IMvision12 commited on
Commit
63b82ba
·
0 Parent(s):

Super-squash branch 'main' using huggingface_hub

Browse files
.gitattributes ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz 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
README.md ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ pipeline_tag: image-text-to-text
3
+ license: mit
4
+ base_model: deepseek-ai/Janus-Pro-7B
5
+ library_name: kerasformers
6
+ tags:
7
+ - keras
8
+ - kerasformers
9
+ - janus
10
+ - janus-pro
11
+ - vision-language
12
+ - image-text-to-text
13
+ - arxiv:2501.17811
14
+ - pytorch
15
+ - jax
16
+ - tf
17
+ ---
18
+
19
+ ## ***See [our collection](https://huggingface.co/collections/kerasformers/janus-pro-6a6eb583a6ae12da1d14c227) for all versions of Janus-Pro.***
20
+
21
+ # Run Janus-Pro with Keras 3: JAX, PyTorch, or TensorFlow
22
+
23
+ [![GitHub](https://img.shields.io/badge/GitHub-KerasFormers-black?logo=github)](https://github.com/IMvision12/KerasFormers) [![Docs](https://img.shields.io/badge/Docs-Janus--Pro-blue)](https://imvision12.github.io/KerasFormers/janus/) [![Collection](https://img.shields.io/badge/HF-Janus--Pro%20collection-yellow)](https://huggingface.co/collections/kerasformers/janus-pro-6a6eb583a6ae12da1d14c227)
24
+
25
+ # kerasformers/janus_pro_7b
26
+
27
+ Paper: [Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling (arXiv:2501.17811)](https://arxiv.org/abs/2501.17811) · [HF Papers](https://huggingface.co/papers/2501.17811)
28
+
29
+ Janus-Pro is a multimodal model (SigLIP tower + GELU aligner + Llama decoder). This KerasFormers port covers the **understanding** path only (image + text → text). Multi-image conversations are supported; VQ image generation is not ported.
30
+
31
+ For more details on the model, please go to the upstream [model card](https://huggingface.co/deepseek-ai/Janus-Pro-7B).
32
+
33
+ Pure-**Keras 3** conversion of [`deepseek-ai/Janus-Pro-7B`](https://huggingface.co/deepseek-ai/Janus-Pro-7B) for [kerasformers](https://github.com/IMvision12/KerasFormers). One implementation runs unmodified on **TensorFlow / Torch / JAX**.
34
+
35
+ This is a **vision-language** checkpoint (`JanusConditionalGenerate`, 7B).
36
+
37
+ ## ✨ Quick start
38
+
39
+ ```python
40
+ import os
41
+ os.environ["KERAS_BACKEND"] = "torch" # or "jax" / "tensorflow"
42
+
43
+ from PIL import Image
44
+ from kerasformers.models.janus import JanusConditionalGenerate, JanusProcessor
45
+
46
+ model = JanusConditionalGenerate.from_weights("kerasformers/janus_pro_7b")
47
+ processor = JanusProcessor.from_weights("kerasformers/janus_pro_7b")
48
+
49
+ image = Image.open("your_image.jpg")
50
+ inputs = processor(
51
+ conversation=[
52
+ {
53
+ "role": "user",
54
+ "content": [
55
+ {"type": "image", "image": image},
56
+ {"type": "text", "text": "Describe this image in one sentence."},
57
+ ],
58
+ }
59
+ ]
60
+ )
61
+ outputs = model.generate(**inputs, max_new_tokens=64)
62
+ print(processor.decode(outputs[0]))
63
+ ```
64
+
65
+ Load any Janus-Pro variant the same way with `from_weights("kerasformers/<variant>")`:
66
+
67
+ | Variant | Hub |
68
+ |---|---|
69
+ | `janus_pro_1b` | [`kerasformers/janus_pro_1b`](https://huggingface.co/kerasformers/janus_pro_1b) |
70
+ | `janus_pro_7b` | [`kerasformers/janus_pro_7b`](https://huggingface.co/kerasformers/janus_pro_7b) |
71
+
72
+ ## Tips
73
+
74
+ - Set `KERAS_BACKEND` **before** importing Keras / kerasformers.
75
+ - Prefer `JanusProcessor.from_weights(...)` so image size and tokenizer match.
76
+ - Add multiple `{"type": "image", ...}` items for multi-image chats.
77
+ - See [Janus-Pro docs](https://imvision12.github.io/KerasFormers/janus/) and [Loading Weights](https://imvision12.github.io/KerasFormers/loading_weights/).
78
+ - Community / upstream safetensors still work via the `hf:` prefix, e.g. `JanusConditionalGenerate.from_weights("hf:deepseek-ai/Janus-Pro-7B")`.
79
+
80
+ ## Special Thanks
81
+
82
+ A huge thank you to the DeepSeek Janus-Pro authors for creating and releasing these models.
83
+
84
+ License: MIT.
kf_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "library_name": "kerasformers",
3
+ "kerasformers_version": "1.2.1",
4
+ "model_module": "kerasformers.models.janus",
5
+ "model_class": "JanusModel",
6
+ "variant": "janus_pro_7b",
7
+ "weights": "model.weights.json",
8
+ "schema_version": 2,
9
+ "weight_dtype": "bfloat16",
10
+ "model_type": "janus",
11
+ "text_config": {
12
+ "vocab_size": 102400,
13
+ "embed_dim": 4096,
14
+ "mlp_dim": 11008,
15
+ "num_layers": 30,
16
+ "num_heads": 32,
17
+ "num_kv_heads": 32,
18
+ "head_dim": 128,
19
+ "norm_eps": 1e-06,
20
+ "rope_theta": 10000.0,
21
+ "tie_embeddings": false
22
+ },
23
+ "vision_config": {
24
+ "embed_dim": 1024,
25
+ "mlp_dim": 4096,
26
+ "num_layers": 24,
27
+ "num_heads": 16,
28
+ "norm_eps": 1e-06,
29
+ "image_size": 384,
30
+ "patch_size": 16
31
+ },
32
+ "image_token_id": 100594
33
+ }
kf_preprocessor.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "library_name": "kerasformers",
3
+ "kerasformers_version": "1.1.3",
4
+ "preprocessor_module": "kerasformers.models.janus",
5
+ "preprocessor_class": "JanusImageProcessor",
6
+ "variant": "janus_pro_7b",
7
+ "size": 384,
8
+ "min_size": 14,
9
+ "background_color": [
10
+ 127,
11
+ 127,
12
+ 127
13
+ ],
14
+ "image_mean": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "image_std": [
20
+ 0.5,
21
+ 0.5,
22
+ 0.5
23
+ ],
24
+ "data_format": null
25
+ }
model.weights.json ADDED
@@ -0,0 +1,868 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 13631207424.0
4
+ },
5
+ "weight_map": {
6
+ "/layers/janus_vision_model/blocks/janus_vision_layer/attention/key/vars": [
7
+ "model_00000.weights.h5"
8
+ ],
9
+ "/layers/janus_vision_model/blocks/janus_vision_layer/attention/output_proj/vars": [
10
+ "model_00000.weights.h5"
11
+ ],
12
+ "/layers/janus_vision_model/blocks/janus_vision_layer/attention/query/vars": [
13
+ "model_00000.weights.h5"
14
+ ],
15
+ "/layers/janus_vision_model/blocks/janus_vision_layer/attention/value/vars": [
16
+ "model_00000.weights.h5"
17
+ ],
18
+ "/layers/janus_vision_model/blocks/janus_vision_layer/fc1/vars": [
19
+ "model_00000.weights.h5"
20
+ ],
21
+ "/layers/janus_vision_model/blocks/janus_vision_layer/fc2/vars": [
22
+ "model_00000.weights.h5"
23
+ ],
24
+ "/layers/janus_vision_model/blocks/janus_vision_layer/layer_norm1/vars": [
25
+ "model_00000.weights.h5"
26
+ ],
27
+ "/layers/janus_vision_model/blocks/janus_vision_layer/layer_norm2/vars": [
28
+ "model_00000.weights.h5"
29
+ ],
30
+ "/layers/janus_vision_model/blocks/janus_vision_layer_1/attention/key/vars": [
31
+ "model_00000.weights.h5"
32
+ ],
33
+ "/layers/janus_vision_model/blocks/janus_vision_layer_1/attention/output_proj/vars": [
34
+ "model_00000.weights.h5"
35
+ ],
36
+ "/layers/janus_vision_model/blocks/janus_vision_layer_1/attention/query/vars": [
37
+ "model_00000.weights.h5"
38
+ ],
39
+ "/layers/janus_vision_model/blocks/janus_vision_layer_1/attention/value/vars": [
40
+ "model_00000.weights.h5"
41
+ ],
42
+ "/layers/janus_vision_model/blocks/janus_vision_layer_1/fc1/vars": [
43
+ "model_00000.weights.h5"
44
+ ],
45
+ "/layers/janus_vision_model/blocks/janus_vision_layer_1/fc2/vars": [
46
+ "model_00000.weights.h5"
47
+ ],
48
+ "/layers/janus_vision_model/blocks/janus_vision_layer_1/layer_norm1/vars": [
49
+ "model_00000.weights.h5"
50
+ ],
51
+ "/layers/janus_vision_model/blocks/janus_vision_layer_1/layer_norm2/vars": [
52
+ "model_00000.weights.h5"
53
+ ],
54
+ "/layers/janus_vision_model/blocks/janus_vision_layer_2/attention/key/vars": [
55
+ "model_00000.weights.h5"
56
+ ],
57
+ "/layers/janus_vision_model/blocks/janus_vision_layer_2/attention/output_proj/vars": [
58
+ "model_00000.weights.h5"
59
+ ],
60
+ "/layers/janus_vision_model/blocks/janus_vision_layer_2/attention/query/vars": [
61
+ "model_00000.weights.h5"
62
+ ],
63
+ "/layers/janus_vision_model/blocks/janus_vision_layer_2/attention/value/vars": [
64
+ "model_00000.weights.h5"
65
+ ],
66
+ "/layers/janus_vision_model/blocks/janus_vision_layer_2/fc1/vars": [
67
+ "model_00000.weights.h5"
68
+ ],
69
+ "/layers/janus_vision_model/blocks/janus_vision_layer_2/fc2/vars": [
70
+ "model_00000.weights.h5"
71
+ ],
72
+ "/layers/janus_vision_model/blocks/janus_vision_layer_2/layer_norm1/vars": [
73
+ "model_00000.weights.h5"
74
+ ],
75
+ "/layers/janus_vision_model/blocks/janus_vision_layer_2/layer_norm2/vars": [
76
+ "model_00000.weights.h5"
77
+ ],
78
+ "/layers/janus_vision_model/blocks/janus_vision_layer_3/attention/key/vars": [
79
+ "model_00000.weights.h5"
80
+ ],
81
+ "/layers/janus_vision_model/blocks/janus_vision_layer_3/attention/output_proj/vars": [
82
+ "model_00000.weights.h5"
83
+ ],
84
+ "/layers/janus_vision_model/blocks/janus_vision_layer_3/attention/query/vars": [
85
+ "model_00000.weights.h5"
86
+ ],
87
+ "/layers/janus_vision_model/blocks/janus_vision_layer_3/attention/value/vars": [
88
+ "model_00000.weights.h5"
89
+ ],
90
+ "/layers/janus_vision_model/blocks/janus_vision_layer_3/fc1/vars": [
91
+ "model_00000.weights.h5"
92
+ ],
93
+ "/layers/janus_vision_model/blocks/janus_vision_layer_3/fc2/vars": [
94
+ "model_00000.weights.h5"
95
+ ],
96
+ "/layers/janus_vision_model/blocks/janus_vision_layer_3/layer_norm1/vars": [
97
+ "model_00000.weights.h5"
98
+ ],
99
+ "/layers/janus_vision_model/blocks/janus_vision_layer_3/layer_norm2/vars": [
100
+ "model_00000.weights.h5"
101
+ ],
102
+ "/layers/janus_vision_model/blocks/janus_vision_layer_4/attention/key/vars": [
103
+ "model_00000.weights.h5"
104
+ ],
105
+ "/layers/janus_vision_model/blocks/janus_vision_layer_4/attention/output_proj/vars": [
106
+ "model_00000.weights.h5"
107
+ ],
108
+ "/layers/janus_vision_model/blocks/janus_vision_layer_4/attention/query/vars": [
109
+ "model_00000.weights.h5"
110
+ ],
111
+ "/layers/janus_vision_model/blocks/janus_vision_layer_4/attention/value/vars": [
112
+ "model_00000.weights.h5"
113
+ ],
114
+ "/layers/janus_vision_model/blocks/janus_vision_layer_4/fc1/vars": [
115
+ "model_00000.weights.h5"
116
+ ],
117
+ "/layers/janus_vision_model/blocks/janus_vision_layer_4/fc2/vars": [
118
+ "model_00000.weights.h5"
119
+ ],
120
+ "/layers/janus_vision_model/blocks/janus_vision_layer_4/layer_norm1/vars": [
121
+ "model_00000.weights.h5"
122
+ ],
123
+ "/layers/janus_vision_model/blocks/janus_vision_layer_4/layer_norm2/vars": [
124
+ "model_00000.weights.h5"
125
+ ],
126
+ "/layers/janus_vision_model/blocks/janus_vision_layer_5/attention/key/vars": [
127
+ "model_00000.weights.h5"
128
+ ],
129
+ "/layers/janus_vision_model/blocks/janus_vision_layer_5/attention/output_proj/vars": [
130
+ "model_00000.weights.h5"
131
+ ],
132
+ "/layers/janus_vision_model/blocks/janus_vision_layer_5/attention/query/vars": [
133
+ "model_00000.weights.h5"
134
+ ],
135
+ "/layers/janus_vision_model/blocks/janus_vision_layer_5/attention/value/vars": [
136
+ "model_00000.weights.h5"
137
+ ],
138
+ "/layers/janus_vision_model/blocks/janus_vision_layer_5/fc1/vars": [
139
+ "model_00000.weights.h5"
140
+ ],
141
+ "/layers/janus_vision_model/blocks/janus_vision_layer_5/fc2/vars": [
142
+ "model_00000.weights.h5"
143
+ ],
144
+ "/layers/janus_vision_model/blocks/janus_vision_layer_5/layer_norm1/vars": [
145
+ "model_00000.weights.h5"
146
+ ],
147
+ "/layers/janus_vision_model/blocks/janus_vision_layer_5/layer_norm2/vars": [
148
+ "model_00000.weights.h5"
149
+ ],
150
+ "/layers/janus_vision_model/blocks/janus_vision_layer_6/attention/key/vars": [
151
+ "model_00000.weights.h5"
152
+ ],
153
+ "/layers/janus_vision_model/blocks/janus_vision_layer_6/attention/output_proj/vars": [
154
+ "model_00000.weights.h5"
155
+ ],
156
+ "/layers/janus_vision_model/blocks/janus_vision_layer_6/attention/query/vars": [
157
+ "model_00000.weights.h5"
158
+ ],
159
+ "/layers/janus_vision_model/blocks/janus_vision_layer_6/attention/value/vars": [
160
+ "model_00000.weights.h5"
161
+ ],
162
+ "/layers/janus_vision_model/blocks/janus_vision_layer_6/fc1/vars": [
163
+ "model_00000.weights.h5"
164
+ ],
165
+ "/layers/janus_vision_model/blocks/janus_vision_layer_6/fc2/vars": [
166
+ "model_00000.weights.h5"
167
+ ],
168
+ "/layers/janus_vision_model/blocks/janus_vision_layer_6/layer_norm1/vars": [
169
+ "model_00000.weights.h5"
170
+ ],
171
+ "/layers/janus_vision_model/blocks/janus_vision_layer_6/layer_norm2/vars": [
172
+ "model_00000.weights.h5"
173
+ ],
174
+ "/layers/janus_vision_model/blocks/janus_vision_layer_7/attention/key/vars": [
175
+ "model_00000.weights.h5"
176
+ ],
177
+ "/layers/janus_vision_model/blocks/janus_vision_layer_7/attention/output_proj/vars": [
178
+ "model_00000.weights.h5"
179
+ ],
180
+ "/layers/janus_vision_model/blocks/janus_vision_layer_7/attention/query/vars": [
181
+ "model_00000.weights.h5"
182
+ ],
183
+ "/layers/janus_vision_model/blocks/janus_vision_layer_7/attention/value/vars": [
184
+ "model_00000.weights.h5"
185
+ ],
186
+ "/layers/janus_vision_model/blocks/janus_vision_layer_7/fc1/vars": [
187
+ "model_00000.weights.h5"
188
+ ],
189
+ "/layers/janus_vision_model/blocks/janus_vision_layer_7/fc2/vars": [
190
+ "model_00000.weights.h5"
191
+ ],
192
+ "/layers/janus_vision_model/blocks/janus_vision_layer_7/layer_norm1/vars": [
193
+ "model_00000.weights.h5"
194
+ ],
195
+ "/layers/janus_vision_model/blocks/janus_vision_layer_7/layer_norm2/vars": [
196
+ "model_00000.weights.h5"
197
+ ],
198
+ "/layers/janus_vision_model/blocks/janus_vision_layer_8/attention/key/vars": [
199
+ "model_00000.weights.h5"
200
+ ],
201
+ "/layers/janus_vision_model/blocks/janus_vision_layer_8/attention/output_proj/vars": [
202
+ "model_00000.weights.h5"
203
+ ],
204
+ "/layers/janus_vision_model/blocks/janus_vision_layer_8/attention/query/vars": [
205
+ "model_00000.weights.h5"
206
+ ],
207
+ "/layers/janus_vision_model/blocks/janus_vision_layer_8/attention/value/vars": [
208
+ "model_00000.weights.h5"
209
+ ],
210
+ "/layers/janus_vision_model/blocks/janus_vision_layer_8/fc1/vars": [
211
+ "model_00000.weights.h5"
212
+ ],
213
+ "/layers/janus_vision_model/blocks/janus_vision_layer_8/fc2/vars": [
214
+ "model_00000.weights.h5"
215
+ ],
216
+ "/layers/janus_vision_model/blocks/janus_vision_layer_8/layer_norm1/vars": [
217
+ "model_00000.weights.h5"
218
+ ],
219
+ "/layers/janus_vision_model/blocks/janus_vision_layer_8/layer_norm2/vars": [
220
+ "model_00000.weights.h5"
221
+ ],
222
+ "/layers/janus_vision_model/blocks/janus_vision_layer_9/attention/key/vars": [
223
+ "model_00000.weights.h5"
224
+ ],
225
+ "/layers/janus_vision_model/blocks/janus_vision_layer_9/attention/output_proj/vars": [
226
+ "model_00000.weights.h5"
227
+ ],
228
+ "/layers/janus_vision_model/blocks/janus_vision_layer_9/attention/query/vars": [
229
+ "model_00000.weights.h5"
230
+ ],
231
+ "/layers/janus_vision_model/blocks/janus_vision_layer_9/attention/value/vars": [
232
+ "model_00000.weights.h5"
233
+ ],
234
+ "/layers/janus_vision_model/blocks/janus_vision_layer_9/fc1/vars": [
235
+ "model_00000.weights.h5"
236
+ ],
237
+ "/layers/janus_vision_model/blocks/janus_vision_layer_9/fc2/vars": [
238
+ "model_00000.weights.h5"
239
+ ],
240
+ "/layers/janus_vision_model/blocks/janus_vision_layer_9/layer_norm1/vars": [
241
+ "model_00000.weights.h5"
242
+ ],
243
+ "/layers/janus_vision_model/blocks/janus_vision_layer_9/layer_norm2/vars": [
244
+ "model_00000.weights.h5"
245
+ ],
246
+ "/layers/janus_vision_model/blocks/janus_vision_layer_10/attention/key/vars": [
247
+ "model_00000.weights.h5"
248
+ ],
249
+ "/layers/janus_vision_model/blocks/janus_vision_layer_10/attention/output_proj/vars": [
250
+ "model_00000.weights.h5"
251
+ ],
252
+ "/layers/janus_vision_model/blocks/janus_vision_layer_10/attention/query/vars": [
253
+ "model_00000.weights.h5"
254
+ ],
255
+ "/layers/janus_vision_model/blocks/janus_vision_layer_10/attention/value/vars": [
256
+ "model_00000.weights.h5"
257
+ ],
258
+ "/layers/janus_vision_model/blocks/janus_vision_layer_10/fc1/vars": [
259
+ "model_00000.weights.h5"
260
+ ],
261
+ "/layers/janus_vision_model/blocks/janus_vision_layer_10/fc2/vars": [
262
+ "model_00000.weights.h5"
263
+ ],
264
+ "/layers/janus_vision_model/blocks/janus_vision_layer_10/layer_norm1/vars": [
265
+ "model_00000.weights.h5"
266
+ ],
267
+ "/layers/janus_vision_model/blocks/janus_vision_layer_10/layer_norm2/vars": [
268
+ "model_00000.weights.h5"
269
+ ],
270
+ "/layers/janus_vision_model/blocks/janus_vision_layer_11/attention/key/vars": [
271
+ "model_00000.weights.h5"
272
+ ],
273
+ "/layers/janus_vision_model/blocks/janus_vision_layer_11/attention/output_proj/vars": [
274
+ "model_00000.weights.h5"
275
+ ],
276
+ "/layers/janus_vision_model/blocks/janus_vision_layer_11/attention/query/vars": [
277
+ "model_00000.weights.h5"
278
+ ],
279
+ "/layers/janus_vision_model/blocks/janus_vision_layer_11/attention/value/vars": [
280
+ "model_00000.weights.h5"
281
+ ],
282
+ "/layers/janus_vision_model/blocks/janus_vision_layer_11/fc1/vars": [
283
+ "model_00000.weights.h5"
284
+ ],
285
+ "/layers/janus_vision_model/blocks/janus_vision_layer_11/fc2/vars": [
286
+ "model_00000.weights.h5"
287
+ ],
288
+ "/layers/janus_vision_model/blocks/janus_vision_layer_11/layer_norm1/vars": [
289
+ "model_00000.weights.h5"
290
+ ],
291
+ "/layers/janus_vision_model/blocks/janus_vision_layer_11/layer_norm2/vars": [
292
+ "model_00000.weights.h5"
293
+ ],
294
+ "/layers/janus_vision_model/blocks/janus_vision_layer_12/attention/key/vars": [
295
+ "model_00000.weights.h5"
296
+ ],
297
+ "/layers/janus_vision_model/blocks/janus_vision_layer_12/attention/output_proj/vars": [
298
+ "model_00000.weights.h5"
299
+ ],
300
+ "/layers/janus_vision_model/blocks/janus_vision_layer_12/attention/query/vars": [
301
+ "model_00000.weights.h5"
302
+ ],
303
+ "/layers/janus_vision_model/blocks/janus_vision_layer_12/attention/value/vars": [
304
+ "model_00000.weights.h5"
305
+ ],
306
+ "/layers/janus_vision_model/blocks/janus_vision_layer_12/fc1/vars": [
307
+ "model_00000.weights.h5"
308
+ ],
309
+ "/layers/janus_vision_model/blocks/janus_vision_layer_12/fc2/vars": [
310
+ "model_00000.weights.h5"
311
+ ],
312
+ "/layers/janus_vision_model/blocks/janus_vision_layer_12/layer_norm1/vars": [
313
+ "model_00000.weights.h5"
314
+ ],
315
+ "/layers/janus_vision_model/blocks/janus_vision_layer_12/layer_norm2/vars": [
316
+ "model_00000.weights.h5"
317
+ ],
318
+ "/layers/janus_vision_model/blocks/janus_vision_layer_13/attention/key/vars": [
319
+ "model_00000.weights.h5"
320
+ ],
321
+ "/layers/janus_vision_model/blocks/janus_vision_layer_13/attention/output_proj/vars": [
322
+ "model_00000.weights.h5"
323
+ ],
324
+ "/layers/janus_vision_model/blocks/janus_vision_layer_13/attention/query/vars": [
325
+ "model_00000.weights.h5"
326
+ ],
327
+ "/layers/janus_vision_model/blocks/janus_vision_layer_13/attention/value/vars": [
328
+ "model_00000.weights.h5"
329
+ ],
330
+ "/layers/janus_vision_model/blocks/janus_vision_layer_13/fc1/vars": [
331
+ "model_00000.weights.h5"
332
+ ],
333
+ "/layers/janus_vision_model/blocks/janus_vision_layer_13/fc2/vars": [
334
+ "model_00000.weights.h5"
335
+ ],
336
+ "/layers/janus_vision_model/blocks/janus_vision_layer_13/layer_norm1/vars": [
337
+ "model_00000.weights.h5"
338
+ ],
339
+ "/layers/janus_vision_model/blocks/janus_vision_layer_13/layer_norm2/vars": [
340
+ "model_00000.weights.h5"
341
+ ],
342
+ "/layers/janus_vision_model/blocks/janus_vision_layer_14/attention/key/vars": [
343
+ "model_00000.weights.h5"
344
+ ],
345
+ "/layers/janus_vision_model/blocks/janus_vision_layer_14/attention/output_proj/vars": [
346
+ "model_00000.weights.h5"
347
+ ],
348
+ "/layers/janus_vision_model/blocks/janus_vision_layer_14/attention/query/vars": [
349
+ "model_00000.weights.h5"
350
+ ],
351
+ "/layers/janus_vision_model/blocks/janus_vision_layer_14/attention/value/vars": [
352
+ "model_00000.weights.h5"
353
+ ],
354
+ "/layers/janus_vision_model/blocks/janus_vision_layer_14/fc1/vars": [
355
+ "model_00000.weights.h5"
356
+ ],
357
+ "/layers/janus_vision_model/blocks/janus_vision_layer_14/fc2/vars": [
358
+ "model_00000.weights.h5"
359
+ ],
360
+ "/layers/janus_vision_model/blocks/janus_vision_layer_14/layer_norm1/vars": [
361
+ "model_00000.weights.h5"
362
+ ],
363
+ "/layers/janus_vision_model/blocks/janus_vision_layer_14/layer_norm2/vars": [
364
+ "model_00000.weights.h5"
365
+ ],
366
+ "/layers/janus_vision_model/blocks/janus_vision_layer_15/attention/key/vars": [
367
+ "model_00000.weights.h5"
368
+ ],
369
+ "/layers/janus_vision_model/blocks/janus_vision_layer_15/attention/output_proj/vars": [
370
+ "model_00000.weights.h5"
371
+ ],
372
+ "/layers/janus_vision_model/blocks/janus_vision_layer_15/attention/query/vars": [
373
+ "model_00000.weights.h5"
374
+ ],
375
+ "/layers/janus_vision_model/blocks/janus_vision_layer_15/attention/value/vars": [
376
+ "model_00000.weights.h5"
377
+ ],
378
+ "/layers/janus_vision_model/blocks/janus_vision_layer_15/fc1/vars": [
379
+ "model_00000.weights.h5"
380
+ ],
381
+ "/layers/janus_vision_model/blocks/janus_vision_layer_15/fc2/vars": [
382
+ "model_00000.weights.h5"
383
+ ],
384
+ "/layers/janus_vision_model/blocks/janus_vision_layer_15/layer_norm1/vars": [
385
+ "model_00000.weights.h5"
386
+ ],
387
+ "/layers/janus_vision_model/blocks/janus_vision_layer_15/layer_norm2/vars": [
388
+ "model_00000.weights.h5"
389
+ ],
390
+ "/layers/janus_vision_model/blocks/janus_vision_layer_16/attention/key/vars": [
391
+ "model_00000.weights.h5"
392
+ ],
393
+ "/layers/janus_vision_model/blocks/janus_vision_layer_16/attention/output_proj/vars": [
394
+ "model_00000.weights.h5"
395
+ ],
396
+ "/layers/janus_vision_model/blocks/janus_vision_layer_16/attention/query/vars": [
397
+ "model_00000.weights.h5"
398
+ ],
399
+ "/layers/janus_vision_model/blocks/janus_vision_layer_16/attention/value/vars": [
400
+ "model_00000.weights.h5"
401
+ ],
402
+ "/layers/janus_vision_model/blocks/janus_vision_layer_16/fc1/vars": [
403
+ "model_00000.weights.h5"
404
+ ],
405
+ "/layers/janus_vision_model/blocks/janus_vision_layer_16/fc2/vars": [
406
+ "model_00000.weights.h5"
407
+ ],
408
+ "/layers/janus_vision_model/blocks/janus_vision_layer_16/layer_norm1/vars": [
409
+ "model_00000.weights.h5"
410
+ ],
411
+ "/layers/janus_vision_model/blocks/janus_vision_layer_16/layer_norm2/vars": [
412
+ "model_00000.weights.h5"
413
+ ],
414
+ "/layers/janus_vision_model/blocks/janus_vision_layer_17/attention/key/vars": [
415
+ "model_00000.weights.h5"
416
+ ],
417
+ "/layers/janus_vision_model/blocks/janus_vision_layer_17/attention/output_proj/vars": [
418
+ "model_00000.weights.h5"
419
+ ],
420
+ "/layers/janus_vision_model/blocks/janus_vision_layer_17/attention/query/vars": [
421
+ "model_00000.weights.h5"
422
+ ],
423
+ "/layers/janus_vision_model/blocks/janus_vision_layer_17/attention/value/vars": [
424
+ "model_00000.weights.h5"
425
+ ],
426
+ "/layers/janus_vision_model/blocks/janus_vision_layer_17/fc1/vars": [
427
+ "model_00000.weights.h5"
428
+ ],
429
+ "/layers/janus_vision_model/blocks/janus_vision_layer_17/fc2/vars": [
430
+ "model_00000.weights.h5"
431
+ ],
432
+ "/layers/janus_vision_model/blocks/janus_vision_layer_17/layer_norm1/vars": [
433
+ "model_00000.weights.h5"
434
+ ],
435
+ "/layers/janus_vision_model/blocks/janus_vision_layer_17/layer_norm2/vars": [
436
+ "model_00000.weights.h5"
437
+ ],
438
+ "/layers/janus_vision_model/blocks/janus_vision_layer_18/attention/key/vars": [
439
+ "model_00000.weights.h5"
440
+ ],
441
+ "/layers/janus_vision_model/blocks/janus_vision_layer_18/attention/output_proj/vars": [
442
+ "model_00000.weights.h5"
443
+ ],
444
+ "/layers/janus_vision_model/blocks/janus_vision_layer_18/attention/query/vars": [
445
+ "model_00000.weights.h5"
446
+ ],
447
+ "/layers/janus_vision_model/blocks/janus_vision_layer_18/attention/value/vars": [
448
+ "model_00000.weights.h5"
449
+ ],
450
+ "/layers/janus_vision_model/blocks/janus_vision_layer_18/fc1/vars": [
451
+ "model_00000.weights.h5"
452
+ ],
453
+ "/layers/janus_vision_model/blocks/janus_vision_layer_18/fc2/vars": [
454
+ "model_00000.weights.h5"
455
+ ],
456
+ "/layers/janus_vision_model/blocks/janus_vision_layer_18/layer_norm1/vars": [
457
+ "model_00000.weights.h5"
458
+ ],
459
+ "/layers/janus_vision_model/blocks/janus_vision_layer_18/layer_norm2/vars": [
460
+ "model_00000.weights.h5"
461
+ ],
462
+ "/layers/janus_vision_model/blocks/janus_vision_layer_19/attention/key/vars": [
463
+ "model_00000.weights.h5"
464
+ ],
465
+ "/layers/janus_vision_model/blocks/janus_vision_layer_19/attention/output_proj/vars": [
466
+ "model_00000.weights.h5"
467
+ ],
468
+ "/layers/janus_vision_model/blocks/janus_vision_layer_19/attention/query/vars": [
469
+ "model_00000.weights.h5"
470
+ ],
471
+ "/layers/janus_vision_model/blocks/janus_vision_layer_19/attention/value/vars": [
472
+ "model_00000.weights.h5"
473
+ ],
474
+ "/layers/janus_vision_model/blocks/janus_vision_layer_19/fc1/vars": [
475
+ "model_00000.weights.h5"
476
+ ],
477
+ "/layers/janus_vision_model/blocks/janus_vision_layer_19/fc2/vars": [
478
+ "model_00000.weights.h5"
479
+ ],
480
+ "/layers/janus_vision_model/blocks/janus_vision_layer_19/layer_norm1/vars": [
481
+ "model_00000.weights.h5"
482
+ ],
483
+ "/layers/janus_vision_model/blocks/janus_vision_layer_19/layer_norm2/vars": [
484
+ "model_00000.weights.h5"
485
+ ],
486
+ "/layers/janus_vision_model/blocks/janus_vision_layer_20/attention/key/vars": [
487
+ "model_00000.weights.h5"
488
+ ],
489
+ "/layers/janus_vision_model/blocks/janus_vision_layer_20/attention/output_proj/vars": [
490
+ "model_00000.weights.h5"
491
+ ],
492
+ "/layers/janus_vision_model/blocks/janus_vision_layer_20/attention/query/vars": [
493
+ "model_00000.weights.h5"
494
+ ],
495
+ "/layers/janus_vision_model/blocks/janus_vision_layer_20/attention/value/vars": [
496
+ "model_00000.weights.h5"
497
+ ],
498
+ "/layers/janus_vision_model/blocks/janus_vision_layer_20/fc1/vars": [
499
+ "model_00000.weights.h5"
500
+ ],
501
+ "/layers/janus_vision_model/blocks/janus_vision_layer_20/fc2/vars": [
502
+ "model_00000.weights.h5"
503
+ ],
504
+ "/layers/janus_vision_model/blocks/janus_vision_layer_20/layer_norm1/vars": [
505
+ "model_00000.weights.h5"
506
+ ],
507
+ "/layers/janus_vision_model/blocks/janus_vision_layer_20/layer_norm2/vars": [
508
+ "model_00000.weights.h5"
509
+ ],
510
+ "/layers/janus_vision_model/blocks/janus_vision_layer_21/attention/key/vars": [
511
+ "model_00000.weights.h5"
512
+ ],
513
+ "/layers/janus_vision_model/blocks/janus_vision_layer_21/attention/output_proj/vars": [
514
+ "model_00000.weights.h5"
515
+ ],
516
+ "/layers/janus_vision_model/blocks/janus_vision_layer_21/attention/query/vars": [
517
+ "model_00000.weights.h5"
518
+ ],
519
+ "/layers/janus_vision_model/blocks/janus_vision_layer_21/attention/value/vars": [
520
+ "model_00000.weights.h5"
521
+ ],
522
+ "/layers/janus_vision_model/blocks/janus_vision_layer_21/fc1/vars": [
523
+ "model_00000.weights.h5"
524
+ ],
525
+ "/layers/janus_vision_model/blocks/janus_vision_layer_21/fc2/vars": [
526
+ "model_00000.weights.h5"
527
+ ],
528
+ "/layers/janus_vision_model/blocks/janus_vision_layer_21/layer_norm1/vars": [
529
+ "model_00000.weights.h5"
530
+ ],
531
+ "/layers/janus_vision_model/blocks/janus_vision_layer_21/layer_norm2/vars": [
532
+ "model_00000.weights.h5"
533
+ ],
534
+ "/layers/janus_vision_model/blocks/janus_vision_layer_22/attention/key/vars": [
535
+ "model_00000.weights.h5"
536
+ ],
537
+ "/layers/janus_vision_model/blocks/janus_vision_layer_22/attention/output_proj/vars": [
538
+ "model_00000.weights.h5"
539
+ ],
540
+ "/layers/janus_vision_model/blocks/janus_vision_layer_22/attention/query/vars": [
541
+ "model_00000.weights.h5"
542
+ ],
543
+ "/layers/janus_vision_model/blocks/janus_vision_layer_22/attention/value/vars": [
544
+ "model_00000.weights.h5"
545
+ ],
546
+ "/layers/janus_vision_model/blocks/janus_vision_layer_22/fc1/vars": [
547
+ "model_00000.weights.h5"
548
+ ],
549
+ "/layers/janus_vision_model/blocks/janus_vision_layer_22/fc2/vars": [
550
+ "model_00000.weights.h5"
551
+ ],
552
+ "/layers/janus_vision_model/blocks/janus_vision_layer_22/layer_norm1/vars": [
553
+ "model_00000.weights.h5"
554
+ ],
555
+ "/layers/janus_vision_model/blocks/janus_vision_layer_22/layer_norm2/vars": [
556
+ "model_00000.weights.h5"
557
+ ],
558
+ "/layers/janus_vision_model/blocks/janus_vision_layer_23/attention/key/vars": [
559
+ "model_00000.weights.h5"
560
+ ],
561
+ "/layers/janus_vision_model/blocks/janus_vision_layer_23/attention/output_proj/vars": [
562
+ "model_00000.weights.h5"
563
+ ],
564
+ "/layers/janus_vision_model/blocks/janus_vision_layer_23/attention/query/vars": [
565
+ "model_00000.weights.h5"
566
+ ],
567
+ "/layers/janus_vision_model/blocks/janus_vision_layer_23/attention/value/vars": [
568
+ "model_00000.weights.h5"
569
+ ],
570
+ "/layers/janus_vision_model/blocks/janus_vision_layer_23/fc1/vars": [
571
+ "model_00000.weights.h5"
572
+ ],
573
+ "/layers/janus_vision_model/blocks/janus_vision_layer_23/fc2/vars": [
574
+ "model_00000.weights.h5"
575
+ ],
576
+ "/layers/janus_vision_model/blocks/janus_vision_layer_23/layer_norm1/vars": [
577
+ "model_00000.weights.h5"
578
+ ],
579
+ "/layers/janus_vision_model/blocks/janus_vision_layer_23/layer_norm2/vars": [
580
+ "model_00000.weights.h5"
581
+ ],
582
+ "/layers/janus_vision_model/patch_embed/vars": [
583
+ "model_00000.weights.h5"
584
+ ],
585
+ "/layers/janus_vision_model/position_embedding/vars": "model_00000.weights.h5",
586
+ "/layers/janus_vision_model/post_layernorm/vars": [
587
+ "model_00000.weights.h5"
588
+ ],
589
+ "/layers/dense/vars": [
590
+ "model_00000.weights.h5"
591
+ ],
592
+ "/layers/dense_1/vars": [
593
+ "model_00000.weights.h5"
594
+ ],
595
+ "/layers/embedding/vars": "model_00000.weights.h5",
596
+ "/layers/janus_text_decoder_layer/attention/key/vars": "model_00000.weights.h5",
597
+ "/layers/janus_text_decoder_layer/attention/output_proj/vars": "model_00000.weights.h5",
598
+ "/layers/janus_text_decoder_layer/attention/query/vars": "model_00000.weights.h5",
599
+ "/layers/janus_text_decoder_layer/attention/value/vars": "model_00000.weights.h5",
600
+ "/layers/janus_text_decoder_layer/attention_norm/vars": "model_00000.weights.h5",
601
+ "/layers/janus_text_decoder_layer/mlp/down/vars": "model_00000.weights.h5",
602
+ "/layers/janus_text_decoder_layer/mlp/gate/vars": "model_00000.weights.h5",
603
+ "/layers/janus_text_decoder_layer/mlp/up/vars": "model_00000.weights.h5",
604
+ "/layers/janus_text_decoder_layer/mlp_norm/vars": "model_00000.weights.h5",
605
+ "/layers/janus_text_decoder_layer_1/attention/key/vars": "model_00000.weights.h5",
606
+ "/layers/janus_text_decoder_layer_1/attention/output_proj/vars": "model_00000.weights.h5",
607
+ "/layers/janus_text_decoder_layer_1/attention/query/vars": "model_00000.weights.h5",
608
+ "/layers/janus_text_decoder_layer_1/attention/value/vars": "model_00000.weights.h5",
609
+ "/layers/janus_text_decoder_layer_1/attention_norm/vars": "model_00000.weights.h5",
610
+ "/layers/janus_text_decoder_layer_1/mlp/down/vars": "model_00000.weights.h5",
611
+ "/layers/janus_text_decoder_layer_1/mlp/gate/vars": "model_00000.weights.h5",
612
+ "/layers/janus_text_decoder_layer_1/mlp/up/vars": "model_00000.weights.h5",
613
+ "/layers/janus_text_decoder_layer_1/mlp_norm/vars": "model_00000.weights.h5",
614
+ "/layers/janus_text_decoder_layer_2/attention/key/vars": "model_00000.weights.h5",
615
+ "/layers/janus_text_decoder_layer_2/attention/output_proj/vars": "model_00000.weights.h5",
616
+ "/layers/janus_text_decoder_layer_2/attention/query/vars": "model_00000.weights.h5",
617
+ "/layers/janus_text_decoder_layer_2/attention/value/vars": "model_00000.weights.h5",
618
+ "/layers/janus_text_decoder_layer_2/attention_norm/vars": "model_00000.weights.h5",
619
+ "/layers/janus_text_decoder_layer_2/mlp/down/vars": "model_00000.weights.h5",
620
+ "/layers/janus_text_decoder_layer_2/mlp/gate/vars": "model_00000.weights.h5",
621
+ "/layers/janus_text_decoder_layer_2/mlp/up/vars": "model_00000.weights.h5",
622
+ "/layers/janus_text_decoder_layer_2/mlp_norm/vars": "model_00000.weights.h5",
623
+ "/layers/janus_text_decoder_layer_3/attention/key/vars": "model_00000.weights.h5",
624
+ "/layers/janus_text_decoder_layer_3/attention/output_proj/vars": "model_00000.weights.h5",
625
+ "/layers/janus_text_decoder_layer_3/attention/query/vars": "model_00000.weights.h5",
626
+ "/layers/janus_text_decoder_layer_3/attention/value/vars": "model_00000.weights.h5",
627
+ "/layers/janus_text_decoder_layer_3/attention_norm/vars": "model_00000.weights.h5",
628
+ "/layers/janus_text_decoder_layer_3/mlp/down/vars": "model_00000.weights.h5",
629
+ "/layers/janus_text_decoder_layer_3/mlp/gate/vars": "model_00000.weights.h5",
630
+ "/layers/janus_text_decoder_layer_3/mlp/up/vars": "model_00000.weights.h5",
631
+ "/layers/janus_text_decoder_layer_3/mlp_norm/vars": "model_00000.weights.h5",
632
+ "/layers/janus_text_decoder_layer_4/attention/key/vars": "model_00000.weights.h5",
633
+ "/layers/janus_text_decoder_layer_4/attention/output_proj/vars": "model_00000.weights.h5",
634
+ "/layers/janus_text_decoder_layer_4/attention/query/vars": "model_00000.weights.h5",
635
+ "/layers/janus_text_decoder_layer_4/attention/value/vars": "model_00000.weights.h5",
636
+ "/layers/janus_text_decoder_layer_4/attention_norm/vars": "model_00000.weights.h5",
637
+ "/layers/janus_text_decoder_layer_4/mlp/down/vars": "model_00000.weights.h5",
638
+ "/layers/janus_text_decoder_layer_4/mlp/gate/vars": "model_00000.weights.h5",
639
+ "/layers/janus_text_decoder_layer_4/mlp/up/vars": "model_00000.weights.h5",
640
+ "/layers/janus_text_decoder_layer_4/mlp_norm/vars": "model_00000.weights.h5",
641
+ "/layers/janus_text_decoder_layer_5/attention/key/vars": "model_00000.weights.h5",
642
+ "/layers/janus_text_decoder_layer_5/attention/output_proj/vars": "model_00000.weights.h5",
643
+ "/layers/janus_text_decoder_layer_5/attention/query/vars": "model_00000.weights.h5",
644
+ "/layers/janus_text_decoder_layer_5/attention/value/vars": "model_00000.weights.h5",
645
+ "/layers/janus_text_decoder_layer_5/attention_norm/vars": "model_00000.weights.h5",
646
+ "/layers/janus_text_decoder_layer_5/mlp/down/vars": "model_00000.weights.h5",
647
+ "/layers/janus_text_decoder_layer_5/mlp/gate/vars": "model_00000.weights.h5",
648
+ "/layers/janus_text_decoder_layer_5/mlp/up/vars": "model_00000.weights.h5",
649
+ "/layers/janus_text_decoder_layer_5/mlp_norm/vars": "model_00000.weights.h5",
650
+ "/layers/janus_text_decoder_layer_6/attention/key/vars": "model_00000.weights.h5",
651
+ "/layers/janus_text_decoder_layer_6/attention/output_proj/vars": "model_00000.weights.h5",
652
+ "/layers/janus_text_decoder_layer_6/attention/query/vars": "model_00000.weights.h5",
653
+ "/layers/janus_text_decoder_layer_6/attention/value/vars": "model_00000.weights.h5",
654
+ "/layers/janus_text_decoder_layer_6/attention_norm/vars": "model_00000.weights.h5",
655
+ "/layers/janus_text_decoder_layer_6/mlp/down/vars": "model_00000.weights.h5",
656
+ "/layers/janus_text_decoder_layer_6/mlp/gate/vars": "model_00000.weights.h5",
657
+ "/layers/janus_text_decoder_layer_6/mlp/up/vars": "model_00000.weights.h5",
658
+ "/layers/janus_text_decoder_layer_6/mlp_norm/vars": "model_00000.weights.h5",
659
+ "/layers/janus_text_decoder_layer_7/attention/key/vars": "model_00000.weights.h5",
660
+ "/layers/janus_text_decoder_layer_7/attention/output_proj/vars": "model_00000.weights.h5",
661
+ "/layers/janus_text_decoder_layer_7/attention/query/vars": "model_00000.weights.h5",
662
+ "/layers/janus_text_decoder_layer_7/attention/value/vars": "model_00000.weights.h5",
663
+ "/layers/janus_text_decoder_layer_7/attention_norm/vars": "model_00000.weights.h5",
664
+ "/layers/janus_text_decoder_layer_7/mlp/down/vars": "model_00000.weights.h5",
665
+ "/layers/janus_text_decoder_layer_7/mlp/gate/vars": "model_00000.weights.h5",
666
+ "/layers/janus_text_decoder_layer_7/mlp/up/vars": "model_00000.weights.h5",
667
+ "/layers/janus_text_decoder_layer_7/mlp_norm/vars": "model_00000.weights.h5",
668
+ "/layers/janus_text_decoder_layer_8/attention/key/vars": "model_00000.weights.h5",
669
+ "/layers/janus_text_decoder_layer_8/attention/output_proj/vars": "model_00000.weights.h5",
670
+ "/layers/janus_text_decoder_layer_8/attention/query/vars": "model_00000.weights.h5",
671
+ "/layers/janus_text_decoder_layer_8/attention/value/vars": "model_00000.weights.h5",
672
+ "/layers/janus_text_decoder_layer_8/attention_norm/vars": "model_00000.weights.h5",
673
+ "/layers/janus_text_decoder_layer_8/mlp/down/vars": "model_00000.weights.h5",
674
+ "/layers/janus_text_decoder_layer_8/mlp/gate/vars": "model_00000.weights.h5",
675
+ "/layers/janus_text_decoder_layer_8/mlp/up/vars": "model_00000.weights.h5",
676
+ "/layers/janus_text_decoder_layer_8/mlp_norm/vars": "model_00000.weights.h5",
677
+ "/layers/janus_text_decoder_layer_9/attention/key/vars": "model_00000.weights.h5",
678
+ "/layers/janus_text_decoder_layer_9/attention/output_proj/vars": "model_00000.weights.h5",
679
+ "/layers/janus_text_decoder_layer_9/attention/query/vars": "model_00000.weights.h5",
680
+ "/layers/janus_text_decoder_layer_9/attention/value/vars": "model_00000.weights.h5",
681
+ "/layers/janus_text_decoder_layer_9/attention_norm/vars": "model_00000.weights.h5",
682
+ "/layers/janus_text_decoder_layer_9/mlp/down/vars": "model_00000.weights.h5",
683
+ "/layers/janus_text_decoder_layer_9/mlp/gate/vars": "model_00001.weights.h5",
684
+ "/layers/janus_text_decoder_layer_9/mlp/up/vars": "model_00001.weights.h5",
685
+ "/layers/janus_text_decoder_layer_9/mlp_norm/vars": "model_00001.weights.h5",
686
+ "/layers/janus_text_decoder_layer_10/attention/key/vars": "model_00001.weights.h5",
687
+ "/layers/janus_text_decoder_layer_10/attention/output_proj/vars": "model_00001.weights.h5",
688
+ "/layers/janus_text_decoder_layer_10/attention/query/vars": "model_00001.weights.h5",
689
+ "/layers/janus_text_decoder_layer_10/attention/value/vars": "model_00001.weights.h5",
690
+ "/layers/janus_text_decoder_layer_10/attention_norm/vars": "model_00001.weights.h5",
691
+ "/layers/janus_text_decoder_layer_10/mlp/down/vars": "model_00001.weights.h5",
692
+ "/layers/janus_text_decoder_layer_10/mlp/gate/vars": "model_00001.weights.h5",
693
+ "/layers/janus_text_decoder_layer_10/mlp/up/vars": "model_00001.weights.h5",
694
+ "/layers/janus_text_decoder_layer_10/mlp_norm/vars": "model_00001.weights.h5",
695
+ "/layers/janus_text_decoder_layer_11/attention/key/vars": "model_00001.weights.h5",
696
+ "/layers/janus_text_decoder_layer_11/attention/output_proj/vars": "model_00001.weights.h5",
697
+ "/layers/janus_text_decoder_layer_11/attention/query/vars": "model_00001.weights.h5",
698
+ "/layers/janus_text_decoder_layer_11/attention/value/vars": "model_00001.weights.h5",
699
+ "/layers/janus_text_decoder_layer_11/attention_norm/vars": "model_00001.weights.h5",
700
+ "/layers/janus_text_decoder_layer_11/mlp/down/vars": "model_00001.weights.h5",
701
+ "/layers/janus_text_decoder_layer_11/mlp/gate/vars": "model_00001.weights.h5",
702
+ "/layers/janus_text_decoder_layer_11/mlp/up/vars": "model_00001.weights.h5",
703
+ "/layers/janus_text_decoder_layer_11/mlp_norm/vars": "model_00001.weights.h5",
704
+ "/layers/janus_text_decoder_layer_12/attention/key/vars": "model_00001.weights.h5",
705
+ "/layers/janus_text_decoder_layer_12/attention/output_proj/vars": "model_00001.weights.h5",
706
+ "/layers/janus_text_decoder_layer_12/attention/query/vars": "model_00001.weights.h5",
707
+ "/layers/janus_text_decoder_layer_12/attention/value/vars": "model_00001.weights.h5",
708
+ "/layers/janus_text_decoder_layer_12/attention_norm/vars": "model_00001.weights.h5",
709
+ "/layers/janus_text_decoder_layer_12/mlp/down/vars": "model_00001.weights.h5",
710
+ "/layers/janus_text_decoder_layer_12/mlp/gate/vars": "model_00001.weights.h5",
711
+ "/layers/janus_text_decoder_layer_12/mlp/up/vars": "model_00001.weights.h5",
712
+ "/layers/janus_text_decoder_layer_12/mlp_norm/vars": "model_00001.weights.h5",
713
+ "/layers/janus_text_decoder_layer_13/attention/key/vars": "model_00001.weights.h5",
714
+ "/layers/janus_text_decoder_layer_13/attention/output_proj/vars": "model_00001.weights.h5",
715
+ "/layers/janus_text_decoder_layer_13/attention/query/vars": "model_00001.weights.h5",
716
+ "/layers/janus_text_decoder_layer_13/attention/value/vars": "model_00001.weights.h5",
717
+ "/layers/janus_text_decoder_layer_13/attention_norm/vars": "model_00001.weights.h5",
718
+ "/layers/janus_text_decoder_layer_13/mlp/down/vars": "model_00001.weights.h5",
719
+ "/layers/janus_text_decoder_layer_13/mlp/gate/vars": "model_00001.weights.h5",
720
+ "/layers/janus_text_decoder_layer_13/mlp/up/vars": "model_00001.weights.h5",
721
+ "/layers/janus_text_decoder_layer_13/mlp_norm/vars": "model_00001.weights.h5",
722
+ "/layers/janus_text_decoder_layer_14/attention/key/vars": "model_00001.weights.h5",
723
+ "/layers/janus_text_decoder_layer_14/attention/output_proj/vars": "model_00001.weights.h5",
724
+ "/layers/janus_text_decoder_layer_14/attention/query/vars": "model_00001.weights.h5",
725
+ "/layers/janus_text_decoder_layer_14/attention/value/vars": "model_00001.weights.h5",
726
+ "/layers/janus_text_decoder_layer_14/attention_norm/vars": "model_00001.weights.h5",
727
+ "/layers/janus_text_decoder_layer_14/mlp/down/vars": "model_00001.weights.h5",
728
+ "/layers/janus_text_decoder_layer_14/mlp/gate/vars": "model_00001.weights.h5",
729
+ "/layers/janus_text_decoder_layer_14/mlp/up/vars": "model_00001.weights.h5",
730
+ "/layers/janus_text_decoder_layer_14/mlp_norm/vars": "model_00001.weights.h5",
731
+ "/layers/janus_text_decoder_layer_15/attention/key/vars": "model_00001.weights.h5",
732
+ "/layers/janus_text_decoder_layer_15/attention/output_proj/vars": "model_00001.weights.h5",
733
+ "/layers/janus_text_decoder_layer_15/attention/query/vars": "model_00001.weights.h5",
734
+ "/layers/janus_text_decoder_layer_15/attention/value/vars": "model_00001.weights.h5",
735
+ "/layers/janus_text_decoder_layer_15/attention_norm/vars": "model_00001.weights.h5",
736
+ "/layers/janus_text_decoder_layer_15/mlp/down/vars": "model_00001.weights.h5",
737
+ "/layers/janus_text_decoder_layer_15/mlp/gate/vars": "model_00001.weights.h5",
738
+ "/layers/janus_text_decoder_layer_15/mlp/up/vars": "model_00001.weights.h5",
739
+ "/layers/janus_text_decoder_layer_15/mlp_norm/vars": "model_00001.weights.h5",
740
+ "/layers/janus_text_decoder_layer_16/attention/key/vars": "model_00001.weights.h5",
741
+ "/layers/janus_text_decoder_layer_16/attention/output_proj/vars": "model_00001.weights.h5",
742
+ "/layers/janus_text_decoder_layer_16/attention/query/vars": "model_00001.weights.h5",
743
+ "/layers/janus_text_decoder_layer_16/attention/value/vars": "model_00001.weights.h5",
744
+ "/layers/janus_text_decoder_layer_16/attention_norm/vars": "model_00001.weights.h5",
745
+ "/layers/janus_text_decoder_layer_16/mlp/down/vars": "model_00001.weights.h5",
746
+ "/layers/janus_text_decoder_layer_16/mlp/gate/vars": "model_00001.weights.h5",
747
+ "/layers/janus_text_decoder_layer_16/mlp/up/vars": "model_00001.weights.h5",
748
+ "/layers/janus_text_decoder_layer_16/mlp_norm/vars": "model_00001.weights.h5",
749
+ "/layers/janus_text_decoder_layer_17/attention/key/vars": "model_00001.weights.h5",
750
+ "/layers/janus_text_decoder_layer_17/attention/output_proj/vars": "model_00001.weights.h5",
751
+ "/layers/janus_text_decoder_layer_17/attention/query/vars": "model_00001.weights.h5",
752
+ "/layers/janus_text_decoder_layer_17/attention/value/vars": "model_00001.weights.h5",
753
+ "/layers/janus_text_decoder_layer_17/attention_norm/vars": "model_00001.weights.h5",
754
+ "/layers/janus_text_decoder_layer_17/mlp/down/vars": "model_00001.weights.h5",
755
+ "/layers/janus_text_decoder_layer_17/mlp/gate/vars": "model_00001.weights.h5",
756
+ "/layers/janus_text_decoder_layer_17/mlp/up/vars": "model_00001.weights.h5",
757
+ "/layers/janus_text_decoder_layer_17/mlp_norm/vars": "model_00001.weights.h5",
758
+ "/layers/janus_text_decoder_layer_18/attention/key/vars": "model_00001.weights.h5",
759
+ "/layers/janus_text_decoder_layer_18/attention/output_proj/vars": "model_00001.weights.h5",
760
+ "/layers/janus_text_decoder_layer_18/attention/query/vars": "model_00001.weights.h5",
761
+ "/layers/janus_text_decoder_layer_18/attention/value/vars": "model_00001.weights.h5",
762
+ "/layers/janus_text_decoder_layer_18/attention_norm/vars": "model_00001.weights.h5",
763
+ "/layers/janus_text_decoder_layer_18/mlp/down/vars": "model_00001.weights.h5",
764
+ "/layers/janus_text_decoder_layer_18/mlp/gate/vars": "model_00001.weights.h5",
765
+ "/layers/janus_text_decoder_layer_18/mlp/up/vars": "model_00001.weights.h5",
766
+ "/layers/janus_text_decoder_layer_18/mlp_norm/vars": "model_00001.weights.h5",
767
+ "/layers/janus_text_decoder_layer_19/attention/key/vars": "model_00001.weights.h5",
768
+ "/layers/janus_text_decoder_layer_19/attention/output_proj/vars": "model_00001.weights.h5",
769
+ "/layers/janus_text_decoder_layer_19/attention/query/vars": "model_00001.weights.h5",
770
+ "/layers/janus_text_decoder_layer_19/attention/value/vars": "model_00001.weights.h5",
771
+ "/layers/janus_text_decoder_layer_19/attention_norm/vars": "model_00001.weights.h5",
772
+ "/layers/janus_text_decoder_layer_19/mlp/down/vars": "model_00001.weights.h5",
773
+ "/layers/janus_text_decoder_layer_19/mlp/gate/vars": "model_00001.weights.h5",
774
+ "/layers/janus_text_decoder_layer_19/mlp/up/vars": "model_00001.weights.h5",
775
+ "/layers/janus_text_decoder_layer_19/mlp_norm/vars": "model_00001.weights.h5",
776
+ "/layers/janus_text_decoder_layer_20/attention/key/vars": "model_00001.weights.h5",
777
+ "/layers/janus_text_decoder_layer_20/attention/output_proj/vars": "model_00001.weights.h5",
778
+ "/layers/janus_text_decoder_layer_20/attention/query/vars": "model_00001.weights.h5",
779
+ "/layers/janus_text_decoder_layer_20/attention/value/vars": "model_00001.weights.h5",
780
+ "/layers/janus_text_decoder_layer_20/attention_norm/vars": "model_00001.weights.h5",
781
+ "/layers/janus_text_decoder_layer_20/mlp/down/vars": "model_00001.weights.h5",
782
+ "/layers/janus_text_decoder_layer_20/mlp/gate/vars": "model_00001.weights.h5",
783
+ "/layers/janus_text_decoder_layer_20/mlp/up/vars": "model_00001.weights.h5",
784
+ "/layers/janus_text_decoder_layer_20/mlp_norm/vars": "model_00001.weights.h5",
785
+ "/layers/janus_text_decoder_layer_21/attention/key/vars": "model_00001.weights.h5",
786
+ "/layers/janus_text_decoder_layer_21/attention/output_proj/vars": "model_00001.weights.h5",
787
+ "/layers/janus_text_decoder_layer_21/attention/query/vars": "model_00001.weights.h5",
788
+ "/layers/janus_text_decoder_layer_21/attention/value/vars": "model_00001.weights.h5",
789
+ "/layers/janus_text_decoder_layer_21/attention_norm/vars": "model_00001.weights.h5",
790
+ "/layers/janus_text_decoder_layer_21/mlp/down/vars": "model_00001.weights.h5",
791
+ "/layers/janus_text_decoder_layer_21/mlp/gate/vars": "model_00001.weights.h5",
792
+ "/layers/janus_text_decoder_layer_21/mlp/up/vars": "model_00001.weights.h5",
793
+ "/layers/janus_text_decoder_layer_21/mlp_norm/vars": "model_00001.weights.h5",
794
+ "/layers/janus_text_decoder_layer_22/attention/key/vars": "model_00001.weights.h5",
795
+ "/layers/janus_text_decoder_layer_22/attention/output_proj/vars": "model_00001.weights.h5",
796
+ "/layers/janus_text_decoder_layer_22/attention/query/vars": "model_00001.weights.h5",
797
+ "/layers/janus_text_decoder_layer_22/attention/value/vars": "model_00001.weights.h5",
798
+ "/layers/janus_text_decoder_layer_22/attention_norm/vars": "model_00001.weights.h5",
799
+ "/layers/janus_text_decoder_layer_22/mlp/down/vars": "model_00001.weights.h5",
800
+ "/layers/janus_text_decoder_layer_22/mlp/gate/vars": "model_00001.weights.h5",
801
+ "/layers/janus_text_decoder_layer_22/mlp/up/vars": "model_00002.weights.h5",
802
+ "/layers/janus_text_decoder_layer_22/mlp_norm/vars": "model_00002.weights.h5",
803
+ "/layers/janus_text_decoder_layer_23/attention/key/vars": "model_00002.weights.h5",
804
+ "/layers/janus_text_decoder_layer_23/attention/output_proj/vars": "model_00002.weights.h5",
805
+ "/layers/janus_text_decoder_layer_23/attention/query/vars": "model_00002.weights.h5",
806
+ "/layers/janus_text_decoder_layer_23/attention/value/vars": "model_00002.weights.h5",
807
+ "/layers/janus_text_decoder_layer_23/attention_norm/vars": "model_00002.weights.h5",
808
+ "/layers/janus_text_decoder_layer_23/mlp/down/vars": "model_00002.weights.h5",
809
+ "/layers/janus_text_decoder_layer_23/mlp/gate/vars": "model_00002.weights.h5",
810
+ "/layers/janus_text_decoder_layer_23/mlp/up/vars": "model_00002.weights.h5",
811
+ "/layers/janus_text_decoder_layer_23/mlp_norm/vars": "model_00002.weights.h5",
812
+ "/layers/janus_text_decoder_layer_24/attention/key/vars": "model_00002.weights.h5",
813
+ "/layers/janus_text_decoder_layer_24/attention/output_proj/vars": "model_00002.weights.h5",
814
+ "/layers/janus_text_decoder_layer_24/attention/query/vars": "model_00002.weights.h5",
815
+ "/layers/janus_text_decoder_layer_24/attention/value/vars": "model_00002.weights.h5",
816
+ "/layers/janus_text_decoder_layer_24/attention_norm/vars": "model_00002.weights.h5",
817
+ "/layers/janus_text_decoder_layer_24/mlp/down/vars": "model_00002.weights.h5",
818
+ "/layers/janus_text_decoder_layer_24/mlp/gate/vars": "model_00002.weights.h5",
819
+ "/layers/janus_text_decoder_layer_24/mlp/up/vars": "model_00002.weights.h5",
820
+ "/layers/janus_text_decoder_layer_24/mlp_norm/vars": "model_00002.weights.h5",
821
+ "/layers/janus_text_decoder_layer_25/attention/key/vars": "model_00002.weights.h5",
822
+ "/layers/janus_text_decoder_layer_25/attention/output_proj/vars": "model_00002.weights.h5",
823
+ "/layers/janus_text_decoder_layer_25/attention/query/vars": "model_00002.weights.h5",
824
+ "/layers/janus_text_decoder_layer_25/attention/value/vars": "model_00002.weights.h5",
825
+ "/layers/janus_text_decoder_layer_25/attention_norm/vars": "model_00002.weights.h5",
826
+ "/layers/janus_text_decoder_layer_25/mlp/down/vars": "model_00002.weights.h5",
827
+ "/layers/janus_text_decoder_layer_25/mlp/gate/vars": "model_00002.weights.h5",
828
+ "/layers/janus_text_decoder_layer_25/mlp/up/vars": "model_00002.weights.h5",
829
+ "/layers/janus_text_decoder_layer_25/mlp_norm/vars": "model_00002.weights.h5",
830
+ "/layers/janus_text_decoder_layer_26/attention/key/vars": "model_00002.weights.h5",
831
+ "/layers/janus_text_decoder_layer_26/attention/output_proj/vars": "model_00002.weights.h5",
832
+ "/layers/janus_text_decoder_layer_26/attention/query/vars": "model_00002.weights.h5",
833
+ "/layers/janus_text_decoder_layer_26/attention/value/vars": "model_00002.weights.h5",
834
+ "/layers/janus_text_decoder_layer_26/attention_norm/vars": "model_00002.weights.h5",
835
+ "/layers/janus_text_decoder_layer_26/mlp/down/vars": "model_00002.weights.h5",
836
+ "/layers/janus_text_decoder_layer_26/mlp/gate/vars": "model_00002.weights.h5",
837
+ "/layers/janus_text_decoder_layer_26/mlp/up/vars": "model_00002.weights.h5",
838
+ "/layers/janus_text_decoder_layer_26/mlp_norm/vars": "model_00002.weights.h5",
839
+ "/layers/janus_text_decoder_layer_27/attention/key/vars": "model_00002.weights.h5",
840
+ "/layers/janus_text_decoder_layer_27/attention/output_proj/vars": "model_00002.weights.h5",
841
+ "/layers/janus_text_decoder_layer_27/attention/query/vars": "model_00002.weights.h5",
842
+ "/layers/janus_text_decoder_layer_27/attention/value/vars": "model_00002.weights.h5",
843
+ "/layers/janus_text_decoder_layer_27/attention_norm/vars": "model_00002.weights.h5",
844
+ "/layers/janus_text_decoder_layer_27/mlp/down/vars": "model_00002.weights.h5",
845
+ "/layers/janus_text_decoder_layer_27/mlp/gate/vars": "model_00002.weights.h5",
846
+ "/layers/janus_text_decoder_layer_27/mlp/up/vars": "model_00002.weights.h5",
847
+ "/layers/janus_text_decoder_layer_27/mlp_norm/vars": "model_00002.weights.h5",
848
+ "/layers/janus_text_decoder_layer_28/attention/key/vars": "model_00002.weights.h5",
849
+ "/layers/janus_text_decoder_layer_28/attention/output_proj/vars": "model_00002.weights.h5",
850
+ "/layers/janus_text_decoder_layer_28/attention/query/vars": "model_00002.weights.h5",
851
+ "/layers/janus_text_decoder_layer_28/attention/value/vars": "model_00002.weights.h5",
852
+ "/layers/janus_text_decoder_layer_28/attention_norm/vars": "model_00002.weights.h5",
853
+ "/layers/janus_text_decoder_layer_28/mlp/down/vars": "model_00002.weights.h5",
854
+ "/layers/janus_text_decoder_layer_28/mlp/gate/vars": "model_00002.weights.h5",
855
+ "/layers/janus_text_decoder_layer_28/mlp/up/vars": "model_00002.weights.h5",
856
+ "/layers/janus_text_decoder_layer_28/mlp_norm/vars": "model_00002.weights.h5",
857
+ "/layers/janus_text_decoder_layer_29/attention/key/vars": "model_00002.weights.h5",
858
+ "/layers/janus_text_decoder_layer_29/attention/output_proj/vars": "model_00002.weights.h5",
859
+ "/layers/janus_text_decoder_layer_29/attention/query/vars": "model_00002.weights.h5",
860
+ "/layers/janus_text_decoder_layer_29/attention/value/vars": "model_00002.weights.h5",
861
+ "/layers/janus_text_decoder_layer_29/attention_norm/vars": "model_00002.weights.h5",
862
+ "/layers/janus_text_decoder_layer_29/mlp/down/vars": "model_00002.weights.h5",
863
+ "/layers/janus_text_decoder_layer_29/mlp/gate/vars": "model_00002.weights.h5",
864
+ "/layers/janus_text_decoder_layer_29/mlp/up/vars": "model_00002.weights.h5",
865
+ "/layers/janus_text_decoder_layer_29/mlp_norm/vars": "model_00002.weights.h5",
866
+ "/layers/janus_text_rms_norm/vars": "model_00002.weights.h5"
867
+ }
868
+ }
model_00000.weights.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bab9d4c3b2a0848eb042b4d4c428f5da2321030b76411e1abfa1d3ffa3c6243e
3
+ size 5356440952
model_00001.weights.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c49ce25cf35e73585514a44ae3b3aa4d4b3b400eca871120ae813cfca7e64f8
3
+ size 5352518784
model_00002.weights.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9c4fd49cf8627fc857946bbcaeecb2f38e14ce9780c1b239fff9f6bb17d960d
3
+ size 2923768784
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff