Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
646
dacorvo HF Staff commited on
Commit
79a75e6
·
verified ·
1 Parent(s): 5b5b7d6

Synchronizing local compiler cache.

Browse files
Files changed (9) hide show
  1. neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev6/stable-diffusion/stabilityai/stable-diffusion-xl-base-1.0/654d6f48e079352175d2.json +173 -0
  2. neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/text_encoder/model.neuron +1 -1
  3. neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/text_encoder_2/model.neuron +1 -1
  4. neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/unet/config.json +2 -2
  5. neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/unet/model.neuron +1 -1
  6. neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/vae_decoder/config.json +4 -4
  7. neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/vae_decoder/model.neuron +1 -1
  8. neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/vae_encoder/config.json +4 -4
  9. neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/vae_encoder/model.neuron +1 -1
neuronxcc-2.19.8089.0+8ab9f450/0_REGISTRY/0.3.1.dev6/stable-diffusion/stabilityai/stable-diffusion-xl-base-1.0/654d6f48e079352175d2.json ADDED
@@ -0,0 +1,173 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "MultiModelCacheEntry",
3
+ "_model_id": "stabilityai/stable-diffusion-xl-base-1.0",
4
+ "_task": null,
5
+ "text_encoder": {
6
+ "architectures": [
7
+ "CLIPTextModel"
8
+ ],
9
+ "attention_dropout": 0.0,
10
+ "dropout": 0.0,
11
+ "export_model_type": "clip-text-model",
12
+ "hidden_act": "quick_gelu",
13
+ "hidden_size": 768,
14
+ "initializer_factor": 1.0,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 3072,
17
+ "layer_norm_eps": 1e-05,
18
+ "max_position_embeddings": 77,
19
+ "model_type": "clip_text_model",
20
+ "neuron": {
21
+ "auto_cast": "matmul",
22
+ "auto_cast_type": "bf16",
23
+ "compiler_type": "neuronx-cc",
24
+ "compiler_version": "2.19.8089.0+8ab9f450",
25
+ "dynamic_batch_size": false,
26
+ "float_dtype": "fp32",
27
+ "inline_weights_to_neff": false,
28
+ "int_dtype": "int64",
29
+ "optlevel": "2",
30
+ "output_attentions": false,
31
+ "output_hidden_states": false,
32
+ "static_batch_size": 1,
33
+ "static_sequence_length": 77,
34
+ "task": "feature-extraction",
35
+ "tensor_parallel_size": 1
36
+ },
37
+ "num_attention_heads": 12,
38
+ "num_hidden_layers": 12,
39
+ "output_hidden_states": true,
40
+ "return_dict": false,
41
+ "vocab_size": 49408
42
+ },
43
+ "text_encoder_2": {
44
+ "architectures": [
45
+ "CLIPTextModelWithProjection"
46
+ ],
47
+ "attention_dropout": 0.0,
48
+ "dropout": 0.0,
49
+ "export_model_type": "clip-text-with-projection",
50
+ "hidden_act": "gelu",
51
+ "hidden_size": 1280,
52
+ "initializer_factor": 1.0,
53
+ "initializer_range": 0.02,
54
+ "intermediate_size": 5120,
55
+ "layer_norm_eps": 1e-05,
56
+ "max_position_embeddings": 77,
57
+ "model_type": "clip_text_model",
58
+ "neuron": {
59
+ "auto_cast": "matmul",
60
+ "auto_cast_type": "bf16",
61
+ "compiler_type": "neuronx-cc",
62
+ "compiler_version": "2.19.8089.0+8ab9f450",
63
+ "dynamic_batch_size": false,
64
+ "float_dtype": "fp32",
65
+ "inline_weights_to_neff": false,
66
+ "int_dtype": "int64",
67
+ "optlevel": "2",
68
+ "output_attentions": false,
69
+ "output_hidden_states": false,
70
+ "static_batch_size": 1,
71
+ "static_sequence_length": 77,
72
+ "task": "feature-extraction",
73
+ "tensor_parallel_size": 1
74
+ },
75
+ "num_attention_heads": 20,
76
+ "num_hidden_layers": 32,
77
+ "output_hidden_states": true,
78
+ "return_dict": false,
79
+ "vocab_size": 49408
80
+ },
81
+ "unet": {
82
+ "_class_name": "UNet2DConditionModel",
83
+ "act_fn": "silu",
84
+ "addition_embed_type": "text_time",
85
+ "addition_embed_type_num_heads": 64,
86
+ "addition_time_embed_dim": 256,
87
+ "attention_head_dim": [
88
+ 5,
89
+ 10,
90
+ 20
91
+ ],
92
+ "attention_type": "default",
93
+ "block_out_channels": [
94
+ 320,
95
+ 640,
96
+ 1280
97
+ ],
98
+ "center_input_sample": false,
99
+ "class_embed_type": null,
100
+ "class_embeddings_concat": false,
101
+ "conv_in_kernel": 3,
102
+ "conv_out_kernel": 3,
103
+ "cross_attention_dim": 2048,
104
+ "cross_attention_norm": null,
105
+ "down_block_types": [
106
+ "DownBlock2D",
107
+ "CrossAttnDownBlock2D",
108
+ "CrossAttnDownBlock2D"
109
+ ],
110
+ "downsample_padding": 1,
111
+ "dropout": 0.0,
112
+ "dual_cross_attention": false,
113
+ "encoder_hid_dim": null,
114
+ "encoder_hid_dim_type": null,
115
+ "flip_sin_to_cos": true,
116
+ "freq_shift": 0,
117
+ "in_channels": 4,
118
+ "layers_per_block": 2,
119
+ "mid_block_only_cross_attention": null,
120
+ "mid_block_scale_factor": 1,
121
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
122
+ "neuron": {
123
+ "auto_cast": "matmul",
124
+ "auto_cast_type": "bf16",
125
+ "compiler_type": "neuronx-cc",
126
+ "compiler_version": "2.19.8089.0+8ab9f450",
127
+ "dynamic_batch_size": false,
128
+ "float_dtype": "fp32",
129
+ "inline_weights_to_neff": true,
130
+ "int_dtype": "int64",
131
+ "optlevel": "2",
132
+ "output_attentions": false,
133
+ "output_hidden_states": false,
134
+ "static_batch_size": 1,
135
+ "static_height": 96,
136
+ "static_num_channels": 4,
137
+ "static_sequence_length": 77,
138
+ "static_vae_scale_factor": 8,
139
+ "static_width": 96,
140
+ "task": "semantic-segmentation",
141
+ "tensor_parallel_size": 1
142
+ },
143
+ "norm_eps": 1e-05,
144
+ "norm_num_groups": 32,
145
+ "num_attention_heads": null,
146
+ "num_class_embeds": null,
147
+ "only_cross_attention": false,
148
+ "out_channels": 4,
149
+ "output_attentions": false,
150
+ "projection_class_embeddings_input_dim": 2816,
151
+ "resnet_out_scale_factor": 1.0,
152
+ "resnet_skip_time_act": false,
153
+ "resnet_time_scale_shift": "default",
154
+ "reverse_transformer_layers_per_block": null,
155
+ "time_cond_proj_dim": null,
156
+ "time_embedding_act_fn": null,
157
+ "time_embedding_dim": null,
158
+ "time_embedding_type": "positional",
159
+ "timestep_post_act": null,
160
+ "transformer_layers_per_block": [
161
+ 1,
162
+ 2,
163
+ 10
164
+ ],
165
+ "up_block_types": [
166
+ "CrossAttnUpBlock2D",
167
+ "CrossAttnUpBlock2D",
168
+ "UpBlock2D"
169
+ ],
170
+ "upcast_attention": null,
171
+ "use_linear_projection": true
172
+ }
173
+ }
neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/text_encoder/model.neuron CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:79b9877c930c8802f04560fbae7b3398e033d55a1bdca9cd39bfd47c5d462366
3
  size 496156793
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7514c064a0d7b144fbcbee6c5a7e8a02a028291cc4f3cbf81e5b0dc0d14fd9e
3
  size 496156793
neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/text_encoder_2/model.neuron CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2824e71e4b85d4e80781d8ff7b37ad2c29ad526699fec80d207d72354f005d90
3
  size 2801804587
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:035359e1ef3cd4f1a539790c308a1ad189f4efa1cd75e31b1b572528a6b0e946
3
  size 2801804587
neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/unet/config.json CHANGED
@@ -2,9 +2,9 @@
2
  "_class_name": "UNet2DConditionModel",
3
  "_diffusers_version": "0.35.1",
4
  "_use_default_values": [
 
5
  "dropout",
6
- "reverse_transformer_layers_per_block",
7
- "attention_type"
8
  ],
9
  "act_fn": "silu",
10
  "addition_embed_type": "text_time",
 
2
  "_class_name": "UNet2DConditionModel",
3
  "_diffusers_version": "0.35.1",
4
  "_use_default_values": [
5
+ "attention_type",
6
  "dropout",
7
+ "reverse_transformer_layers_per_block"
 
8
  ],
9
  "act_fn": "silu",
10
  "addition_embed_type": "text_time",
neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/unet/model.neuron CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9eabf4a3b4adfb01a6fddfc2b06ecdda5b530296c7f8ef8a25ae030680005814
3
  size 4157584957
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b65242707e3a4046c164443ae058ecf9d5f811cdb40c1db7af26812c20a1dffb
3
  size 4157584957
neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/vae_decoder/config.json CHANGED
@@ -2,12 +2,12 @@
2
  "_class_name": "AutoencoderKL",
3
  "_diffusers_version": "0.35.1",
4
  "_use_default_values": [
 
 
 
5
  "latents_std",
6
  "use_post_quant_conv",
7
- "mid_block_add_attention",
8
- "latents_mean",
9
- "use_quant_conv",
10
- "shift_factor"
11
  ],
12
  "act_fn": "silu",
13
  "block_out_channels": [
 
2
  "_class_name": "AutoencoderKL",
3
  "_diffusers_version": "0.35.1",
4
  "_use_default_values": [
5
+ "shift_factor",
6
+ "latents_mean",
7
+ "mid_block_add_attention",
8
  "latents_std",
9
  "use_post_quant_conv",
10
+ "use_quant_conv"
 
 
 
11
  ],
12
  "act_fn": "silu",
13
  "block_out_channels": [
neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/vae_decoder/model.neuron CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:490d45db4b0afd0d167ef4b3c5ea8e04477820785a600ca1bdeaa7e019aaa16a
3
  size 349015667
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a7e836235316d5adce16bba5407f1f05679fc30923e90d81d97c1e43463cb57
3
  size 349015667
neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/vae_encoder/config.json CHANGED
@@ -2,12 +2,12 @@
2
  "_class_name": "AutoencoderKL",
3
  "_diffusers_version": "0.35.1",
4
  "_use_default_values": [
 
 
 
5
  "latents_std",
6
  "use_post_quant_conv",
7
- "mid_block_add_attention",
8
- "latents_mean",
9
- "use_quant_conv",
10
- "shift_factor"
11
  ],
12
  "act_fn": "silu",
13
  "block_out_channels": [
 
2
  "_class_name": "AutoencoderKL",
3
  "_diffusers_version": "0.35.1",
4
  "_use_default_values": [
5
+ "shift_factor",
6
+ "latents_mean",
7
+ "mid_block_add_attention",
8
  "latents_std",
9
  "use_post_quant_conv",
10
+ "use_quant_conv"
 
 
 
11
  ],
12
  "act_fn": "silu",
13
  "block_out_channels": [
neuronxcc-2.19.8089.0+8ab9f450/MODULE_654d6f48e079352175d2/vae_encoder/model.neuron CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7f97985c4363c0860163ecbd7c59d92518d59c7e279c0250c5382231c07fffa4
3
  size 201275507
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e312b094986e1d7f58b1e20924e5625858426844eca77107458667859f4d67c
3
  size 201275507