Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
596
dacorvo HF Staff commited on
Commit
8a6ec65
·
verified ·
1 Parent(s): 024b363

Synchronizing local compiler cache.

Browse files
Files changed (8) hide show
  1. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0.dev4/stable-diffusion/stabilityai/sdxl-turbo/c4162c03a5bca6d7620e.json +162 -0
  2. neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/text_encoder/model.neuron +1 -1
  3. neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/text_encoder_2/model.neuron +1 -1
  4. neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/unet/model.neuron +1 -1
  5. neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/vae_decoder/config.json +4 -4
  6. neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/vae_decoder/model.neuron +1 -1
  7. neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/vae_encoder/config.json +4 -4
  8. neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/vae_encoder/model.neuron +1 -1
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0.dev4/stable-diffusion/stabilityai/sdxl-turbo/c4162c03a5bca6d7620e.json ADDED
@@ -0,0 +1,162 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "MultiModelCacheEntry",
3
+ "_model_id": "stabilityai/sdxl-turbo",
4
+ "_task": null,
5
+ "text_encoder": {
6
+ "architectures": [
7
+ "CLIPTextModel"
8
+ ],
9
+ "attention_dropout": 0.0,
10
+ "dropout": 0.0,
11
+ "hidden_act": "quick_gelu",
12
+ "hidden_size": 768,
13
+ "initializer_factor": 1.0,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 3072,
16
+ "layer_norm_eps": 1e-05,
17
+ "max_position_embeddings": 77,
18
+ "model_type": "clip_text_model",
19
+ "neuron": {
20
+ "auto_cast": "matmul",
21
+ "auto_cast_type": "bf16",
22
+ "compiler_type": "neuronx-cc",
23
+ "compiler_version": "2.17.194.0+d312836f",
24
+ "dynamic_batch_size": false,
25
+ "inline_weights_to_neff": false,
26
+ "optlevel": "2",
27
+ "output_attentions": false,
28
+ "output_hidden_states": false,
29
+ "static_batch_size": 1,
30
+ "static_sequence_length": 77,
31
+ "task": "feature-extraction",
32
+ "tensor_parallel_size": 1
33
+ },
34
+ "num_attention_heads": 12,
35
+ "num_hidden_layers": 12,
36
+ "output_hidden_states": true,
37
+ "vocab_size": 49408
38
+ },
39
+ "text_encoder_2": {
40
+ "architectures": [
41
+ "CLIPTextModelWithProjection"
42
+ ],
43
+ "attention_dropout": 0.0,
44
+ "dropout": 0.0,
45
+ "hidden_act": "gelu",
46
+ "hidden_size": 1280,
47
+ "initializer_factor": 1.0,
48
+ "initializer_range": 0.02,
49
+ "intermediate_size": 5120,
50
+ "layer_norm_eps": 1e-05,
51
+ "max_position_embeddings": 77,
52
+ "model_type": "clip_text_model",
53
+ "neuron": {
54
+ "auto_cast": "matmul",
55
+ "auto_cast_type": "bf16",
56
+ "compiler_type": "neuronx-cc",
57
+ "compiler_version": "2.17.194.0+d312836f",
58
+ "dynamic_batch_size": false,
59
+ "inline_weights_to_neff": false,
60
+ "optlevel": "2",
61
+ "output_attentions": false,
62
+ "output_hidden_states": false,
63
+ "static_batch_size": 1,
64
+ "static_sequence_length": 77,
65
+ "task": "feature-extraction",
66
+ "tensor_parallel_size": 1
67
+ },
68
+ "num_attention_heads": 20,
69
+ "num_hidden_layers": 32,
70
+ "output_hidden_states": true,
71
+ "vocab_size": 49408
72
+ },
73
+ "unet": {
74
+ "_class_name": "UNet2DConditionModel",
75
+ "act_fn": "silu",
76
+ "addition_embed_type": "text_time",
77
+ "addition_embed_type_num_heads": 64,
78
+ "addition_time_embed_dim": 256,
79
+ "attention_head_dim": [
80
+ 5,
81
+ 10,
82
+ 20
83
+ ],
84
+ "attention_type": "default",
85
+ "block_out_channels": [
86
+ 320,
87
+ 640,
88
+ 1280
89
+ ],
90
+ "center_input_sample": false,
91
+ "class_embed_type": null,
92
+ "class_embeddings_concat": false,
93
+ "conv_in_kernel": 3,
94
+ "conv_out_kernel": 3,
95
+ "cross_attention_dim": 2048,
96
+ "cross_attention_norm": null,
97
+ "down_block_types": [
98
+ "DownBlock2D",
99
+ "CrossAttnDownBlock2D",
100
+ "CrossAttnDownBlock2D"
101
+ ],
102
+ "downsample_padding": 1,
103
+ "dropout": 0.0,
104
+ "dual_cross_attention": false,
105
+ "encoder_hid_dim": null,
106
+ "encoder_hid_dim_type": null,
107
+ "flip_sin_to_cos": true,
108
+ "freq_shift": 0,
109
+ "in_channels": 4,
110
+ "layers_per_block": 2,
111
+ "mid_block_only_cross_attention": null,
112
+ "mid_block_scale_factor": 1,
113
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
114
+ "neuron": {
115
+ "auto_cast": "matmul",
116
+ "auto_cast_type": "bf16",
117
+ "compiler_type": "neuronx-cc",
118
+ "compiler_version": "2.17.194.0+d312836f",
119
+ "dynamic_batch_size": false,
120
+ "inline_weights_to_neff": true,
121
+ "optlevel": "2",
122
+ "output_attentions": false,
123
+ "output_hidden_states": false,
124
+ "static_batch_size": 1,
125
+ "static_height": 96,
126
+ "static_num_channels": 4,
127
+ "static_sequence_length": 77,
128
+ "static_vae_scale_factor": 8,
129
+ "static_width": 96,
130
+ "task": "semantic-segmentation",
131
+ "tensor_parallel_size": 1
132
+ },
133
+ "norm_eps": 1e-05,
134
+ "norm_num_groups": 32,
135
+ "num_attention_heads": null,
136
+ "num_class_embeds": null,
137
+ "only_cross_attention": false,
138
+ "out_channels": 4,
139
+ "projection_class_embeddings_input_dim": 2816,
140
+ "resnet_out_scale_factor": 1.0,
141
+ "resnet_skip_time_act": false,
142
+ "resnet_time_scale_shift": "default",
143
+ "reverse_transformer_layers_per_block": null,
144
+ "time_cond_proj_dim": null,
145
+ "time_embedding_act_fn": null,
146
+ "time_embedding_dim": null,
147
+ "time_embedding_type": "positional",
148
+ "timestep_post_act": null,
149
+ "transformer_layers_per_block": [
150
+ 1,
151
+ 2,
152
+ 10
153
+ ],
154
+ "up_block_types": [
155
+ "CrossAttnUpBlock2D",
156
+ "CrossAttnUpBlock2D",
157
+ "UpBlock2D"
158
+ ],
159
+ "upcast_attention": null,
160
+ "use_linear_projection": true
161
+ }
162
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/text_encoder/model.neuron CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b06e47e21ffb869d1ea14f2c214a6f3f9ac318a657d0826cb672a14b570b1d82
3
  size 495950649
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f7394ce24b82b25da26e8b8ceb37bbc41fe281d7e08d59fdf0396cb22d357a7
3
  size 495950649
neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/text_encoder_2/model.neuron CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4588f57683173b6de48513709970d9b5626af7a15f0e4399dc1629395183e2d5
3
  size 2801222123
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99ef153a1400658d0be9f943a67be6e32f4ae05930ac30a3cacbf69160129de4
3
  size 2801222123
neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/unet/model.neuron CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2a8144f92e7c6719c38473410958bfe3a2b1076146067e3b7cd3f15e23cfc2d0
3
  size 4156474685
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30de0423291f38222dbce50ae7989dcf8a00bb01e90c6d43509cf317b5632493
3
  size 4156474685
neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/vae_decoder/config.json CHANGED
@@ -4,12 +4,12 @@
4
  "_diffusers_version": "0.30.3",
5
  "_name_or_path": "/home/runner/.cache/huggingface/hub/models--stabilityai--sdxl-turbo/snapshots/71153311d3dbb46851df1931d3ca6e939de83304/vae",
6
  "_use_default_values": [
7
- "mid_block_add_attention",
8
- "use_quant_conv",
9
- "latents_mean",
10
  "latents_std",
11
  "use_post_quant_conv",
12
- "shift_factor"
 
 
 
13
  ],
14
  "act_fn": "silu",
15
  "block_out_channels": [
 
4
  "_diffusers_version": "0.30.3",
5
  "_name_or_path": "/home/runner/.cache/huggingface/hub/models--stabilityai--sdxl-turbo/snapshots/71153311d3dbb46851df1931d3ca6e939de83304/vae",
6
  "_use_default_values": [
 
 
 
7
  "latents_std",
8
  "use_post_quant_conv",
9
+ "shift_factor",
10
+ "latents_mean",
11
+ "mid_block_add_attention",
12
+ "use_quant_conv"
13
  ],
14
  "act_fn": "silu",
15
  "block_out_channels": [
neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/vae_decoder/model.neuron CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f9638e64a6fe9fb8949375315d80107852382a82ceda886377d55600fd6dbe7b
3
  size 509313587
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:105a7027a726e87564fe8174301cb8a9614eb1542dc553afcd236ae19dc185b9
3
  size 509313587
neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/vae_encoder/config.json CHANGED
@@ -4,12 +4,12 @@
4
  "_diffusers_version": "0.30.3",
5
  "_name_or_path": "/home/runner/.cache/huggingface/hub/models--stabilityai--sdxl-turbo/snapshots/71153311d3dbb46851df1931d3ca6e939de83304/vae",
6
  "_use_default_values": [
7
- "mid_block_add_attention",
8
- "use_quant_conv",
9
- "latents_mean",
10
  "latents_std",
11
  "use_post_quant_conv",
12
- "shift_factor"
 
 
 
13
  ],
14
  "act_fn": "silu",
15
  "block_out_channels": [
 
4
  "_diffusers_version": "0.30.3",
5
  "_name_or_path": "/home/runner/.cache/huggingface/hub/models--stabilityai--sdxl-turbo/snapshots/71153311d3dbb46851df1931d3ca6e939de83304/vae",
6
  "_use_default_values": [
 
 
 
7
  "latents_std",
8
  "use_post_quant_conv",
9
+ "shift_factor",
10
+ "latents_mean",
11
+ "mid_block_add_attention",
12
+ "use_quant_conv"
13
  ],
14
  "act_fn": "silu",
15
  "block_out_channels": [
neuronxcc-2.17.194.0+d312836f/MODULE_c4162c03a5bca6d7620e/vae_encoder/model.neuron CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:71294ce2d791c789e8149e954e8b7d4989e7ac1e20c00f6c2713376248fa529a
3
  size 285682227
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a2f0f62b95ebc64fd5770797a74f83ce80d21754f3a3d590ca3eb3c8bdf60ad
3
  size 285682227