Hugging Face's logo

aws-neuron
/
optimum-neuron-cache

Model card Files
xet
Community
584
dacorvo HF Staff commited on
Commit
4fbe8ac
·
verified ·
1 Parent(s): d1ed9e7

Synchronizing local compiler cache.

Browse files
Files changed (17) hide show
  1. .gitattributes +4 -0
  2. neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0.dev2/inference/stable-diffusion/stabilityai/stable-diffusion-2-1/ee34e363190e7c6310a6.json +127 -0
  3. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/feature_extractor/preprocessor_config.json +27 -0
  4. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/model_index.json +38 -0
  5. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/scheduler/scheduler_config.json +20 -0
  6. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/text_encoder/config.json +50 -0
  7. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/text_encoder/model.neuron +3 -0
  8. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/tokenizer/merges.txt +0 -0
  9. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/tokenizer/special_tokens_map.json +24 -0
  10. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/tokenizer/tokenizer_config.json +39 -0
  11. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/tokenizer/vocab.json +0 -0
  12. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/unet/config.json +131 -0
  13. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/unet/model.neuron +3 -0
  14. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/vae_decoder/config.json +74 -0
  15. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/vae_decoder/model.neuron +3 -0
  16. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/vae_encoder/config.json +74 -0
  17. neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/vae_encoder/model.neuron +3 -0
.gitattributes CHANGED
@@ -7859,3 +7859,7 @@ neuronxcc-2.17.194.0+d312836f/MODULE_290d6364f00ca86a0a51/vae_decoder/model.neur
7859
  neuronxcc-2.17.194.0+d312836f/MODULE_290d6364f00ca86a0a51/vae_encoder/model.neuron filter=lfs diff=lfs merge=lfs -text
7860
  neuronxcc-2.17.194.0+d312836f/MODULE_c477e2d21d208f7e36a7+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
7861
  neuronxcc-2.17.194.0+d312836f/MODULE_d9090a85eb256ecbcf5a+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
7859
  neuronxcc-2.17.194.0+d312836f/MODULE_290d6364f00ca86a0a51/vae_encoder/model.neuron filter=lfs diff=lfs merge=lfs -text
7860
  neuronxcc-2.17.194.0+d312836f/MODULE_c477e2d21d208f7e36a7+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
7861
  neuronxcc-2.17.194.0+d312836f/MODULE_d9090a85eb256ecbcf5a+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
7862
+ neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/text_encoder/model.neuron filter=lfs diff=lfs merge=lfs -text
7863
+ neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/unet/model.neuron filter=lfs diff=lfs merge=lfs -text
7864
+ neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/vae_decoder/model.neuron filter=lfs diff=lfs merge=lfs -text
7865
+ neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/vae_encoder/model.neuron filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.17.194.0+d312836f/0_REGISTRY/0.2.0.dev2/inference/stable-diffusion/stabilityai/stable-diffusion-2-1/ee34e363190e7c6310a6.json ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "MultiModelCacheEntry",
3
+ "_model_id": "stabilityai/stable-diffusion-2-1",
4
+ "_task": null,
5
+ "text_encoder": {
6
+ "architectures": [
7
+ "CLIPTextModel"
8
+ ],
9
+ "attention_dropout": 0.0,
10
+ "dropout": 0.0,
11
+ "hidden_act": "gelu",
12
+ "hidden_size": 1024,
13
+ "initializer_factor": 1.0,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 4096,
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": 16,
35
+ "num_hidden_layers": 23,
36
+ "vocab_size": 49408
37
+ },
38
+ "unet": {
39
+ "_class_name": "UNet2DConditionModel",
40
+ "act_fn": "silu",
41
+ "addition_embed_type": null,
42
+ "addition_embed_type_num_heads": 64,
43
+ "addition_time_embed_dim": null,
44
+ "attention_head_dim": [
45
+ 5,
46
+ 10,
47
+ 20,
48
+ 20
49
+ ],
50
+ "attention_type": "default",
51
+ "block_out_channels": [
52
+ 320,
53
+ 640,
54
+ 1280,
55
+ 1280
56
+ ],
57
+ "center_input_sample": false,
58
+ "class_embed_type": null,
59
+ "class_embeddings_concat": false,
60
+ "conv_in_kernel": 3,
61
+ "conv_out_kernel": 3,
62
+ "cross_attention_dim": 1024,
63
+ "cross_attention_norm": null,
64
+ "down_block_types": [
65
+ "CrossAttnDownBlock2D",
66
+ "CrossAttnDownBlock2D",
67
+ "CrossAttnDownBlock2D",
68
+ "DownBlock2D"
69
+ ],
70
+ "downsample_padding": 1,
71
+ "dropout": 0.0,
72
+ "dual_cross_attention": false,
73
+ "encoder_hid_dim": null,
74
+ "encoder_hid_dim_type": null,
75
+ "flip_sin_to_cos": true,
76
+ "freq_shift": 0,
77
+ "in_channels": 4,
78
+ "layers_per_block": 2,
79
+ "mid_block_only_cross_attention": null,
80
+ "mid_block_scale_factor": 1,
81
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
82
+ "neuron": {
83
+ "auto_cast": "matmul",
84
+ "auto_cast_type": "bf16",
85
+ "compiler_type": "neuronx-cc",
86
+ "compiler_version": "2.17.194.0+d312836f",
87
+ "dynamic_batch_size": false,
88
+ "inline_weights_to_neff": false,
89
+ "optlevel": "2",
90
+ "output_attentions": false,
91
+ "output_hidden_states": false,
92
+ "static_batch_size": 1,
93
+ "static_height": 64,
94
+ "static_num_channels": 4,
95
+ "static_sequence_length": 77,
96
+ "static_vae_scale_factor": 8,
97
+ "static_width": 64,
98
+ "task": "semantic-segmentation",
99
+ "tensor_parallel_size": 1
100
+ },
101
+ "norm_eps": 1e-05,
102
+ "norm_num_groups": 32,
103
+ "num_attention_heads": null,
104
+ "num_class_embeds": null,
105
+ "only_cross_attention": false,
106
+ "out_channels": 4,
107
+ "projection_class_embeddings_input_dim": null,
108
+ "resnet_out_scale_factor": 1.0,
109
+ "resnet_skip_time_act": false,
110
+ "resnet_time_scale_shift": "default",
111
+ "reverse_transformer_layers_per_block": null,
112
+ "time_cond_proj_dim": null,
113
+ "time_embedding_act_fn": null,
114
+ "time_embedding_dim": null,
115
+ "time_embedding_type": "positional",
116
+ "timestep_post_act": null,
117
+ "transformer_layers_per_block": 1,
118
+ "up_block_types": [
119
+ "UpBlock2D",
120
+ "CrossAttnUpBlock2D",
121
+ "CrossAttnUpBlock2D",
122
+ "CrossAttnUpBlock2D"
123
+ ],
124
+ "upcast_attention": true,
125
+ "use_linear_projection": true
126
+ }
127
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/feature_extractor/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "CLIPImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/model_index.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "StableDiffusionPipeline",
3
+ "_diffusers_version": "0.30.3",
4
+ "_name_or_path": "stabilityai/stable-diffusion-2-1",
5
+ "feature_extractor": [
6
+ "transformers",
7
+ "CLIPImageProcessor"
8
+ ],
9
+ "image_encoder": [
10
+ null,
11
+ null
12
+ ],
13
+ "requires_safety_checker": false,
14
+ "safety_checker": [
15
+ null,
16
+ null
17
+ ],
18
+ "scheduler": [
19
+ "diffusers",
20
+ "DDIMScheduler"
21
+ ],
22
+ "text_encoder": [
23
+ "transformers",
24
+ "CLIPTextModel"
25
+ ],
26
+ "tokenizer": [
27
+ "transformers",
28
+ "CLIPTokenizer"
29
+ ],
30
+ "unet": [
31
+ "diffusers",
32
+ "UNet2DConditionModel"
33
+ ],
34
+ "vae": [
35
+ "diffusers",
36
+ "AutoencoderKL"
37
+ ]
38
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/scheduler/scheduler_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "DDIMScheduler",
3
+ "_diffusers_version": "0.30.3",
4
+ "beta_end": 0.012,
5
+ "beta_schedule": "scaled_linear",
6
+ "beta_start": 0.00085,
7
+ "clip_sample": false,
8
+ "clip_sample_range": 1.0,
9
+ "dynamic_thresholding_ratio": 0.995,
10
+ "num_train_timesteps": 1000,
11
+ "prediction_type": "v_prediction",
12
+ "rescale_betas_zero_snr": false,
13
+ "sample_max_value": 1.0,
14
+ "set_alpha_to_one": false,
15
+ "skip_prk_steps": true,
16
+ "steps_offset": 1,
17
+ "thresholding": false,
18
+ "timestep_spacing": "leading",
19
+ "trained_betas": null
20
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/text_encoder/config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "_name_or_path": "/home/runner/.cache/huggingface/hub/models--stabilityai--stable-diffusion-2-1/snapshots/5cae40e6a2745ae2b01ad92ae5043f95f23644d6/text_encoder",
4
+ "architectures": [
5
+ "CLIPTextModel"
6
+ ],
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 0,
9
+ "dropout": 0.0,
10
+ "eos_token_id": 2,
11
+ "hidden_act": "gelu",
12
+ "hidden_size": 1024,
13
+ "initializer_factor": 1.0,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 4096,
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
+ "input_names": [
27
+ "input_ids"
28
+ ],
29
+ "model_type": "clip-text-model",
30
+ "optlevel": "2",
31
+ "output_attentions": false,
32
+ "output_hidden_states": false,
33
+ "output_names": [
34
+ "last_hidden_state",
35
+ "pooler_output"
36
+ ],
37
+ "static_batch_size": 1,
38
+ "static_sequence_length": 77,
39
+ "task": "feature-extraction",
40
+ "tensor_parallel_size": 1
41
+ },
42
+ "num_attention_heads": 16,
43
+ "num_hidden_layers": 23,
44
+ "pad_token_id": 1,
45
+ "projection_dim": 512,
46
+ "torch_dtype": "float32",
47
+ "torchscript": true,
48
+ "transformers_version": "4.49.0",
49
+ "vocab_size": 49408
50
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/text_encoder/model.neuron ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7837f9c701b2e4c4c75229efa5c6133477759fa784fde010e1311c338ae8358
3
+ size 1372305465
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/tokenizer/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/tokenizer/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|startoftext|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "!",
17
+ "unk_token": {
18
+ "content": "<|endoftext|>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "!",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "49406": {
13
+ "content": "<|startoftext|>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "49407": {
21
+ "content": "<|endoftext|>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ }
28
+ },
29
+ "bos_token": "<|startoftext|>",
30
+ "clean_up_tokenization_spaces": false,
31
+ "do_lower_case": true,
32
+ "eos_token": "<|endoftext|>",
33
+ "errors": "replace",
34
+ "extra_special_tokens": {},
35
+ "model_max_length": 77,
36
+ "pad_token": "!",
37
+ "tokenizer_class": "CLIPTokenizer",
38
+ "unk_token": "<|endoftext|>"
39
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/tokenizer/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/unet/config.json ADDED
@@ -0,0 +1,131 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_commit_hash": null,
4
+ "_diffusers_version": "0.30.3",
5
+ "_name_or_path": "/home/runner/.cache/huggingface/hub/models--stabilityai--stable-diffusion-2-1/snapshots/5cae40e6a2745ae2b01ad92ae5043f95f23644d6/unet",
6
+ "_use_default_values": [
7
+ "time_embedding_dim",
8
+ "conv_out_kernel",
9
+ "encoder_hid_dim",
10
+ "class_embeddings_concat",
11
+ "addition_embed_type_num_heads",
12
+ "mid_block_only_cross_attention",
13
+ "addition_embed_type",
14
+ "reverse_transformer_layers_per_block",
15
+ "transformer_layers_per_block",
16
+ "mid_block_type",
17
+ "attention_type",
18
+ "time_embedding_type",
19
+ "time_embedding_act_fn",
20
+ "encoder_hid_dim_type",
21
+ "addition_time_embed_dim",
22
+ "resnet_skip_time_act",
23
+ "resnet_time_scale_shift",
24
+ "projection_class_embeddings_input_dim",
25
+ "cross_attention_norm",
26
+ "dropout",
27
+ "class_embed_type",
28
+ "num_attention_heads",
29
+ "timestep_post_act",
30
+ "resnet_out_scale_factor",
31
+ "conv_in_kernel",
32
+ "time_cond_proj_dim"
33
+ ],
34
+ "act_fn": "silu",
35
+ "addition_embed_type": null,
36
+ "addition_embed_type_num_heads": 64,
37
+ "addition_time_embed_dim": null,
38
+ "attention_head_dim": [
39
+ 5,
40
+ 10,
41
+ 20,
42
+ 20
43
+ ],
44
+ "attention_type": "default",
45
+ "block_out_channels": [
46
+ 320,
47
+ 640,
48
+ 1280,
49
+ 1280
50
+ ],
51
+ "center_input_sample": false,
52
+ "class_embed_type": null,
53
+ "class_embeddings_concat": false,
54
+ "conv_in_kernel": 3,
55
+ "conv_out_kernel": 3,
56
+ "cross_attention_dim": 1024,
57
+ "cross_attention_norm": null,
58
+ "down_block_types": [
59
+ "CrossAttnDownBlock2D",
60
+ "CrossAttnDownBlock2D",
61
+ "CrossAttnDownBlock2D",
62
+ "DownBlock2D"
63
+ ],
64
+ "downsample_padding": 1,
65
+ "dropout": 0.0,
66
+ "dual_cross_attention": false,
67
+ "encoder_hid_dim": null,
68
+ "encoder_hid_dim_type": null,
69
+ "flip_sin_to_cos": true,
70
+ "freq_shift": 0,
71
+ "in_channels": 4,
72
+ "layers_per_block": 2,
73
+ "mid_block_only_cross_attention": null,
74
+ "mid_block_scale_factor": 1,
75
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
76
+ "neuron": {
77
+ "auto_cast": "matmul",
78
+ "auto_cast_type": "bf16",
79
+ "compiler_type": "neuronx-cc",
80
+ "compiler_version": "2.17.194.0+d312836f",
81
+ "dynamic_batch_size": false,
82
+ "inline_weights_to_neff": false,
83
+ "input_names": [
84
+ "sample",
85
+ "timestep",
86
+ "encoder_hidden_states"
87
+ ],
88
+ "model_type": "unet",
89
+ "optlevel": "2",
90
+ "output_attentions": false,
91
+ "output_hidden_states": false,
92
+ "output_names": [
93
+ "sample"
94
+ ],
95
+ "static_batch_size": 1,
96
+ "static_height": 64,
97
+ "static_num_channels": 4,
98
+ "static_sequence_length": 77,
99
+ "static_vae_scale_factor": 8,
100
+ "static_width": 64,
101
+ "task": "semantic-segmentation",
102
+ "tensor_parallel_size": 1
103
+ },
104
+ "norm_eps": 1e-05,
105
+ "norm_num_groups": 32,
106
+ "num_attention_heads": null,
107
+ "num_class_embeds": null,
108
+ "only_cross_attention": false,
109
+ "out_channels": 4,
110
+ "projection_class_embeddings_input_dim": null,
111
+ "resnet_out_scale_factor": 1.0,
112
+ "resnet_skip_time_act": false,
113
+ "resnet_time_scale_shift": "default",
114
+ "reverse_transformer_layers_per_block": null,
115
+ "sample_size": 96,
116
+ "time_cond_proj_dim": null,
117
+ "time_embedding_act_fn": null,
118
+ "time_embedding_dim": null,
119
+ "time_embedding_type": "positional",
120
+ "timestep_post_act": null,
121
+ "transformer_layers_per_block": 1,
122
+ "transformers_version": null,
123
+ "up_block_types": [
124
+ "UpBlock2D",
125
+ "CrossAttnUpBlock2D",
126
+ "CrossAttnUpBlock2D",
127
+ "CrossAttnUpBlock2D"
128
+ ],
129
+ "upcast_attention": true,
130
+ "use_linear_projection": true
131
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/unet/model.neuron ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d21f1ccbd8cad70ddaf3ed5e31e6a70925238d038ab6f3a0a2997ded42a5f71b
3
+ size 3572109069
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/vae_decoder/config.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "AutoencoderKL",
3
+ "_commit_hash": null,
4
+ "_diffusers_version": "0.30.3",
5
+ "_name_or_path": "/home/runner/.cache/huggingface/hub/models--stabilityai--stable-diffusion-2-1/snapshots/5cae40e6a2745ae2b01ad92ae5043f95f23644d6/vae",
6
+ "_use_default_values": [
7
+ "shift_factor",
8
+ "force_upcast",
9
+ "use_post_quant_conv",
10
+ "scaling_factor",
11
+ "mid_block_add_attention",
12
+ "latents_mean",
13
+ "latents_std",
14
+ "use_quant_conv"
15
+ ],
16
+ "act_fn": "silu",
17
+ "block_out_channels": [
18
+ 128,
19
+ 256,
20
+ 512,
21
+ 512
22
+ ],
23
+ "down_block_types": [
24
+ "DownEncoderBlock2D",
25
+ "DownEncoderBlock2D",
26
+ "DownEncoderBlock2D",
27
+ "DownEncoderBlock2D"
28
+ ],
29
+ "force_upcast": true,
30
+ "in_channels": 3,
31
+ "latent_channels": 4,
32
+ "latents_mean": null,
33
+ "latents_std": null,
34
+ "layers_per_block": 2,
35
+ "mid_block_add_attention": true,
36
+ "neuron": {
37
+ "auto_cast": "matmul",
38
+ "auto_cast_type": "bf16",
39
+ "compiler_type": "neuronx-cc",
40
+ "compiler_version": "2.17.194.0+d312836f",
41
+ "dynamic_batch_size": false,
42
+ "inline_weights_to_neff": false,
43
+ "input_names": [
44
+ "latent_sample"
45
+ ],
46
+ "model_type": "vae-decoder",
47
+ "optlevel": "2",
48
+ "output_attentions": false,
49
+ "output_hidden_states": false,
50
+ "output_names": [
51
+ "sample"
52
+ ],
53
+ "static_batch_size": 1,
54
+ "static_height": 64,
55
+ "static_num_channels": 4,
56
+ "static_width": 64,
57
+ "task": "semantic-segmentation",
58
+ "tensor_parallel_size": 1
59
+ },
60
+ "norm_num_groups": 32,
61
+ "out_channels": 3,
62
+ "sample_size": 768,
63
+ "scaling_factor": 0.18215,
64
+ "shift_factor": null,
65
+ "transformers_version": null,
66
+ "up_block_types": [
67
+ "UpDecoderBlock2D",
68
+ "UpDecoderBlock2D",
69
+ "UpDecoderBlock2D",
70
+ "UpDecoderBlock2D"
71
+ ],
72
+ "use_post_quant_conv": true,
73
+ "use_quant_conv": true
74
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/vae_decoder/model.neuron ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33d6069cdf7144c01531f2318c422eaaab2e9190f14e967540fc5b19c056fc71
3
+ size 367178901
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/vae_encoder/config.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "AutoencoderKL",
3
+ "_commit_hash": null,
4
+ "_diffusers_version": "0.30.3",
5
+ "_name_or_path": "/home/runner/.cache/huggingface/hub/models--stabilityai--stable-diffusion-2-1/snapshots/5cae40e6a2745ae2b01ad92ae5043f95f23644d6/vae",
6
+ "_use_default_values": [
7
+ "shift_factor",
8
+ "force_upcast",
9
+ "use_post_quant_conv",
10
+ "scaling_factor",
11
+ "mid_block_add_attention",
12
+ "latents_mean",
13
+ "latents_std",
14
+ "use_quant_conv"
15
+ ],
16
+ "act_fn": "silu",
17
+ "block_out_channels": [
18
+ 128,
19
+ 256,
20
+ 512,
21
+ 512
22
+ ],
23
+ "down_block_types": [
24
+ "DownEncoderBlock2D",
25
+ "DownEncoderBlock2D",
26
+ "DownEncoderBlock2D",
27
+ "DownEncoderBlock2D"
28
+ ],
29
+ "force_upcast": true,
30
+ "in_channels": 3,
31
+ "latent_channels": 4,
32
+ "latents_mean": null,
33
+ "latents_std": null,
34
+ "layers_per_block": 2,
35
+ "mid_block_add_attention": true,
36
+ "neuron": {
37
+ "auto_cast": "matmul",
38
+ "auto_cast_type": "bf16",
39
+ "compiler_type": "neuronx-cc",
40
+ "compiler_version": "2.17.194.0+d312836f",
41
+ "dynamic_batch_size": false,
42
+ "inline_weights_to_neff": false,
43
+ "input_names": [
44
+ "sample"
45
+ ],
46
+ "model_type": "vae-encoder",
47
+ "optlevel": "2",
48
+ "output_attentions": false,
49
+ "output_hidden_states": false,
50
+ "output_names": [
51
+ "latent_parameters"
52
+ ],
53
+ "static_batch_size": 1,
54
+ "static_height": 512,
55
+ "static_num_channels": 3,
56
+ "static_width": 512,
57
+ "task": "semantic-segmentation",
58
+ "tensor_parallel_size": 1
59
+ },
60
+ "norm_num_groups": 32,
61
+ "out_channels": 3,
62
+ "sample_size": 768,
63
+ "scaling_factor": 0.18215,
64
+ "shift_factor": null,
65
+ "transformers_version": null,
66
+ "up_block_types": [
67
+ "UpDecoderBlock2D",
68
+ "UpDecoderBlock2D",
69
+ "UpDecoderBlock2D",
70
+ "UpDecoderBlock2D"
71
+ ],
72
+ "use_post_quant_conv": true,
73
+ "use_quant_conv": true
74
+ }
neuronxcc-2.17.194.0+d312836f/MODULE_ee34e363190e7c6310a6/vae_encoder/model.neuron ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:088f4298499fc3999bb0ccd457f4cb0c1dd24d93255b92dcc3b32929826bd7dd
3
+ size 225662483