2089022e134d61c4ca007bfe84dab0ae748440d35387b
3
+ size 4954793008
model-00006-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95a9d064a758656b78aa449f2fc3b398f7587fab0c96b13189561ae19351eb70
3
+ size 4954793016
model-00007-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d244811ada2a89bb93f1589e28e30eede0e01f7279d569011a535c192cb060b5
3
+ size 4954793008
model-00008-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7117643b3a1cb23fc307647408dc873acc59107e0ce9d8caa0329156e88b69ad
3
+ size 4954793016
model-00009-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a9743d7c16f9ee772caf959cf3a24400dae22751f747b0a95288fda3e595181
3
+ size 4954793008
model-00010-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eda2273975ca3c9eaeddd2f547b5d50787c77312a5ec02f31812cba2c98a6cbf
3
+ size 4954793016
model-00011-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3df53ad3c515176fc1ab3a7cee6ab52d03848efa211aec75c9ca019fa851dbf2
3
+ size 4997942592
model-00012-of-00012.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:938c1a631ba2c5f657ccf9c2010ba6b7d40adced93be1fea78b4449e099746f7
3
+ size 340608920
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
preprocessor_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": null,
3
+ "do_normalize": true,
4
+ "do_pan_and_scan": null,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "image_mean": [
8
+ 0.5,
9
+ 0.5,
10
+ 0.5
11
+ ],
12
+ "image_processor_type": "Gemma3ImageProcessor",
13
+ "image_seq_length": 256,
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "pan_and_scan_max_num_crops": null,
20
+ "pan_and_scan_min_crop_size": null,
21
+ "pan_and_scan_min_ratio_to_activate": null,
22
+ "processor_class": "Gemma3Processor",
23
+ "resample": 2,
24
+ "rescale_factor": 0.00392156862745098,
25
+ "size": {
26
+ "height": 896,
27
+ "width": 896
28
+ }
29
+ }
processor_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "image_seq_length": 256,
3
+ "processor_class": "Gemma3Processor"
4
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "boi_token": "<start_of_image>",
3
+ "bos_token": {
4
+ "content": "<bos>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "eoi_token": "<end_of_image>",
11
+ "eos_token": {
12
+ "content": "<eos>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "image_token": "<image_soft_token>",
19
+ "pad_token": {
20
+ "content": "<pad>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "unk_token": {
27
+ "content": "<unk>",
28
+ "lstrip": false,
29
+ "normalized": false,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ }
33
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
3
+ size 33384568
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff