skx618 commited on
Commit
d7adcab
·
verified ·
1 Parent(s): 3e7ccea

Publish in ModelOpt's unified Hugging Face layout

Browse files

Re-express the calibrated NVFP4 components in NVIDIA Model Optimizer's unified Hugging Face layout: packed E2M1 values under `.weight`, the FP32 weight scale as `.weight_scale_2`, the static activation scale as `.input_scale` (fp32(amax)/2688), and a `quantization_config` (`quant_method: modelopt`, `quant_algo: NVFP4`) in each quantized component's config.json. Values and scales are unchanged bit for bit; the packed layout's manifests are removed. Produced by UniServe's scripts/convert_h3_modelopt_checkpoint.py (commit c002a175).

README.md CHANGED
@@ -19,15 +19,14 @@ tags:
19
 
20
  > This UniServe deployment checkpoint applies calibrated NVIDIA Model Optimizer NVFP4 W4A4 to the fused gate/up and down projections in all 50 H3 denoiser gated MLPs and to the qkv, attention-output, fused gate/up, and down projections in all 36 Video VAE decoder Transformer blocks; denoiser attention, the BF16 text encoder, token refiner, modulation, normalization, latent and conditioning boundaries, convolutions, residual/affine paths, and audio decoder remain BF16/FP32. Calibration used 1,000 prompts with complete four-forward denoising trajectories and deployment-distribution VAE decoding, ModelOpt commit `6a4b3f147e14a6fec690fedbced8df402344085d`, E2M1 values, K16 FP8 E4M3 block scales, FP32 tensor scales, and calibration-derived activation tensor scales.
21
 
22
- This repository is a packed UniServe checkpoint derived from [FastVideo/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree](https://huggingface.co/FastVideo/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree). It generates synchronized video and audio with four transformer forwards and retains the source checkpoint's VSA-H3 execution contract.
23
 
24
  ## Run with UniServe
25
 
26
- UniServe detects and validates `modelopt_manifest.json` automatically. The packed checkpoint does not require NVIDIA Model Optimizer at deployment time and rejects runtime precision presets or component overrides that would conflict with its calibrated weights and scales.
27
 
28
  ```bash
29
  uniserve serve skx618/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree-NVFP4 \
30
- --model-description minimax-h3 \
31
  --served-model-name MiniMax-H3 \
32
  --dtype bfloat16
33
  ```
@@ -38,7 +37,7 @@ The checkpoint was validated on four NVIDIA B200 GPUs. The GPU count must divide
38
 
39
  Packed values and scales passed 39/39 projection conformance checks against ModelOpt fake quantization. Candidate A reduced measured end-to-end latency by 16.3–28.4% and peak aggregate GPU memory by 12.8–17.7% versus the BF16/FP16 reference over the fixed benchmark points. Same-seed metrics measure trajectory drift rather than absolute perceptual quality; the frozen blinded human review has not been completed, so this checkpoint is not claimed to be perceptually equivalent to the reference.
40
 
41
- A separate executable-checkpoint review generated five 5-second, 1344×768, 24-fps synchronized audio/video samples from this packed checkpoint and five same-prompt, same-seed quality-reference samples successfully. Across those five pairs, mean RGB PSNR was 13.81 dB, SSIM 0.4645, AlexNet LPIPS 0.4729, and audio log-mel cosine 0.9260. These prompts came from the calibration-source dataset, so the measurements describe trajectory drift and generation viability, not independent held-out quality acceptance.
42
 
43
  ## License
44
 
 
19
 
20
  > This UniServe deployment checkpoint applies calibrated NVIDIA Model Optimizer NVFP4 W4A4 to the fused gate/up and down projections in all 50 H3 denoiser gated MLPs and to the qkv, attention-output, fused gate/up, and down projections in all 36 Video VAE decoder Transformer blocks; denoiser attention, the BF16 text encoder, token refiner, modulation, normalization, latent and conditioning boundaries, convolutions, residual/affine paths, and audio decoder remain BF16/FP32. Calibration used 1,000 prompts with complete four-forward denoising trajectories and deployment-distribution VAE decoding, ModelOpt commit `6a4b3f147e14a6fec690fedbced8df402344085d`, E2M1 values, K16 FP8 E4M3 block scales, FP32 tensor scales, and calibration-derived activation tensor scales.
21
 
22
+ This repository is a UniServe deployment checkpoint in NVIDIA Model Optimizer's unified Hugging Face layout, derived from [FastVideo/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree](https://huggingface.co/FastVideo/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree). It generates synchronized video and audio with four transformer forwards and retains the source checkpoint's VSA-H3 execution contract.
23
 
24
  ## Run with UniServe
25
 
26
+ The quantized `transformer` and `vae` components store packed E2M1 values in `.weight`, K16 FP8 E4M3 block scales in `.weight_scale`, the FP32 weight scale in `.weight_scale_2` and the static activation scale in `.input_scale`, and each declares a `quantization_config` (`quant_method: modelopt`, `quant_algo: NVFP4`) in its `config.json`; layers listed under `ignore` stay dense. UniServe detects this configuration automatically. The checkpoint does not require NVIDIA Model Optimizer at deployment time and rejects runtime precision presets or component overrides that would conflict with its calibrated weights and scales. The tensors are the calibrated values and scales of this repository's earlier packed layout, renamed and described bit for bit, so the validation below applies unchanged.
27
 
28
  ```bash
29
  uniserve serve skx618/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree-NVFP4 \
 
30
  --served-model-name MiniMax-H3 \
31
  --dtype bfloat16
32
  ```
 
37
 
38
  Packed values and scales passed 39/39 projection conformance checks against ModelOpt fake quantization. Candidate A reduced measured end-to-end latency by 16.3–28.4% and peak aggregate GPU memory by 12.8–17.7% versus the BF16/FP16 reference over the fixed benchmark points. Same-seed metrics measure trajectory drift rather than absolute perceptual quality; the frozen blinded human review has not been completed, so this checkpoint is not claimed to be perceptually equivalent to the reference.
39
 
40
+ A separate executable-checkpoint review generated five 5-second, 1344×768, 24-fps synchronized audio/video samples from this checkpoint and five same-prompt, same-seed quality-reference samples successfully. Across those five pairs, mean RGB PSNR was 13.81 dB, SSIM 0.4645, AlexNet LPIPS 0.4729, and audio log-mel cosine 0.9260. These prompts came from the calibration-source dataset, so the measurements describe trajectory drift and generation viability, not independent held-out quality acceptance.
41
 
42
  ## License
43
 
modelopt_manifest.json DELETED
@@ -1,1280 +0,0 @@
1
- {
2
- "calibration": {
3
- "activation_block_scale": "runtime dynamic K16 encoding",
4
- "activation_global_scale": "ModelOpt NVFP4 activation headroom calibration",
5
- "denoising_forwards_per_record": 4,
6
- "records": 1000,
7
- "sample_sha256": "5baf06f8a8a811d4021efd4b372e984d3a8a5ea71912b150d80ace804046dd90",
8
- "weight_scale": "static MSE, complete E4M3 scale sweep"
9
- },
10
- "components": {
11
- "denoiser": {
12
- "enabled": true,
13
- "excluded_patterns": [
14
- "attention",
15
- "token_refiner",
16
- "adaln",
17
- "norm",
18
- "latent_heads"
19
- ],
20
- "modules": {
21
- "denoiser.transformer.layers.0.mlp.down": {
22
- "activation_amax": 602.3152465820312
23
- },
24
- "denoiser.transformer.layers.0.mlp.gate_up.projections.gate": {
25
- "activation_amax": 485.0114440917969
26
- },
27
- "denoiser.transformer.layers.0.mlp.gate_up.projections.up": {
28
- "activation_amax": 485.0114440917969
29
- },
30
- "denoiser.transformer.layers.1.mlp.down": {
31
- "activation_amax": 390.5531311035156
32
- },
33
- "denoiser.transformer.layers.1.mlp.gate_up.projections.gate": {
34
- "activation_amax": 485.0114440917969
35
- },
36
- "denoiser.transformer.layers.1.mlp.gate_up.projections.up": {
37
- "activation_amax": 485.0114440917969
38
- },
39
- "denoiser.transformer.layers.10.mlp.down": {
40
- "activation_amax": 5254.6357421875
41
- },
42
- "denoiser.transformer.layers.10.mlp.gate_up.projections.gate": {
43
- "activation_amax": 2743.63916015625
44
- },
45
- "denoiser.transformer.layers.10.mlp.gate_up.projections.up": {
46
- "activation_amax": 2743.63916015625
47
- },
48
- "denoiser.transformer.layers.11.mlp.down": {
49
- "activation_amax": 5855.6962890625
50
- },
51
- "denoiser.transformer.layers.11.mlp.gate_up.projections.gate": {
52
- "activation_amax": 2743.63916015625
53
- },
54
- "denoiser.transformer.layers.11.mlp.gate_up.projections.up": {
55
- "activation_amax": 2743.63916015625
56
- },
57
- "denoiser.transformer.layers.12.mlp.down": {
58
- "activation_amax": 5855.6962890625
59
- },
60
- "denoiser.transformer.layers.12.mlp.gate_up.projections.gate": {
61
- "activation_amax": 2743.63916015625
62
- },
63
- "denoiser.transformer.layers.12.mlp.gate_up.projections.up": {
64
- "activation_amax": 2743.63916015625
65
- },
66
- "denoiser.transformer.layers.13.mlp.down": {
67
- "activation_amax": 5855.6962890625
68
- },
69
- "denoiser.transformer.layers.13.mlp.gate_up.projections.gate": {
70
- "activation_amax": 3057.47509765625
71
- },
72
- "denoiser.transformer.layers.13.mlp.gate_up.projections.up": {
73
- "activation_amax": 3057.47509765625
74
- },
75
- "denoiser.transformer.layers.14.mlp.down": {
76
- "activation_amax": 5254.6357421875
77
- },
78
- "denoiser.transformer.layers.14.mlp.gate_up.projections.gate": {
79
- "activation_amax": 2743.63916015625
80
- },
81
- "denoiser.transformer.layers.14.mlp.gate_up.projections.up": {
82
- "activation_amax": 2743.63916015625
83
- },
84
- "denoiser.transformer.layers.15.mlp.down": {
85
- "activation_amax": 5855.6962890625
86
- },
87
- "denoiser.transformer.layers.15.mlp.gate_up.projections.gate": {
88
- "activation_amax": 2743.63916015625
89
- },
90
- "denoiser.transformer.layers.15.mlp.gate_up.projections.up": {
91
- "activation_amax": 2743.63916015625
92
- },
93
- "denoiser.transformer.layers.16.mlp.down": {
94
- "activation_amax": 5855.6962890625
95
- },
96
- "denoiser.transformer.layers.16.mlp.gate_up.projections.gate": {
97
- "activation_amax": 3407.209716796875
98
- },
99
- "denoiser.transformer.layers.16.mlp.gate_up.projections.up": {
100
- "activation_amax": 3407.209716796875
101
- },
102
- "denoiser.transformer.layers.17.mlp.down": {
103
- "activation_amax": 9030.7177734375
104
- },
105
- "denoiser.transformer.layers.17.mlp.gate_up.projections.gate": {
106
- "activation_amax": 3057.47509765625
107
- },
108
- "denoiser.transformer.layers.17.mlp.gate_up.projections.up": {
109
- "activation_amax": 3057.47509765625
110
- },
111
- "denoiser.transformer.layers.18.mlp.down": {
112
- "activation_amax": 9030.7177734375
113
- },
114
- "denoiser.transformer.layers.18.mlp.gate_up.projections.gate": {
115
- "activation_amax": 2743.63916015625
116
- },
117
- "denoiser.transformer.layers.18.mlp.gate_up.projections.up": {
118
- "activation_amax": 2743.63916015625
119
- },
120
- "denoiser.transformer.layers.19.mlp.down": {
121
- "activation_amax": 9030.7177734375
122
- },
123
- "denoiser.transformer.layers.19.mlp.gate_up.projections.gate": {
124
- "activation_amax": 2743.63916015625
125
- },
126
- "denoiser.transformer.layers.19.mlp.gate_up.projections.up": {
127
- "activation_amax": 2743.63916015625
128
- },
129
- "denoiser.transformer.layers.2.mlp.down": {
130
- "activation_amax": 1779.0299072265625
131
- },
132
- "denoiser.transformer.layers.2.mlp.gate_up.projections.gate": {
133
- "activation_amax": 928.8970947265625
134
- },
135
- "denoiser.transformer.layers.2.mlp.gate_up.projections.up": {
136
- "activation_amax": 928.8970947265625
137
- },
138
- "denoiser.transformer.layers.20.mlp.down": {
139
- "activation_amax": 10063.712890625
140
- },
141
- "denoiser.transformer.layers.20.mlp.gate_up.projections.gate": {
142
- "activation_amax": 3057.47509765625
143
- },
144
- "denoiser.transformer.layers.20.mlp.gate_up.projections.up": {
145
- "activation_amax": 3057.47509765625
146
- },
147
- "denoiser.transformer.layers.21.mlp.down": {
148
- "activation_amax": 9030.7177734375
149
- },
150
- "denoiser.transformer.layers.21.mlp.gate_up.projections.gate": {
151
- "activation_amax": 3057.47509765625
152
- },
153
- "denoiser.transformer.layers.21.mlp.gate_up.projections.up": {
154
- "activation_amax": 3057.47509765625
155
- },
156
- "denoiser.transformer.layers.22.mlp.down": {
157
- "activation_amax": 9030.7177734375
158
- },
159
- "denoiser.transformer.layers.22.mlp.gate_up.projections.gate": {
160
- "activation_amax": 3407.209716796875
161
- },
162
- "denoiser.transformer.layers.22.mlp.gate_up.projections.up": {
163
- "activation_amax": 3407.209716796875
164
- },
165
- "denoiser.transformer.layers.23.mlp.down": {
166
- "activation_amax": 11214.8681640625
167
- },
168
- "denoiser.transformer.layers.23.mlp.gate_up.projections.gate": {
169
- "activation_amax": 3407.209716796875
170
- },
171
- "denoiser.transformer.layers.23.mlp.gate_up.projections.up": {
172
- "activation_amax": 3407.209716796875
173
- },
174
- "denoiser.transformer.layers.24.mlp.down": {
175
- "activation_amax": 12497.7001953125
176
- },
177
- "denoiser.transformer.layers.24.mlp.gate_up.projections.gate": {
178
- "activation_amax": 3796.94921875
179
- },
180
- "denoiser.transformer.layers.24.mlp.gate_up.projections.up": {
181
- "activation_amax": 3796.94921875
182
- },
183
- "denoiser.transformer.layers.25.mlp.down": {
184
- "activation_amax": 17295.69140625
185
- },
186
- "denoiser.transformer.layers.25.mlp.gate_up.projections.gate": {
187
- "activation_amax": 3796.94921875
188
- },
189
- "denoiser.transformer.layers.25.mlp.gate_up.projections.up": {
190
- "activation_amax": 3796.94921875
191
- },
192
- "denoiser.transformer.layers.26.mlp.down": {
193
- "activation_amax": 21478.7890625
194
- },
195
- "denoiser.transformer.layers.26.mlp.gate_up.projections.gate": {
196
- "activation_amax": 3796.94921875
197
- },
198
- "denoiser.transformer.layers.26.mlp.gate_up.projections.up": {
199
- "activation_amax": 3796.94921875
200
- },
201
- "denoiser.transformer.layers.27.mlp.down": {
202
- "activation_amax": 26673.6015625
203
- },
204
- "denoiser.transformer.layers.27.mlp.gate_up.projections.gate": {
205
- "activation_amax": 4715.27099609375
206
- },
207
- "denoiser.transformer.layers.27.mlp.gate_up.projections.up": {
208
- "activation_amax": 4715.27099609375
209
- },
210
- "denoiser.transformer.layers.28.mlp.down": {
211
- "activation_amax": 45841.765625
212
- },
213
- "denoiser.transformer.layers.28.mlp.gate_up.projections.gate": {
214
- "activation_amax": 4715.27099609375
215
- },
216
- "denoiser.transformer.layers.28.mlp.gate_up.projections.up": {
217
- "activation_amax": 4715.27099609375
218
- },
219
- "denoiser.transformer.layers.29.mlp.down": {
220
- "activation_amax": 51085.45703125
221
- },
222
- "denoiser.transformer.layers.29.mlp.gate_up.projections.gate": {
223
- "activation_amax": 4715.27099609375
224
- },
225
- "denoiser.transformer.layers.29.mlp.gate_up.projections.up": {
226
- "activation_amax": 4715.27099609375
227
- },
228
- "denoiser.transformer.layers.3.mlp.down": {
229
- "activation_amax": 1596.4205322265625
230
- },
231
- "denoiser.transformer.layers.3.mlp.gate_up.projections.gate": {
232
- "activation_amax": 833.550048828125
233
- },
234
- "denoiser.transformer.layers.3.mlp.gate_up.projections.up": {
235
- "activation_amax": 833.550048828125
236
- },
237
- "denoiser.transformer.layers.30.mlp.down": {
238
- "activation_amax": 70697.6640625
239
- },
240
- "denoiser.transformer.layers.30.mlp.gate_up.projections.gate": {
241
- "activation_amax": 5254.6357421875
242
- },
243
- "denoiser.transformer.layers.30.mlp.gate_up.projections.up": {
244
- "activation_amax": 5254.6357421875
245
- },
246
- "denoiser.transformer.layers.31.mlp.down": {
247
- "activation_amax": 97839.203125
248
- },
249
- "denoiser.transformer.layers.31.mlp.gate_up.projections.gate": {
250
- "activation_amax": 5254.6357421875
251
- },
252
- "denoiser.transformer.layers.31.mlp.gate_up.projections.up": {
253
- "activation_amax": 5254.6357421875
254
- },
255
- "denoiser.transformer.layers.32.mlp.down": {
256
- "activation_amax": 135400.640625
257
- },
258
- "denoiser.transformer.layers.32.mlp.gate_up.projections.gate": {
259
- "activation_amax": 5855.6962890625
260
- },
261
- "denoiser.transformer.layers.32.mlp.gate_up.projections.up": {
262
- "activation_amax": 5855.6962890625
263
- },
264
- "denoiser.transformer.layers.33.mlp.down": {
265
- "activation_amax": 168148.34375
266
- },
267
- "denoiser.transformer.layers.33.mlp.gate_up.projections.gate": {
268
- "activation_amax": 5855.6962890625
269
- },
270
- "denoiser.transformer.layers.33.mlp.gate_up.projections.up": {
271
- "activation_amax": 5855.6962890625
272
- },
273
- "denoiser.transformer.layers.34.mlp.down": {
274
- "activation_amax": 187382.28125
275
- },
276
- "denoiser.transformer.layers.34.mlp.gate_up.projections.gate": {
277
- "activation_amax": 6525.51025390625
278
- },
279
- "denoiser.transformer.layers.34.mlp.gate_up.projections.up": {
280
- "activation_amax": 6525.51025390625
281
- },
282
- "denoiser.transformer.layers.35.mlp.down": {
283
- "activation_amax": 232702.140625
284
- },
285
- "denoiser.transformer.layers.35.mlp.gate_up.projections.gate": {
286
- "activation_amax": 7271.94189453125
287
- },
288
- "denoiser.transformer.layers.35.mlp.gate_up.projections.up": {
289
- "activation_amax": 7271.94189453125
290
- },
291
- "denoiser.transformer.layers.36.mlp.down": {
292
- "activation_amax": 232702.140625
293
- },
294
- "denoiser.transformer.layers.36.mlp.gate_up.projections.gate": {
295
- "activation_amax": 7271.94189453125
296
- },
297
- "denoiser.transformer.layers.36.mlp.gate_up.projections.up": {
298
- "activation_amax": 7271.94189453125
299
- },
300
- "denoiser.transformer.layers.37.mlp.down": {
301
- "activation_amax": 232702.140625
302
- },
303
- "denoiser.transformer.layers.37.mlp.gate_up.projections.gate": {
304
- "activation_amax": 6525.51025390625
305
- },
306
- "denoiser.transformer.layers.37.mlp.gate_up.projections.up": {
307
- "activation_amax": 6525.51025390625
308
- },
309
- "denoiser.transformer.layers.38.mlp.down": {
310
- "activation_amax": 288982.96875
311
- },
312
- "denoiser.transformer.layers.38.mlp.gate_up.projections.gate": {
313
- "activation_amax": 6525.51025390625
314
- },
315
- "denoiser.transformer.layers.38.mlp.gate_up.projections.up": {
316
- "activation_amax": 6525.51025390625
317
- },
318
- "denoiser.transformer.layers.39.mlp.down": {
319
- "activation_amax": 399926.40625
320
- },
321
- "denoiser.transformer.layers.39.mlp.gate_up.projections.gate": {
322
- "activation_amax": 7271.94189453125
323
- },
324
- "denoiser.transformer.layers.39.mlp.gate_up.projections.up": {
325
- "activation_amax": 7271.94189453125
326
- },
327
- "denoiser.transformer.layers.4.mlp.down": {
328
- "activation_amax": 1432.55517578125
329
- },
330
- "denoiser.transformer.layers.4.mlp.gate_up.projections.gate": {
331
- "activation_amax": 833.550048828125
332
- },
333
- "denoiser.transformer.layers.4.mlp.gate_up.projections.up": {
334
- "activation_amax": 833.550048828125
335
- },
336
- "denoiser.transformer.layers.40.mlp.down": {
337
- "activation_amax": 445672.6875
338
- },
339
- "denoiser.transformer.layers.40.mlp.gate_up.projections.gate": {
340
- "activation_amax": 8103.755859375
341
- },
342
- "denoiser.transformer.layers.40.mlp.gate_up.projections.up": {
343
- "activation_amax": 8103.755859375
344
- },
345
- "denoiser.transformer.layers.41.mlp.down": {
346
- "activation_amax": 496651.71875
347
- },
348
- "denoiser.transformer.layers.41.mlp.gate_up.projections.gate": {
349
- "activation_amax": 8103.755859375
350
- },
351
- "denoiser.transformer.layers.41.mlp.gate_up.projections.up": {
352
- "activation_amax": 8103.755859375
353
- },
354
- "denoiser.transformer.layers.42.mlp.down": {
355
- "activation_amax": 616770.8125
356
- },
357
- "denoiser.transformer.layers.42.mlp.gate_up.projections.gate": {
358
- "activation_amax": 9030.7177734375
359
- },
360
- "denoiser.transformer.layers.42.mlp.gate_up.projections.up": {
361
- "activation_amax": 9030.7177734375
362
- },
363
- "denoiser.transformer.layers.43.mlp.down": {
364
- "activation_amax": 616770.8125
365
- },
366
- "denoiser.transformer.layers.43.mlp.gate_up.projections.gate": {
367
- "activation_amax": 9030.7177734375
368
- },
369
- "denoiser.transformer.layers.43.mlp.gate_up.projections.up": {
370
- "activation_amax": 9030.7177734375
371
- },
372
- "denoiser.transformer.layers.44.mlp.down": {
373
- "activation_amax": 687321.25
374
- },
375
- "denoiser.transformer.layers.44.mlp.gate_up.projections.gate": {
376
- "activation_amax": 10063.712890625
377
- },
378
- "denoiser.transformer.layers.44.mlp.gate_up.projections.up": {
379
- "activation_amax": 10063.712890625
380
- },
381
- "denoiser.transformer.layers.45.mlp.down": {
382
- "activation_amax": 853555.25
383
- },
384
- "denoiser.transformer.layers.45.mlp.gate_up.projections.gate": {
385
- "activation_amax": 11214.8681640625
386
- },
387
- "denoiser.transformer.layers.45.mlp.gate_up.projections.up": {
388
- "activation_amax": 11214.8681640625
389
- },
390
- "denoiser.transformer.layers.46.mlp.down": {
391
- "activation_amax": 951190.625
392
- },
393
- "denoiser.transformer.layers.46.mlp.gate_up.projections.gate": {
394
- "activation_amax": 12497.7001953125
395
- },
396
- "denoiser.transformer.layers.46.mlp.gate_up.projections.up": {
397
- "activation_amax": 12497.7001953125
398
- },
399
- "denoiser.transformer.layers.47.mlp.down": {
400
- "activation_amax": 1316362.125
401
- },
402
- "denoiser.transformer.layers.47.mlp.gate_up.projections.gate": {
403
- "activation_amax": 13927.271484375
404
- },
405
- "denoiser.transformer.layers.47.mlp.gate_up.projections.up": {
406
- "activation_amax": 13927.271484375
407
- },
408
- "denoiser.transformer.layers.48.mlp.down": {
409
- "activation_amax": 2030107.875
410
- },
411
- "denoiser.transformer.layers.48.mlp.gate_up.projections.gate": {
412
- "activation_amax": 17295.69140625
413
- },
414
- "denoiser.transformer.layers.48.mlp.gate_up.projections.up": {
415
- "activation_amax": 17295.69140625
416
- },
417
- "denoiser.transformer.layers.49.mlp.down": {
418
- "activation_amax": 2809486.5
419
- },
420
- "denoiser.transformer.layers.49.mlp.gate_up.projections.gate": {
421
- "activation_amax": 19274.087890625
422
- },
423
- "denoiser.transformer.layers.49.mlp.gate_up.projections.up": {
424
- "activation_amax": 19274.087890625
425
- },
426
- "denoiser.transformer.layers.5.mlp.down": {
427
- "activation_amax": 5254.6357421875
428
- },
429
- "denoiser.transformer.layers.5.mlp.gate_up.projections.gate": {
430
- "activation_amax": 2209.302001953125
431
- },
432
- "denoiser.transformer.layers.5.mlp.gate_up.projections.up": {
433
- "activation_amax": 2209.302001953125
434
- },
435
- "denoiser.transformer.layers.6.mlp.down": {
436
- "activation_amax": 5254.6357421875
437
- },
438
- "denoiser.transformer.layers.6.mlp.gate_up.projections.gate": {
439
- "activation_amax": 2209.302001953125
440
- },
441
- "denoiser.transformer.layers.6.mlp.gate_up.projections.up": {
442
- "activation_amax": 2209.302001953125
443
- },
444
- "denoiser.transformer.layers.7.mlp.down": {
445
- "activation_amax": 4231.27001953125
446
- },
447
- "denoiser.transformer.layers.7.mlp.gate_up.projections.gate": {
448
- "activation_amax": 2462.01708984375
449
- },
450
- "denoiser.transformer.layers.7.mlp.gate_up.projections.up": {
451
- "activation_amax": 2462.01708984375
452
- },
453
- "denoiser.transformer.layers.8.mlp.down": {
454
- "activation_amax": 5254.6357421875
455
- },
456
- "denoiser.transformer.layers.8.mlp.gate_up.projections.gate": {
457
- "activation_amax": 2462.01708984375
458
- },
459
- "denoiser.transformer.layers.8.mlp.gate_up.projections.up": {
460
- "activation_amax": 2462.01708984375
461
- },
462
- "denoiser.transformer.layers.9.mlp.down": {
463
- "activation_amax": 5254.6357421875
464
- },
465
- "denoiser.transformer.layers.9.mlp.gate_up.projections.gate": {
466
- "activation_amax": 2743.63916015625
467
- },
468
- "denoiser.transformer.layers.9.mlp.gate_up.projections.up": {
469
- "activation_amax": 2743.63916015625
470
- }
471
- },
472
- "quantized_patterns": [
473
- "transformer_blocks.*.ff.net.0.proj",
474
- "transformer_blocks.*.ff.net.2"
475
- ]
476
- },
477
- "text_encoder": {
478
- "enabled": false,
479
- "excluded_patterns": [
480
- "attention",
481
- "embedding",
482
- "norm",
483
- "rope",
484
- "softmax",
485
- "checkpoint_layers.50:64",
486
- "conditioning_output"
487
- ],
488
- "modules": {},
489
- "quantized_patterns": [
490
- "retained_layers.0:50.mlp.{gate,up,down}"
491
- ]
492
- },
493
- "video_vae": {
494
- "enabled": true,
495
- "excluded_patterns": [
496
- "post_quant_conv",
497
- "decoder_boundaries",
498
- "conv",
499
- "norm",
500
- "rope",
501
- "softmax",
502
- "residual",
503
- "affine",
504
- "pixel_postprocess"
505
- ],
506
- "modules": {
507
- "video_decoder.decoder.decoder.decoder.layers.0.mlp.down": {
508
- "activation_amax": 540.4903564453125
509
- },
510
- "video_decoder.decoder.decoder.decoder.layers.0.mlp.gate_up.projections.gate": {
511
- "activation_amax": 3796.94921875
512
- },
513
- "video_decoder.decoder.decoder.decoder.layers.0.mlp.gate_up.projections.up": {
514
- "activation_amax": 3796.94921875
515
- },
516
- "video_decoder.decoder.decoder.decoder.layers.0.output": {
517
- "activation_amax": 4231.27001953125
518
- },
519
- "video_decoder.decoder.decoder.decoder.layers.0.qkv.projections.k": {
520
- "activation_amax": 5855.6962890625
521
- },
522
- "video_decoder.decoder.decoder.decoder.layers.0.qkv.projections.q": {
523
- "activation_amax": 5855.6962890625
524
- },
525
- "video_decoder.decoder.decoder.decoder.layers.0.qkv.projections.v": {
526
- "activation_amax": 5855.6962890625
527
- },
528
- "video_decoder.decoder.decoder.decoder.layers.1.mlp.down": {
529
- "activation_amax": 833.550048828125
530
- },
531
- "video_decoder.decoder.decoder.decoder.layers.1.mlp.gate_up.projections.gate": {
532
- "activation_amax": 3057.47509765625
533
- },
534
- "video_decoder.decoder.decoder.decoder.layers.1.mlp.gate_up.projections.up": {
535
- "activation_amax": 3057.47509765625
536
- },
537
- "video_decoder.decoder.decoder.decoder.layers.1.output": {
538
- "activation_amax": 3407.209716796875
539
- },
540
- "video_decoder.decoder.decoder.decoder.layers.1.qkv.projections.k": {
541
- "activation_amax": 3407.209716796875
542
- },
543
- "video_decoder.decoder.decoder.decoder.layers.1.qkv.projections.q": {
544
- "activation_amax": 3407.209716796875
545
- },
546
- "video_decoder.decoder.decoder.decoder.layers.1.qkv.projections.v": {
547
- "activation_amax": 3407.209716796875
548
- },
549
- "video_decoder.decoder.decoder.decoder.layers.10.mlp.down": {
550
- "activation_amax": 1982.5272216796875
551
- },
552
- "video_decoder.decoder.decoder.decoder.layers.10.mlp.gate_up.projections.gate": {
553
- "activation_amax": 9030.7177734375
554
- },
555
- "video_decoder.decoder.decoder.decoder.layers.10.mlp.gate_up.projections.up": {
556
- "activation_amax": 9030.7177734375
557
- },
558
- "video_decoder.decoder.decoder.decoder.layers.10.output": {
559
- "activation_amax": 4715.27099609375
560
- },
561
- "video_decoder.decoder.decoder.decoder.layers.10.qkv.projections.k": {
562
- "activation_amax": 10063.712890625
563
- },
564
- "video_decoder.decoder.decoder.decoder.layers.10.qkv.projections.q": {
565
- "activation_amax": 10063.712890625
566
- },
567
- "video_decoder.decoder.decoder.decoder.layers.10.qkv.projections.v": {
568
- "activation_amax": 10063.712890625
569
- },
570
- "video_decoder.decoder.decoder.decoder.layers.11.mlp.down": {
571
- "activation_amax": 1982.5272216796875
572
- },
573
- "video_decoder.decoder.decoder.decoder.layers.11.mlp.gate_up.projections.gate": {
574
- "activation_amax": 9030.7177734375
575
- },
576
- "video_decoder.decoder.decoder.decoder.layers.11.mlp.gate_up.projections.up": {
577
- "activation_amax": 9030.7177734375
578
- },
579
- "video_decoder.decoder.decoder.decoder.layers.11.output": {
580
- "activation_amax": 4231.27001953125
581
- },
582
- "video_decoder.decoder.decoder.decoder.layers.11.qkv.projections.k": {
583
- "activation_amax": 10063.712890625
584
- },
585
- "video_decoder.decoder.decoder.decoder.layers.11.qkv.projections.q": {
586
- "activation_amax": 10063.712890625
587
- },
588
- "video_decoder.decoder.decoder.decoder.layers.11.qkv.projections.v": {
589
- "activation_amax": 10063.712890625
590
- },
591
- "video_decoder.decoder.decoder.decoder.layers.12.mlp.down": {
592
- "activation_amax": 1982.5272216796875
593
- },
594
- "video_decoder.decoder.decoder.decoder.layers.12.mlp.gate_up.projections.gate": {
595
- "activation_amax": 9030.7177734375
596
- },
597
- "video_decoder.decoder.decoder.decoder.layers.12.mlp.gate_up.projections.up": {
598
- "activation_amax": 9030.7177734375
599
- },
600
- "video_decoder.decoder.decoder.decoder.layers.12.output": {
601
- "activation_amax": 3407.209716796875
602
- },
603
- "video_decoder.decoder.decoder.decoder.layers.12.qkv.projections.k": {
604
- "activation_amax": 10063.712890625
605
- },
606
- "video_decoder.decoder.decoder.decoder.layers.12.qkv.projections.q": {
607
- "activation_amax": 10063.712890625
608
- },
609
- "video_decoder.decoder.decoder.decoder.layers.12.qkv.projections.v": {
610
- "activation_amax": 10063.712890625
611
- },
612
- "video_decoder.decoder.decoder.decoder.layers.13.mlp.down": {
613
- "activation_amax": 1779.0299072265625
614
- },
615
- "video_decoder.decoder.decoder.decoder.layers.13.mlp.gate_up.projections.gate": {
616
- "activation_amax": 9030.7177734375
617
- },
618
- "video_decoder.decoder.decoder.decoder.layers.13.mlp.gate_up.projections.up": {
619
- "activation_amax": 9030.7177734375
620
- },
621
- "video_decoder.decoder.decoder.decoder.layers.13.output": {
622
- "activation_amax": 3407.209716796875
623
- },
624
- "video_decoder.decoder.decoder.decoder.layers.13.qkv.projections.k": {
625
- "activation_amax": 9030.7177734375
626
- },
627
- "video_decoder.decoder.decoder.decoder.layers.13.qkv.projections.q": {
628
- "activation_amax": 9030.7177734375
629
- },
630
- "video_decoder.decoder.decoder.decoder.layers.13.qkv.projections.v": {
631
- "activation_amax": 9030.7177734375
632
- },
633
- "video_decoder.decoder.decoder.decoder.layers.14.mlp.down": {
634
- "activation_amax": 1779.0299072265625
635
- },
636
- "video_decoder.decoder.decoder.decoder.layers.14.mlp.gate_up.projections.gate": {
637
- "activation_amax": 8103.755859375
638
- },
639
- "video_decoder.decoder.decoder.decoder.layers.14.mlp.gate_up.projections.up": {
640
- "activation_amax": 8103.755859375
641
- },
642
- "video_decoder.decoder.decoder.decoder.layers.14.output": {
643
- "activation_amax": 2743.63916015625
644
- },
645
- "video_decoder.decoder.decoder.decoder.layers.14.qkv.projections.k": {
646
- "activation_amax": 9030.7177734375
647
- },
648
- "video_decoder.decoder.decoder.decoder.layers.14.qkv.projections.q": {
649
- "activation_amax": 9030.7177734375
650
- },
651
- "video_decoder.decoder.decoder.decoder.layers.14.qkv.projections.v": {
652
- "activation_amax": 9030.7177734375
653
- },
654
- "video_decoder.decoder.decoder.decoder.layers.15.mlp.down": {
655
- "activation_amax": 1779.0299072265625
656
- },
657
- "video_decoder.decoder.decoder.decoder.layers.15.mlp.gate_up.projections.gate": {
658
- "activation_amax": 8103.755859375
659
- },
660
- "video_decoder.decoder.decoder.decoder.layers.15.mlp.gate_up.projections.up": {
661
- "activation_amax": 8103.755859375
662
- },
663
- "video_decoder.decoder.decoder.decoder.layers.15.output": {
664
- "activation_amax": 2462.01708984375
665
- },
666
- "video_decoder.decoder.decoder.decoder.layers.15.qkv.projections.k": {
667
- "activation_amax": 9030.7177734375
668
- },
669
- "video_decoder.decoder.decoder.decoder.layers.15.qkv.projections.q": {
670
- "activation_amax": 9030.7177734375
671
- },
672
- "video_decoder.decoder.decoder.decoder.layers.15.qkv.projections.v": {
673
- "activation_amax": 9030.7177734375
674
- },
675
- "video_decoder.decoder.decoder.decoder.layers.16.mlp.down": {
676
- "activation_amax": 1596.4205322265625
677
- },
678
- "video_decoder.decoder.decoder.decoder.layers.16.mlp.gate_up.projections.gate": {
679
- "activation_amax": 7271.94189453125
680
- },
681
- "video_decoder.decoder.decoder.decoder.layers.16.mlp.gate_up.projections.up": {
682
- "activation_amax": 7271.94189453125
683
- },
684
- "video_decoder.decoder.decoder.decoder.layers.16.output": {
685
- "activation_amax": 2743.63916015625
686
- },
687
- "video_decoder.decoder.decoder.decoder.layers.16.qkv.projections.k": {
688
- "activation_amax": 8103.755859375
689
- },
690
- "video_decoder.decoder.decoder.decoder.layers.16.qkv.projections.q": {
691
- "activation_amax": 8103.755859375
692
- },
693
- "video_decoder.decoder.decoder.decoder.layers.16.qkv.projections.v": {
694
- "activation_amax": 8103.755859375
695
- },
696
- "video_decoder.decoder.decoder.decoder.layers.17.mlp.down": {
697
- "activation_amax": 1432.55517578125
698
- },
699
- "video_decoder.decoder.decoder.decoder.layers.17.mlp.gate_up.projections.gate": {
700
- "activation_amax": 7271.94189453125
701
- },
702
- "video_decoder.decoder.decoder.decoder.layers.17.mlp.gate_up.projections.up": {
703
- "activation_amax": 7271.94189453125
704
- },
705
- "video_decoder.decoder.decoder.decoder.layers.17.output": {
706
- "activation_amax": 1982.5272216796875
707
- },
708
- "video_decoder.decoder.decoder.decoder.layers.17.qkv.projections.k": {
709
- "activation_amax": 7271.94189453125
710
- },
711
- "video_decoder.decoder.decoder.decoder.layers.17.qkv.projections.q": {
712
- "activation_amax": 7271.94189453125
713
- },
714
- "video_decoder.decoder.decoder.decoder.layers.17.qkv.projections.v": {
715
- "activation_amax": 7271.94189453125
716
- },
717
- "video_decoder.decoder.decoder.decoder.layers.18.mlp.down": {
718
- "activation_amax": 1035.150634765625
719
- },
720
- "video_decoder.decoder.decoder.decoder.layers.18.mlp.gate_up.projections.gate": {
721
- "activation_amax": 5855.6962890625
722
- },
723
- "video_decoder.decoder.decoder.decoder.layers.18.mlp.gate_up.projections.up": {
724
- "activation_amax": 5855.6962890625
725
- },
726
- "video_decoder.decoder.decoder.decoder.layers.18.output": {
727
- "activation_amax": 3407.209716796875
728
- },
729
- "video_decoder.decoder.decoder.decoder.layers.18.qkv.projections.k": {
730
- "activation_amax": 6525.51025390625
731
- },
732
- "video_decoder.decoder.decoder.decoder.layers.18.qkv.projections.q": {
733
- "activation_amax": 6525.51025390625
734
- },
735
- "video_decoder.decoder.decoder.decoder.layers.18.qkv.projections.v": {
736
- "activation_amax": 6525.51025390625
737
- },
738
- "video_decoder.decoder.decoder.decoder.layers.19.mlp.down": {
739
- "activation_amax": 833.550048828125
740
- },
741
- "video_decoder.decoder.decoder.decoder.layers.19.mlp.gate_up.projections.gate": {
742
- "activation_amax": 5254.6357421875
743
- },
744
- "video_decoder.decoder.decoder.decoder.layers.19.mlp.gate_up.projections.up": {
745
- "activation_amax": 5254.6357421875
746
- },
747
- "video_decoder.decoder.decoder.decoder.layers.19.output": {
748
- "activation_amax": 5855.6962890625
749
- },
750
- "video_decoder.decoder.decoder.decoder.layers.19.qkv.projections.k": {
751
- "activation_amax": 5855.6962890625
752
- },
753
- "video_decoder.decoder.decoder.decoder.layers.19.qkv.projections.q": {
754
- "activation_amax": 5855.6962890625
755
- },
756
- "video_decoder.decoder.decoder.decoder.layers.19.qkv.projections.v": {
757
- "activation_amax": 5855.6962890625
758
- },
759
- "video_decoder.decoder.decoder.decoder.layers.2.mlp.down": {
760
- "activation_amax": 1153.55810546875
761
- },
762
- "video_decoder.decoder.decoder.decoder.layers.2.mlp.gate_up.projections.gate": {
763
- "activation_amax": 4231.27001953125
764
- },
765
- "video_decoder.decoder.decoder.decoder.layers.2.mlp.gate_up.projections.up": {
766
- "activation_amax": 4231.27001953125
767
- },
768
- "video_decoder.decoder.decoder.decoder.layers.2.output": {
769
- "activation_amax": 2743.63916015625
770
- },
771
- "video_decoder.decoder.decoder.decoder.layers.2.qkv.projections.k": {
772
- "activation_amax": 3407.209716796875
773
- },
774
- "video_decoder.decoder.decoder.decoder.layers.2.qkv.projections.q": {
775
- "activation_amax": 3407.209716796875
776
- },
777
- "video_decoder.decoder.decoder.decoder.layers.2.qkv.projections.v": {
778
- "activation_amax": 3407.209716796875
779
- },
780
- "video_decoder.decoder.decoder.decoder.layers.20.mlp.down": {
781
- "activation_amax": 671.212158203125
782
- },
783
- "video_decoder.decoder.decoder.decoder.layers.20.mlp.gate_up.projections.gate": {
784
- "activation_amax": 5254.6357421875
785
- },
786
- "video_decoder.decoder.decoder.decoder.layers.20.mlp.gate_up.projections.up": {
787
- "activation_amax": 5254.6357421875
788
- },
789
- "video_decoder.decoder.decoder.decoder.layers.20.output": {
790
- "activation_amax": 6525.51025390625
791
- },
792
- "video_decoder.decoder.decoder.decoder.layers.20.qkv.projections.k": {
793
- "activation_amax": 5254.6357421875
794
- },
795
- "video_decoder.decoder.decoder.decoder.layers.20.qkv.projections.q": {
796
- "activation_amax": 5254.6357421875
797
- },
798
- "video_decoder.decoder.decoder.decoder.layers.20.qkv.projections.v": {
799
- "activation_amax": 5254.6357421875
800
- },
801
- "video_decoder.decoder.decoder.decoder.layers.21.mlp.down": {
802
- "activation_amax": 540.4903564453125
803
- },
804
- "video_decoder.decoder.decoder.decoder.layers.21.mlp.gate_up.projections.gate": {
805
- "activation_amax": 4715.27099609375
806
- },
807
- "video_decoder.decoder.decoder.decoder.layers.21.mlp.gate_up.projections.up": {
808
- "activation_amax": 4715.27099609375
809
- },
810
- "video_decoder.decoder.decoder.decoder.layers.21.output": {
811
- "activation_amax": 3407.209716796875
812
- },
813
- "video_decoder.decoder.decoder.decoder.layers.21.qkv.projections.k": {
814
- "activation_amax": 4715.27099609375
815
- },
816
- "video_decoder.decoder.decoder.decoder.layers.21.qkv.projections.q": {
817
- "activation_amax": 4715.27099609375
818
- },
819
- "video_decoder.decoder.decoder.decoder.layers.21.qkv.projections.v": {
820
- "activation_amax": 4715.27099609375
821
- },
822
- "video_decoder.decoder.decoder.decoder.layers.22.mlp.down": {
823
- "activation_amax": 390.5531311035156
824
- },
825
- "video_decoder.decoder.decoder.decoder.layers.22.mlp.gate_up.projections.gate": {
826
- "activation_amax": 4231.27001953125
827
- },
828
- "video_decoder.decoder.decoder.decoder.layers.22.mlp.gate_up.projections.up": {
829
- "activation_amax": 4231.27001953125
830
- },
831
- "video_decoder.decoder.decoder.decoder.layers.22.output": {
832
- "activation_amax": 5855.6962890625
833
- },
834
- "video_decoder.decoder.decoder.decoder.layers.22.qkv.projections.k": {
835
- "activation_amax": 4715.27099609375
836
- },
837
- "video_decoder.decoder.decoder.decoder.layers.22.qkv.projections.q": {
838
- "activation_amax": 4715.27099609375
839
- },
840
- "video_decoder.decoder.decoder.decoder.layers.22.qkv.projections.v": {
841
- "activation_amax": 4715.27099609375
842
- },
843
- "video_decoder.decoder.decoder.decoder.layers.23.mlp.down": {
844
- "activation_amax": 390.5531311035156
845
- },
846
- "video_decoder.decoder.decoder.decoder.layers.23.mlp.gate_up.projections.gate": {
847
- "activation_amax": 4231.27001953125
848
- },
849
- "video_decoder.decoder.decoder.decoder.layers.23.mlp.gate_up.projections.up": {
850
- "activation_amax": 4231.27001953125
851
- },
852
- "video_decoder.decoder.decoder.decoder.layers.23.output": {
853
- "activation_amax": 1432.55517578125
854
- },
855
- "video_decoder.decoder.decoder.decoder.layers.23.qkv.projections.k": {
856
- "activation_amax": 4231.27001953125
857
- },
858
- "video_decoder.decoder.decoder.decoder.layers.23.qkv.projections.q": {
859
- "activation_amax": 4231.27001953125
860
- },
861
- "video_decoder.decoder.decoder.decoder.layers.23.qkv.projections.v": {
862
- "activation_amax": 4231.27001953125
863
- },
864
- "video_decoder.decoder.decoder.decoder.layers.24.mlp.down": {
865
- "activation_amax": 390.5531311035156
866
- },
867
- "video_decoder.decoder.decoder.decoder.layers.24.mlp.gate_up.projections.gate": {
868
- "activation_amax": 3796.94921875
869
- },
870
- "video_decoder.decoder.decoder.decoder.layers.24.mlp.gate_up.projections.up": {
871
- "activation_amax": 3796.94921875
872
- },
873
- "video_decoder.decoder.decoder.decoder.layers.24.output": {
874
- "activation_amax": 5254.6357421875
875
- },
876
- "video_decoder.decoder.decoder.decoder.layers.24.qkv.projections.k": {
877
- "activation_amax": 4231.27001953125
878
- },
879
- "video_decoder.decoder.decoder.decoder.layers.24.qkv.projections.q": {
880
- "activation_amax": 4231.27001953125
881
- },
882
- "video_decoder.decoder.decoder.decoder.layers.24.qkv.projections.v": {
883
- "activation_amax": 4231.27001953125
884
- },
885
- "video_decoder.decoder.decoder.decoder.layers.25.mlp.down": {
886
- "activation_amax": 1285.5098876953125
887
- },
888
- "video_decoder.decoder.decoder.decoder.layers.25.mlp.gate_up.projections.gate": {
889
- "activation_amax": 3407.209716796875
890
- },
891
- "video_decoder.decoder.decoder.decoder.layers.25.mlp.gate_up.projections.up": {
892
- "activation_amax": 3407.209716796875
893
- },
894
- "video_decoder.decoder.decoder.decoder.layers.25.output": {
895
- "activation_amax": 1982.5272216796875
896
- },
897
- "video_decoder.decoder.decoder.decoder.layers.25.qkv.projections.k": {
898
- "activation_amax": 3407.209716796875
899
- },
900
- "video_decoder.decoder.decoder.decoder.layers.25.qkv.projections.q": {
901
- "activation_amax": 3407.209716796875
902
- },
903
- "video_decoder.decoder.decoder.decoder.layers.25.qkv.projections.v": {
904
- "activation_amax": 3407.209716796875
905
- },
906
- "video_decoder.decoder.decoder.decoder.layers.26.mlp.down": {
907
- "activation_amax": 350.4646301269531
908
- },
909
- "video_decoder.decoder.decoder.decoder.layers.26.mlp.gate_up.projections.gate": {
910
- "activation_amax": 3407.209716796875
911
- },
912
- "video_decoder.decoder.decoder.decoder.layers.26.mlp.gate_up.projections.up": {
913
- "activation_amax": 3407.209716796875
914
- },
915
- "video_decoder.decoder.decoder.decoder.layers.26.output": {
916
- "activation_amax": 1982.5272216796875
917
- },
918
- "video_decoder.decoder.decoder.decoder.layers.26.qkv.projections.k": {
919
- "activation_amax": 3407.209716796875
920
- },
921
- "video_decoder.decoder.decoder.decoder.layers.26.qkv.projections.q": {
922
- "activation_amax": 3407.209716796875
923
- },
924
- "video_decoder.decoder.decoder.decoder.layers.26.qkv.projections.v": {
925
- "activation_amax": 3407.209716796875
926
- },
927
- "video_decoder.decoder.decoder.decoder.layers.27.mlp.down": {
928
- "activation_amax": 350.4646301269531
929
- },
930
- "video_decoder.decoder.decoder.decoder.layers.27.mlp.gate_up.projections.gate": {
931
- "activation_amax": 3407.209716796875
932
- },
933
- "video_decoder.decoder.decoder.decoder.layers.27.mlp.gate_up.projections.up": {
934
- "activation_amax": 3407.209716796875
935
- },
936
- "video_decoder.decoder.decoder.decoder.layers.27.output": {
937
- "activation_amax": 4715.27099609375
938
- },
939
- "video_decoder.decoder.decoder.decoder.layers.27.qkv.projections.k": {
940
- "activation_amax": 3407.209716796875
941
- },
942
- "video_decoder.decoder.decoder.decoder.layers.27.qkv.projections.q": {
943
- "activation_amax": 3407.209716796875
944
- },
945
- "video_decoder.decoder.decoder.decoder.layers.27.qkv.projections.v": {
946
- "activation_amax": 3407.209716796875
947
- },
948
- "video_decoder.decoder.decoder.decoder.layers.28.mlp.down": {
949
- "activation_amax": 314.49102783203125
950
- },
951
- "video_decoder.decoder.decoder.decoder.layers.28.mlp.gate_up.projections.gate": {
952
- "activation_amax": 3407.209716796875
953
- },
954
- "video_decoder.decoder.decoder.decoder.layers.28.mlp.gate_up.projections.up": {
955
- "activation_amax": 3407.209716796875
956
- },
957
- "video_decoder.decoder.decoder.decoder.layers.28.output": {
958
- "activation_amax": 747.9899291992188
959
- },
960
- "video_decoder.decoder.decoder.decoder.layers.28.qkv.projections.k": {
961
- "activation_amax": 3057.47509765625
962
- },
963
- "video_decoder.decoder.decoder.decoder.layers.28.qkv.projections.q": {
964
- "activation_amax": 3057.47509765625
965
- },
966
- "video_decoder.decoder.decoder.decoder.layers.28.qkv.projections.v": {
967
- "activation_amax": 3057.47509765625
968
- },
969
- "video_decoder.decoder.decoder.decoder.layers.29.mlp.down": {
970
- "activation_amax": 350.4646301269531
971
- },
972
- "video_decoder.decoder.decoder.decoder.layers.29.mlp.gate_up.projections.gate": {
973
- "activation_amax": 3407.209716796875
974
- },
975
- "video_decoder.decoder.decoder.decoder.layers.29.mlp.gate_up.projections.up": {
976
- "activation_amax": 3407.209716796875
977
- },
978
- "video_decoder.decoder.decoder.decoder.layers.29.output": {
979
- "activation_amax": 1035.150634765625
980
- },
981
- "video_decoder.decoder.decoder.decoder.layers.29.qkv.projections.k": {
982
- "activation_amax": 3057.47509765625
983
- },
984
- "video_decoder.decoder.decoder.decoder.layers.29.qkv.projections.q": {
985
- "activation_amax": 3057.47509765625
986
- },
987
- "video_decoder.decoder.decoder.decoder.layers.29.qkv.projections.v": {
988
- "activation_amax": 3057.47509765625
989
- },
990
- "video_decoder.decoder.decoder.decoder.layers.3.mlp.down": {
991
- "activation_amax": 1035.150634765625
992
- },
993
- "video_decoder.decoder.decoder.decoder.layers.3.mlp.gate_up.projections.gate": {
994
- "activation_amax": 4715.27099609375
995
- },
996
- "video_decoder.decoder.decoder.decoder.layers.3.mlp.gate_up.projections.up": {
997
- "activation_amax": 4715.27099609375
998
- },
999
- "video_decoder.decoder.decoder.decoder.layers.3.output": {
1000
- "activation_amax": 2209.302001953125
1001
- },
1002
- "video_decoder.decoder.decoder.decoder.layers.3.qkv.projections.k": {
1003
- "activation_amax": 4715.27099609375
1004
- },
1005
- "video_decoder.decoder.decoder.decoder.layers.3.qkv.projections.q": {
1006
- "activation_amax": 4715.27099609375
1007
- },
1008
- "video_decoder.decoder.decoder.decoder.layers.3.qkv.projections.v": {
1009
- "activation_amax": 4715.27099609375
1010
- },
1011
- "video_decoder.decoder.decoder.decoder.layers.30.mlp.down": {
1012
- "activation_amax": 1779.0299072265625
1013
- },
1014
- "video_decoder.decoder.decoder.decoder.layers.30.mlp.gate_up.projections.gate": {
1015
- "activation_amax": 3407.209716796875
1016
- },
1017
- "video_decoder.decoder.decoder.decoder.layers.30.mlp.gate_up.projections.up": {
1018
- "activation_amax": 3407.209716796875
1019
- },
1020
- "video_decoder.decoder.decoder.decoder.layers.30.output": {
1021
- "activation_amax": 1285.5098876953125
1022
- },
1023
- "video_decoder.decoder.decoder.decoder.layers.30.qkv.projections.k": {
1024
- "activation_amax": 3057.47509765625
1025
- },
1026
- "video_decoder.decoder.decoder.decoder.layers.30.qkv.projections.q": {
1027
- "activation_amax": 3057.47509765625
1028
- },
1029
- "video_decoder.decoder.decoder.decoder.layers.30.qkv.projections.v": {
1030
- "activation_amax": 3057.47509765625
1031
- },
1032
- "video_decoder.decoder.decoder.decoder.layers.31.mlp.down": {
1033
- "activation_amax": 1285.5098876953125
1034
- },
1035
- "video_decoder.decoder.decoder.decoder.layers.31.mlp.gate_up.projections.gate": {
1036
- "activation_amax": 1285.5098876953125
1037
- },
1038
- "video_decoder.decoder.decoder.decoder.layers.31.mlp.gate_up.projections.up": {
1039
- "activation_amax": 1285.5098876953125
1040
- },
1041
- "video_decoder.decoder.decoder.decoder.layers.31.output": {
1042
- "activation_amax": 3407.209716796875
1043
- },
1044
- "video_decoder.decoder.decoder.decoder.layers.31.qkv.projections.k": {
1045
- "activation_amax": 1285.5098876953125
1046
- },
1047
- "video_decoder.decoder.decoder.decoder.layers.31.qkv.projections.q": {
1048
- "activation_amax": 1285.5098876953125
1049
- },
1050
- "video_decoder.decoder.decoder.decoder.layers.31.qkv.projections.v": {
1051
- "activation_amax": 1285.5098876953125
1052
- },
1053
- "video_decoder.decoder.decoder.decoder.layers.32.mlp.down": {
1054
- "activation_amax": 1596.4205322265625
1055
- },
1056
- "video_decoder.decoder.decoder.decoder.layers.32.mlp.gate_up.projections.gate": {
1057
- "activation_amax": 1285.5098876953125
1058
- },
1059
- "video_decoder.decoder.decoder.decoder.layers.32.mlp.gate_up.projections.up": {
1060
- "activation_amax": 1285.5098876953125
1061
- },
1062
- "video_decoder.decoder.decoder.decoder.layers.32.output": {
1063
- "activation_amax": 3796.94921875
1064
- },
1065
- "video_decoder.decoder.decoder.decoder.layers.32.qkv.projections.k": {
1066
- "activation_amax": 1153.55810546875
1067
- },
1068
- "video_decoder.decoder.decoder.decoder.layers.32.qkv.projections.q": {
1069
- "activation_amax": 1153.55810546875
1070
- },
1071
- "video_decoder.decoder.decoder.decoder.layers.32.qkv.projections.v": {
1072
- "activation_amax": 1153.55810546875
1073
- },
1074
- "video_decoder.decoder.decoder.decoder.layers.33.mlp.down": {
1075
- "activation_amax": 1779.0299072265625
1076
- },
1077
- "video_decoder.decoder.decoder.decoder.layers.33.mlp.gate_up.projections.gate": {
1078
- "activation_amax": 1153.55810546875
1079
- },
1080
- "video_decoder.decoder.decoder.decoder.layers.33.mlp.gate_up.projections.up": {
1081
- "activation_amax": 1153.55810546875
1082
- },
1083
- "video_decoder.decoder.decoder.decoder.layers.33.output": {
1084
- "activation_amax": 4715.27099609375
1085
- },
1086
- "video_decoder.decoder.decoder.decoder.layers.33.qkv.projections.k": {
1087
- "activation_amax": 1035.150634765625
1088
- },
1089
- "video_decoder.decoder.decoder.decoder.layers.33.qkv.projections.q": {
1090
- "activation_amax": 1035.150634765625
1091
- },
1092
- "video_decoder.decoder.decoder.decoder.layers.33.qkv.projections.v": {
1093
- "activation_amax": 1035.150634765625
1094
- },
1095
- "video_decoder.decoder.decoder.decoder.layers.34.mlp.down": {
1096
- "activation_amax": 1779.0299072265625
1097
- },
1098
- "video_decoder.decoder.decoder.decoder.layers.34.mlp.gate_up.projections.gate": {
1099
- "activation_amax": 928.8970947265625
1100
- },
1101
- "video_decoder.decoder.decoder.decoder.layers.34.mlp.gate_up.projections.up": {
1102
- "activation_amax": 928.8970947265625
1103
- },
1104
- "video_decoder.decoder.decoder.decoder.layers.34.output": {
1105
- "activation_amax": 3407.209716796875
1106
- },
1107
- "video_decoder.decoder.decoder.decoder.layers.34.qkv.projections.k": {
1108
- "activation_amax": 928.8970947265625
1109
- },
1110
- "video_decoder.decoder.decoder.decoder.layers.34.qkv.projections.q": {
1111
- "activation_amax": 928.8970947265625
1112
- },
1113
- "video_decoder.decoder.decoder.decoder.layers.34.qkv.projections.v": {
1114
- "activation_amax": 928.8970947265625
1115
- },
1116
- "video_decoder.decoder.decoder.decoder.layers.35.mlp.down": {
1117
- "activation_amax": 2462.01708984375
1118
- },
1119
- "video_decoder.decoder.decoder.decoder.layers.35.mlp.gate_up.projections.gate": {
1120
- "activation_amax": 833.550048828125
1121
- },
1122
- "video_decoder.decoder.decoder.decoder.layers.35.mlp.gate_up.projections.up": {
1123
- "activation_amax": 833.550048828125
1124
- },
1125
- "video_decoder.decoder.decoder.decoder.layers.35.output": {
1126
- "activation_amax": 4231.27001953125
1127
- },
1128
- "video_decoder.decoder.decoder.decoder.layers.35.qkv.projections.k": {
1129
- "activation_amax": 833.550048828125
1130
- },
1131
- "video_decoder.decoder.decoder.decoder.layers.35.qkv.projections.q": {
1132
- "activation_amax": 833.550048828125
1133
- },
1134
- "video_decoder.decoder.decoder.decoder.layers.35.qkv.projections.v": {
1135
- "activation_amax": 833.550048828125
1136
- },
1137
- "video_decoder.decoder.decoder.decoder.layers.4.mlp.down": {
1138
- "activation_amax": 1153.55810546875
1139
- },
1140
- "video_decoder.decoder.decoder.decoder.layers.4.mlp.gate_up.projections.gate": {
1141
- "activation_amax": 5254.6357421875
1142
- },
1143
- "video_decoder.decoder.decoder.decoder.layers.4.mlp.gate_up.projections.up": {
1144
- "activation_amax": 5254.6357421875
1145
- },
1146
- "video_decoder.decoder.decoder.decoder.layers.4.output": {
1147
- "activation_amax": 1779.0299072265625
1148
- },
1149
- "video_decoder.decoder.decoder.decoder.layers.4.qkv.projections.k": {
1150
- "activation_amax": 5254.6357421875
1151
- },
1152
- "video_decoder.decoder.decoder.decoder.layers.4.qkv.projections.q": {
1153
- "activation_amax": 5254.6357421875
1154
- },
1155
- "video_decoder.decoder.decoder.decoder.layers.4.qkv.projections.v": {
1156
- "activation_amax": 5254.6357421875
1157
- },
1158
- "video_decoder.decoder.decoder.decoder.layers.5.mlp.down": {
1159
- "activation_amax": 1432.55517578125
1160
- },
1161
- "video_decoder.decoder.decoder.decoder.layers.5.mlp.gate_up.projections.gate": {
1162
- "activation_amax": 5855.6962890625
1163
- },
1164
- "video_decoder.decoder.decoder.decoder.layers.5.mlp.gate_up.projections.up": {
1165
- "activation_amax": 5855.6962890625
1166
- },
1167
- "video_decoder.decoder.decoder.decoder.layers.5.output": {
1168
- "activation_amax": 2209.302001953125
1169
- },
1170
- "video_decoder.decoder.decoder.decoder.layers.5.qkv.projections.k": {
1171
- "activation_amax": 6525.51025390625
1172
- },
1173
- "video_decoder.decoder.decoder.decoder.layers.5.qkv.projections.q": {
1174
- "activation_amax": 6525.51025390625
1175
- },
1176
- "video_decoder.decoder.decoder.decoder.layers.5.qkv.projections.v": {
1177
- "activation_amax": 6525.51025390625
1178
- },
1179
- "video_decoder.decoder.decoder.decoder.layers.6.mlp.down": {
1180
- "activation_amax": 1596.4205322265625
1181
- },
1182
- "video_decoder.decoder.decoder.decoder.layers.6.mlp.gate_up.projections.gate": {
1183
- "activation_amax": 6525.51025390625
1184
- },
1185
- "video_decoder.decoder.decoder.decoder.layers.6.mlp.gate_up.projections.up": {
1186
- "activation_amax": 6525.51025390625
1187
- },
1188
- "video_decoder.decoder.decoder.decoder.layers.6.output": {
1189
- "activation_amax": 3057.47509765625
1190
- },
1191
- "video_decoder.decoder.decoder.decoder.layers.6.qkv.projections.k": {
1192
- "activation_amax": 7271.94189453125
1193
- },
1194
- "video_decoder.decoder.decoder.decoder.layers.6.qkv.projections.q": {
1195
- "activation_amax": 7271.94189453125
1196
- },
1197
- "video_decoder.decoder.decoder.decoder.layers.6.qkv.projections.v": {
1198
- "activation_amax": 7271.94189453125
1199
- },
1200
- "video_decoder.decoder.decoder.decoder.layers.7.mlp.down": {
1201
- "activation_amax": 1596.4205322265625
1202
- },
1203
- "video_decoder.decoder.decoder.decoder.layers.7.mlp.gate_up.projections.gate": {
1204
- "activation_amax": 7271.94189453125
1205
- },
1206
- "video_decoder.decoder.decoder.decoder.layers.7.mlp.gate_up.projections.up": {
1207
- "activation_amax": 7271.94189453125
1208
- },
1209
- "video_decoder.decoder.decoder.decoder.layers.7.output": {
1210
- "activation_amax": 3057.47509765625
1211
- },
1212
- "video_decoder.decoder.decoder.decoder.layers.7.qkv.projections.k": {
1213
- "activation_amax": 8103.755859375
1214
- },
1215
- "video_decoder.decoder.decoder.decoder.layers.7.qkv.projections.q": {
1216
- "activation_amax": 8103.755859375
1217
- },
1218
- "video_decoder.decoder.decoder.decoder.layers.7.qkv.projections.v": {
1219
- "activation_amax": 8103.755859375
1220
- },
1221
- "video_decoder.decoder.decoder.decoder.layers.8.mlp.down": {
1222
- "activation_amax": 1779.0299072265625
1223
- },
1224
- "video_decoder.decoder.decoder.decoder.layers.8.mlp.gate_up.projections.gate": {
1225
- "activation_amax": 8103.755859375
1226
- },
1227
- "video_decoder.decoder.decoder.decoder.layers.8.mlp.gate_up.projections.up": {
1228
- "activation_amax": 8103.755859375
1229
- },
1230
- "video_decoder.decoder.decoder.decoder.layers.8.output": {
1231
- "activation_amax": 3796.94921875
1232
- },
1233
- "video_decoder.decoder.decoder.decoder.layers.8.qkv.projections.k": {
1234
- "activation_amax": 9030.7177734375
1235
- },
1236
- "video_decoder.decoder.decoder.decoder.layers.8.qkv.projections.q": {
1237
- "activation_amax": 9030.7177734375
1238
- },
1239
- "video_decoder.decoder.decoder.decoder.layers.8.qkv.projections.v": {
1240
- "activation_amax": 9030.7177734375
1241
- },
1242
- "video_decoder.decoder.decoder.decoder.layers.9.mlp.down": {
1243
- "activation_amax": 1982.5272216796875
1244
- },
1245
- "video_decoder.decoder.decoder.decoder.layers.9.mlp.gate_up.projections.gate": {
1246
- "activation_amax": 8103.755859375
1247
- },
1248
- "video_decoder.decoder.decoder.decoder.layers.9.mlp.gate_up.projections.up": {
1249
- "activation_amax": 8103.755859375
1250
- },
1251
- "video_decoder.decoder.decoder.decoder.layers.9.output": {
1252
- "activation_amax": 3407.209716796875
1253
- },
1254
- "video_decoder.decoder.decoder.decoder.layers.9.qkv.projections.k": {
1255
- "activation_amax": 9030.7177734375
1256
- },
1257
- "video_decoder.decoder.decoder.decoder.layers.9.qkv.projections.q": {
1258
- "activation_amax": 9030.7177734375
1259
- },
1260
- "video_decoder.decoder.decoder.decoder.layers.9.qkv.projections.v": {
1261
- "activation_amax": 9030.7177734375
1262
- }
1263
- },
1264
- "quantized_patterns": [
1265
- "decoder.transformer_blocks.*.{q,k,v,out,gate,up,down}"
1266
- ]
1267
- }
1268
- },
1269
- "modelopt_commit": "6a4b3f147e14a6fec690fedbced8df402344085d",
1270
- "numerical_format": {
1271
- "activation": "a4",
1272
- "block_scale": "fp8_e4m3",
1273
- "block_size": 16,
1274
- "output": "bf16",
1275
- "tensor_scale": "fp32",
1276
- "values": "e2m1",
1277
- "weight": "w4"
1278
- },
1279
- "schema_version": 1
1280
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
transformer/config.json CHANGED
@@ -22,5 +22,354 @@
22
  "rope_theta": 10000.0,
23
  "norm_eps": 1e-05,
24
  "qk_norm_eps": 1e-05,
25
- "final_norm_eps": 1e-05
26
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
22
  "rope_theta": 10000.0,
23
  "norm_eps": 1e-05,
24
  "qk_norm_eps": 1e-05,
25
+ "final_norm_eps": 1e-05,
26
+ "quantization_config": {
27
+ "config_groups": {
28
+ "group_0": {
29
+ "input_activations": {
30
+ "dynamic": false,
31
+ "num_bits": 4,
32
+ "type": "float",
33
+ "group_size": 16
34
+ },
35
+ "weights": {
36
+ "dynamic": false,
37
+ "num_bits": 4,
38
+ "type": "float",
39
+ "group_size": 16
40
+ },
41
+ "targets": [
42
+ "Linear"
43
+ ]
44
+ }
45
+ },
46
+ "ignore": [
47
+ "audio_proj_in",
48
+ "audio_proj_out",
49
+ "context_embedder",
50
+ "norm_out.linear",
51
+ "proj_in",
52
+ "proj_out",
53
+ "time_embedder.linear_1",
54
+ "time_embedder.linear_2",
55
+ "token_refiner.refiner_blocks.0.attn.to_k",
56
+ "token_refiner.refiner_blocks.0.attn.to_out.0",
57
+ "token_refiner.refiner_blocks.0.attn.to_q",
58
+ "token_refiner.refiner_blocks.0.attn.to_v",
59
+ "token_refiner.refiner_blocks.0.ff.net.0.proj",
60
+ "token_refiner.refiner_blocks.0.ff.net.2",
61
+ "token_refiner.refiner_blocks.1.attn.to_k",
62
+ "token_refiner.refiner_blocks.1.attn.to_out.0",
63
+ "token_refiner.refiner_blocks.1.attn.to_q",
64
+ "token_refiner.refiner_blocks.1.attn.to_v",
65
+ "token_refiner.refiner_blocks.1.ff.net.0.proj",
66
+ "token_refiner.refiner_blocks.1.ff.net.2",
67
+ "transformer_blocks.0.adaln_proj.linear",
68
+ "transformer_blocks.0.attn.to_gate_compress",
69
+ "transformer_blocks.0.attn.to_k",
70
+ "transformer_blocks.0.attn.to_out.0",
71
+ "transformer_blocks.0.attn.to_q",
72
+ "transformer_blocks.0.attn.to_v",
73
+ "transformer_blocks.1.adaln_proj.linear",
74
+ "transformer_blocks.1.attn.to_gate_compress",
75
+ "transformer_blocks.1.attn.to_k",
76
+ "transformer_blocks.1.attn.to_out.0",
77
+ "transformer_blocks.1.attn.to_q",
78
+ "transformer_blocks.1.attn.to_v",
79
+ "transformer_blocks.10.adaln_proj.linear",
80
+ "transformer_blocks.10.attn.to_gate_compress",
81
+ "transformer_blocks.10.attn.to_k",
82
+ "transformer_blocks.10.attn.to_out.0",
83
+ "transformer_blocks.10.attn.to_q",
84
+ "transformer_blocks.10.attn.to_v",
85
+ "transformer_blocks.11.adaln_proj.linear",
86
+ "transformer_blocks.11.attn.to_gate_compress",
87
+ "transformer_blocks.11.attn.to_k",
88
+ "transformer_blocks.11.attn.to_out.0",
89
+ "transformer_blocks.11.attn.to_q",
90
+ "transformer_blocks.11.attn.to_v",
91
+ "transformer_blocks.12.adaln_proj.linear",
92
+ "transformer_blocks.12.attn.to_gate_compress",
93
+ "transformer_blocks.12.attn.to_k",
94
+ "transformer_blocks.12.attn.to_out.0",
95
+ "transformer_blocks.12.attn.to_q",
96
+ "transformer_blocks.12.attn.to_v",
97
+ "transformer_blocks.13.adaln_proj.linear",
98
+ "transformer_blocks.13.attn.to_gate_compress",
99
+ "transformer_blocks.13.attn.to_k",
100
+ "transformer_blocks.13.attn.to_out.0",
101
+ "transformer_blocks.13.attn.to_q",
102
+ "transformer_blocks.13.attn.to_v",
103
+ "transformer_blocks.14.adaln_proj.linear",
104
+ "transformer_blocks.14.attn.to_gate_compress",
105
+ "transformer_blocks.14.attn.to_k",
106
+ "transformer_blocks.14.attn.to_out.0",
107
+ "transformer_blocks.14.attn.to_q",
108
+ "transformer_blocks.14.attn.to_v",
109
+ "transformer_blocks.15.adaln_proj.linear",
110
+ "transformer_blocks.15.attn.to_gate_compress",
111
+ "transformer_blocks.15.attn.to_k",
112
+ "transformer_blocks.15.attn.to_out.0",
113
+ "transformer_blocks.15.attn.to_q",
114
+ "transformer_blocks.15.attn.to_v",
115
+ "transformer_blocks.16.adaln_proj.linear",
116
+ "transformer_blocks.16.attn.to_gate_compress",
117
+ "transformer_blocks.16.attn.to_k",
118
+ "transformer_blocks.16.attn.to_out.0",
119
+ "transformer_blocks.16.attn.to_q",
120
+ "transformer_blocks.16.attn.to_v",
121
+ "transformer_blocks.17.adaln_proj.linear",
122
+ "transformer_blocks.17.attn.to_gate_compress",
123
+ "transformer_blocks.17.attn.to_k",
124
+ "transformer_blocks.17.attn.to_out.0",
125
+ "transformer_blocks.17.attn.to_q",
126
+ "transformer_blocks.17.attn.to_v",
127
+ "transformer_blocks.18.adaln_proj.linear",
128
+ "transformer_blocks.18.attn.to_gate_compress",
129
+ "transformer_blocks.18.attn.to_k",
130
+ "transformer_blocks.18.attn.to_out.0",
131
+ "transformer_blocks.18.attn.to_q",
132
+ "transformer_blocks.18.attn.to_v",
133
+ "transformer_blocks.19.adaln_proj.linear",
134
+ "transformer_blocks.19.attn.to_gate_compress",
135
+ "transformer_blocks.19.attn.to_k",
136
+ "transformer_blocks.19.attn.to_out.0",
137
+ "transformer_blocks.19.attn.to_q",
138
+ "transformer_blocks.19.attn.to_v",
139
+ "transformer_blocks.2.adaln_proj.linear",
140
+ "transformer_blocks.2.attn.to_gate_compress",
141
+ "transformer_blocks.2.attn.to_k",
142
+ "transformer_blocks.2.attn.to_out.0",
143
+ "transformer_blocks.2.attn.to_q",
144
+ "transformer_blocks.2.attn.to_v",
145
+ "transformer_blocks.20.adaln_proj.linear",
146
+ "transformer_blocks.20.attn.to_gate_compress",
147
+ "transformer_blocks.20.attn.to_k",
148
+ "transformer_blocks.20.attn.to_out.0",
149
+ "transformer_blocks.20.attn.to_q",
150
+ "transformer_blocks.20.attn.to_v",
151
+ "transformer_blocks.21.adaln_proj.linear",
152
+ "transformer_blocks.21.attn.to_gate_compress",
153
+ "transformer_blocks.21.attn.to_k",
154
+ "transformer_blocks.21.attn.to_out.0",
155
+ "transformer_blocks.21.attn.to_q",
156
+ "transformer_blocks.21.attn.to_v",
157
+ "transformer_blocks.22.adaln_proj.linear",
158
+ "transformer_blocks.22.attn.to_gate_compress",
159
+ "transformer_blocks.22.attn.to_k",
160
+ "transformer_blocks.22.attn.to_out.0",
161
+ "transformer_blocks.22.attn.to_q",
162
+ "transformer_blocks.22.attn.to_v",
163
+ "transformer_blocks.23.adaln_proj.linear",
164
+ "transformer_blocks.23.attn.to_gate_compress",
165
+ "transformer_blocks.23.attn.to_k",
166
+ "transformer_blocks.23.attn.to_out.0",
167
+ "transformer_blocks.23.attn.to_q",
168
+ "transformer_blocks.23.attn.to_v",
169
+ "transformer_blocks.24.adaln_proj.linear",
170
+ "transformer_blocks.24.attn.to_gate_compress",
171
+ "transformer_blocks.24.attn.to_k",
172
+ "transformer_blocks.24.attn.to_out.0",
173
+ "transformer_blocks.24.attn.to_q",
174
+ "transformer_blocks.24.attn.to_v",
175
+ "transformer_blocks.25.adaln_proj.linear",
176
+ "transformer_blocks.25.attn.to_gate_compress",
177
+ "transformer_blocks.25.attn.to_k",
178
+ "transformer_blocks.25.attn.to_out.0",
179
+ "transformer_blocks.25.attn.to_q",
180
+ "transformer_blocks.25.attn.to_v",
181
+ "transformer_blocks.26.adaln_proj.linear",
182
+ "transformer_blocks.26.attn.to_gate_compress",
183
+ "transformer_blocks.26.attn.to_k",
184
+ "transformer_blocks.26.attn.to_out.0",
185
+ "transformer_blocks.26.attn.to_q",
186
+ "transformer_blocks.26.attn.to_v",
187
+ "transformer_blocks.27.adaln_proj.linear",
188
+ "transformer_blocks.27.attn.to_gate_compress",
189
+ "transformer_blocks.27.attn.to_k",
190
+ "transformer_blocks.27.attn.to_out.0",
191
+ "transformer_blocks.27.attn.to_q",
192
+ "transformer_blocks.27.attn.to_v",
193
+ "transformer_blocks.28.adaln_proj.linear",
194
+ "transformer_blocks.28.attn.to_gate_compress",
195
+ "transformer_blocks.28.attn.to_k",
196
+ "transformer_blocks.28.attn.to_out.0",
197
+ "transformer_blocks.28.attn.to_q",
198
+ "transformer_blocks.28.attn.to_v",
199
+ "transformer_blocks.29.adaln_proj.linear",
200
+ "transformer_blocks.29.attn.to_gate_compress",
201
+ "transformer_blocks.29.attn.to_k",
202
+ "transformer_blocks.29.attn.to_out.0",
203
+ "transformer_blocks.29.attn.to_q",
204
+ "transformer_blocks.29.attn.to_v",
205
+ "transformer_blocks.3.adaln_proj.linear",
206
+ "transformer_blocks.3.attn.to_gate_compress",
207
+ "transformer_blocks.3.attn.to_k",
208
+ "transformer_blocks.3.attn.to_out.0",
209
+ "transformer_blocks.3.attn.to_q",
210
+ "transformer_blocks.3.attn.to_v",
211
+ "transformer_blocks.30.adaln_proj.linear",
212
+ "transformer_blocks.30.attn.to_gate_compress",
213
+ "transformer_blocks.30.attn.to_k",
214
+ "transformer_blocks.30.attn.to_out.0",
215
+ "transformer_blocks.30.attn.to_q",
216
+ "transformer_blocks.30.attn.to_v",
217
+ "transformer_blocks.31.adaln_proj.linear",
218
+ "transformer_blocks.31.attn.to_gate_compress",
219
+ "transformer_blocks.31.attn.to_k",
220
+ "transformer_blocks.31.attn.to_out.0",
221
+ "transformer_blocks.31.attn.to_q",
222
+ "transformer_blocks.31.attn.to_v",
223
+ "transformer_blocks.32.adaln_proj.linear",
224
+ "transformer_blocks.32.attn.to_gate_compress",
225
+ "transformer_blocks.32.attn.to_k",
226
+ "transformer_blocks.32.attn.to_out.0",
227
+ "transformer_blocks.32.attn.to_q",
228
+ "transformer_blocks.32.attn.to_v",
229
+ "transformer_blocks.33.adaln_proj.linear",
230
+ "transformer_blocks.33.attn.to_gate_compress",
231
+ "transformer_blocks.33.attn.to_k",
232
+ "transformer_blocks.33.attn.to_out.0",
233
+ "transformer_blocks.33.attn.to_q",
234
+ "transformer_blocks.33.attn.to_v",
235
+ "transformer_blocks.34.adaln_proj.linear",
236
+ "transformer_blocks.34.attn.to_gate_compress",
237
+ "transformer_blocks.34.attn.to_k",
238
+ "transformer_blocks.34.attn.to_out.0",
239
+ "transformer_blocks.34.attn.to_q",
240
+ "transformer_blocks.34.attn.to_v",
241
+ "transformer_blocks.35.adaln_proj.linear",
242
+ "transformer_blocks.35.attn.to_gate_compress",
243
+ "transformer_blocks.35.attn.to_k",
244
+ "transformer_blocks.35.attn.to_out.0",
245
+ "transformer_blocks.35.attn.to_q",
246
+ "transformer_blocks.35.attn.to_v",
247
+ "transformer_blocks.36.adaln_proj.linear",
248
+ "transformer_blocks.36.attn.to_gate_compress",
249
+ "transformer_blocks.36.attn.to_k",
250
+ "transformer_blocks.36.attn.to_out.0",
251
+ "transformer_blocks.36.attn.to_q",
252
+ "transformer_blocks.36.attn.to_v",
253
+ "transformer_blocks.37.adaln_proj.linear",
254
+ "transformer_blocks.37.attn.to_gate_compress",
255
+ "transformer_blocks.37.attn.to_k",
256
+ "transformer_blocks.37.attn.to_out.0",
257
+ "transformer_blocks.37.attn.to_q",
258
+ "transformer_blocks.37.attn.to_v",
259
+ "transformer_blocks.38.adaln_proj.linear",
260
+ "transformer_blocks.38.attn.to_gate_compress",
261
+ "transformer_blocks.38.attn.to_k",
262
+ "transformer_blocks.38.attn.to_out.0",
263
+ "transformer_blocks.38.attn.to_q",
264
+ "transformer_blocks.38.attn.to_v",
265
+ "transformer_blocks.39.adaln_proj.linear",
266
+ "transformer_blocks.39.attn.to_gate_compress",
267
+ "transformer_blocks.39.attn.to_k",
268
+ "transformer_blocks.39.attn.to_out.0",
269
+ "transformer_blocks.39.attn.to_q",
270
+ "transformer_blocks.39.attn.to_v",
271
+ "transformer_blocks.4.adaln_proj.linear",
272
+ "transformer_blocks.4.attn.to_gate_compress",
273
+ "transformer_blocks.4.attn.to_k",
274
+ "transformer_blocks.4.attn.to_out.0",
275
+ "transformer_blocks.4.attn.to_q",
276
+ "transformer_blocks.4.attn.to_v",
277
+ "transformer_blocks.40.adaln_proj.linear",
278
+ "transformer_blocks.40.attn.to_gate_compress",
279
+ "transformer_blocks.40.attn.to_k",
280
+ "transformer_blocks.40.attn.to_out.0",
281
+ "transformer_blocks.40.attn.to_q",
282
+ "transformer_blocks.40.attn.to_v",
283
+ "transformer_blocks.41.adaln_proj.linear",
284
+ "transformer_blocks.41.attn.to_gate_compress",
285
+ "transformer_blocks.41.attn.to_k",
286
+ "transformer_blocks.41.attn.to_out.0",
287
+ "transformer_blocks.41.attn.to_q",
288
+ "transformer_blocks.41.attn.to_v",
289
+ "transformer_blocks.42.adaln_proj.linear",
290
+ "transformer_blocks.42.attn.to_gate_compress",
291
+ "transformer_blocks.42.attn.to_k",
292
+ "transformer_blocks.42.attn.to_out.0",
293
+ "transformer_blocks.42.attn.to_q",
294
+ "transformer_blocks.42.attn.to_v",
295
+ "transformer_blocks.43.adaln_proj.linear",
296
+ "transformer_blocks.43.attn.to_gate_compress",
297
+ "transformer_blocks.43.attn.to_k",
298
+ "transformer_blocks.43.attn.to_out.0",
299
+ "transformer_blocks.43.attn.to_q",
300
+ "transformer_blocks.43.attn.to_v",
301
+ "transformer_blocks.44.adaln_proj.linear",
302
+ "transformer_blocks.44.attn.to_gate_compress",
303
+ "transformer_blocks.44.attn.to_k",
304
+ "transformer_blocks.44.attn.to_out.0",
305
+ "transformer_blocks.44.attn.to_q",
306
+ "transformer_blocks.44.attn.to_v",
307
+ "transformer_blocks.45.adaln_proj.linear",
308
+ "transformer_blocks.45.attn.to_gate_compress",
309
+ "transformer_blocks.45.attn.to_k",
310
+ "transformer_blocks.45.attn.to_out.0",
311
+ "transformer_blocks.45.attn.to_q",
312
+ "transformer_blocks.45.attn.to_v",
313
+ "transformer_blocks.46.adaln_proj.linear",
314
+ "transformer_blocks.46.attn.to_gate_compress",
315
+ "transformer_blocks.46.attn.to_k",
316
+ "transformer_blocks.46.attn.to_out.0",
317
+ "transformer_blocks.46.attn.to_q",
318
+ "transformer_blocks.46.attn.to_v",
319
+ "transformer_blocks.47.adaln_proj.linear",
320
+ "transformer_blocks.47.attn.to_gate_compress",
321
+ "transformer_blocks.47.attn.to_k",
322
+ "transformer_blocks.47.attn.to_out.0",
323
+ "transformer_blocks.47.attn.to_q",
324
+ "transformer_blocks.47.attn.to_v",
325
+ "transformer_blocks.48.adaln_proj.linear",
326
+ "transformer_blocks.48.attn.to_gate_compress",
327
+ "transformer_blocks.48.attn.to_k",
328
+ "transformer_blocks.48.attn.to_out.0",
329
+ "transformer_blocks.48.attn.to_q",
330
+ "transformer_blocks.48.attn.to_v",
331
+ "transformer_blocks.49.adaln_proj.linear",
332
+ "transformer_blocks.49.attn.to_gate_compress",
333
+ "transformer_blocks.49.attn.to_k",
334
+ "transformer_blocks.49.attn.to_out.0",
335
+ "transformer_blocks.49.attn.to_q",
336
+ "transformer_blocks.49.attn.to_v",
337
+ "transformer_blocks.5.adaln_proj.linear",
338
+ "transformer_blocks.5.attn.to_gate_compress",
339
+ "transformer_blocks.5.attn.to_k",
340
+ "transformer_blocks.5.attn.to_out.0",
341
+ "transformer_blocks.5.attn.to_q",
342
+ "transformer_blocks.5.attn.to_v",
343
+ "transformer_blocks.6.adaln_proj.linear",
344
+ "transformer_blocks.6.attn.to_gate_compress",
345
+ "transformer_blocks.6.attn.to_k",
346
+ "transformer_blocks.6.attn.to_out.0",
347
+ "transformer_blocks.6.attn.to_q",
348
+ "transformer_blocks.6.attn.to_v",
349
+ "transformer_blocks.7.adaln_proj.linear",
350
+ "transformer_blocks.7.attn.to_gate_compress",
351
+ "transformer_blocks.7.attn.to_k",
352
+ "transformer_blocks.7.attn.to_out.0",
353
+ "transformer_blocks.7.attn.to_q",
354
+ "transformer_blocks.7.attn.to_v",
355
+ "transformer_blocks.8.adaln_proj.linear",
356
+ "transformer_blocks.8.attn.to_gate_compress",
357
+ "transformer_blocks.8.attn.to_k",
358
+ "transformer_blocks.8.attn.to_out.0",
359
+ "transformer_blocks.8.attn.to_q",
360
+ "transformer_blocks.8.attn.to_v",
361
+ "transformer_blocks.9.adaln_proj.linear",
362
+ "transformer_blocks.9.attn.to_gate_compress",
363
+ "transformer_blocks.9.attn.to_k",
364
+ "transformer_blocks.9.attn.to_out.0",
365
+ "transformer_blocks.9.attn.to_q",
366
+ "transformer_blocks.9.attn.to_v"
367
+ ],
368
+ "quant_algo": "NVFP4",
369
+ "producer": {
370
+ "name": "modelopt",
371
+ "version": "0.47.0rc0-63-g6a4b3f14"
372
+ },
373
+ "quant_method": "modelopt"
374
+ }
375
+ }
transformer/diffusion_pytorch_model-00001-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:66dd07f1629c38e19c4c18384405cef84cc9b8dd424bedd1de5f953c05991e41
3
- size 4666277856
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71eb3dde330647e18478c7e9aa3acbdec58abd0e42c4b9e96ed78ae6dffcf17a
3
+ size 4666278208
transformer/diffusion_pytorch_model-00002-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2a11227cb6c67446498fc44a92bf45683425da4e06ca09282a2707a260e90584
3
- size 3989259848
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cbc2ba2c5dc7b6e9f3c93b54b305e1f4005a2dffd2fac7ed93101854ebab6a5
3
+ size 3989260560
transformer/diffusion_pytorch_model-00003-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4aafaa4c41e4022d58f236b990672545e829c9629a678a0dc4e220a4d29a1446
3
- size 3989259864
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:566f5f8597effa27c69548ad3186e56f5e3e46d1caef6724a23f3175a7190803
3
+ size 3989260584
transformer/diffusion_pytorch_model-00004-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:072ba683a4dbb184ca4c0d9a561563b1a7259957c342475f3536d7c7f5712c48
3
- size 3546104664
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3597343c1b46d5196699cf406df7ee3b4ad9cdf25a2c4ea7c39faadfc5a615f
3
+ size 3546105376
transformer/diffusion_pytorch_model-00005-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1762c23119da1924ed7a2be99eba18b05f7ee25be320c791067bea99e8f281bc
3
- size 4099832988
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c637251aa345f5c92c11a7cbf3dd7c6232a03d17cc44897b83dadd54f183b192
3
+ size 4099833616
transformer/diffusion_pytorch_model-00006-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6faddc05d3478d965faea9948f65443a3e908f2528a7de25af795615c84ee640
3
- size 4056696260
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2646c95d33f7719f177ba8d2e408640a592fe3bfbaa7a9d6e142d627b60d8dda
3
+ size 4056696880
transformer/diffusion_pytorch_model-00007-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d8ceaad5fdfd8bb70ffff4c4c5d0cb164f05e609fc3a428b1e171d2ad519c6a6
3
- size 3989259832
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:366a8574a0293bb69e5c24a291b4d112cb0cf63a633756bd638e065d6387fc1a
3
+ size 3989260544
transformer/diffusion_pytorch_model-00008-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f29d771bad140fffc723d3c9fa4910cf3d450058387c83f74b974ee61b0dee5e
3
- size 3989259864
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:10ffc495acf04998fabfcbfcea4a229e0ddbd5a3a730fc500491580a32dbf39e
3
+ size 3989260584
transformer/diffusion_pytorch_model-00009-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:49da116f9b79e3a1b50b3085cbaa62ed1511f498ec1e209544a1567183456dab
3
- size 3546104680
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05ab97ea2ba92a40f46bf57948b9c566f373d70038b46b87f99cccbc68c342dc
3
+ size 3546105400
transformer/diffusion_pytorch_model-00010-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:550290b051f89ca10a40209fd3d79334e91110088a522cc9544776b43aa17da8
3
- size 4099832972
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa4a0ae7316673d5763cd2c0594b69670773c4068a06ba2261bd303e4a5045e6
3
+ size 4099833600
transformer/diffusion_pytorch_model-00011-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a3626720879942a3afb460a00b98a375911d3e1d07b52a885b5ba4843f5def55
3
- size 4056696260
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8eef1af665574360d2df4049656ce0fc1a983352cbbe82f2d6b4bec83341026d
3
+ size 4056696880
transformer/diffusion_pytorch_model-00012-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cd2be36a2032ee027083be015675949970732d4d579ada810a34b854fe557731
3
- size 3989259848
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0f2c96b23922b5c0bc158ddc5c237ad4d8937058be596a0bf56a76afa849294
3
+ size 3989260560
transformer/diffusion_pytorch_model-00013-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a63129e407c505b446b434cbb8375874156ab27a6031d440e48eac1eb32db981
3
- size 3989259808
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:436079ac0ff83ce3ae79f7e58d07ff61b8af5ad0cbadff62a52c2cf039d92e5a
3
+ size 3989260520
transformer/diffusion_pytorch_model-00014-of-00014.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aa1081318d90d28644ee9250314298639bcf323e770e5e758d527886cc246bb3
3
- size 1474210592
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3990ddfd96432a9d7b76c09ada2c7814329787503f7b96b7b12de049e18beec
3
+ size 1474210952
transformer/diffusion_pytorch_model.safetensors.index.json CHANGED
@@ -50,12 +50,14 @@
50
  "transformer_blocks.0.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
51
  "transformer_blocks.0.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
52
  "transformer_blocks.0.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
53
- "transformer_blocks.0.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00001-of-00014.safetensors",
 
54
  "transformer_blocks.0.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
55
- "transformer_blocks.0.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
56
- "transformer_blocks.0.ff.net.2.weight_packed": "diffusion_pytorch_model-00001-of-00014.safetensors",
 
57
  "transformer_blocks.0.ff.net.2.weight_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
58
- "transformer_blocks.0.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
59
  "transformer_blocks.0.norm1.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
60
  "transformer_blocks.0.norm2.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
61
  "transformer_blocks.1.adaln_proj.linear.bias": "diffusion_pytorch_model-00001-of-00014.safetensors",
@@ -67,12 +69,14 @@
67
  "transformer_blocks.1.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
68
  "transformer_blocks.1.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
69
  "transformer_blocks.1.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
70
- "transformer_blocks.1.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00001-of-00014.safetensors",
 
71
  "transformer_blocks.1.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
72
- "transformer_blocks.1.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
73
- "transformer_blocks.1.ff.net.2.weight_packed": "diffusion_pytorch_model-00001-of-00014.safetensors",
 
74
  "transformer_blocks.1.ff.net.2.weight_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
75
- "transformer_blocks.1.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
76
  "transformer_blocks.1.norm1.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
77
  "transformer_blocks.1.norm2.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
78
  "transformer_blocks.10.adaln_proj.linear.bias": "diffusion_pytorch_model-00001-of-00014.safetensors",
@@ -84,12 +88,14 @@
84
  "transformer_blocks.10.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
85
  "transformer_blocks.10.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
86
  "transformer_blocks.10.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
87
- "transformer_blocks.10.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
88
  "transformer_blocks.10.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
89
- "transformer_blocks.10.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
90
- "transformer_blocks.10.ff.net.2.weight_packed": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
91
  "transformer_blocks.10.ff.net.2.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
92
- "transformer_blocks.10.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
93
  "transformer_blocks.10.norm1.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
94
  "transformer_blocks.10.norm2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
95
  "transformer_blocks.11.adaln_proj.linear.bias": "diffusion_pytorch_model-00002-of-00014.safetensors",
@@ -101,12 +107,14 @@
101
  "transformer_blocks.11.attn.to_out.0.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
102
  "transformer_blocks.11.attn.to_q.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
103
  "transformer_blocks.11.attn.to_v.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
104
- "transformer_blocks.11.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
105
  "transformer_blocks.11.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
106
- "transformer_blocks.11.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
107
- "transformer_blocks.11.ff.net.2.weight_packed": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
108
  "transformer_blocks.11.ff.net.2.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
109
- "transformer_blocks.11.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
110
  "transformer_blocks.11.norm1.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
111
  "transformer_blocks.11.norm2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
112
  "transformer_blocks.12.adaln_proj.linear.bias": "diffusion_pytorch_model-00002-of-00014.safetensors",
@@ -118,12 +126,14 @@
118
  "transformer_blocks.12.attn.to_out.0.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
119
  "transformer_blocks.12.attn.to_q.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
120
  "transformer_blocks.12.attn.to_v.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
121
- "transformer_blocks.12.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
122
  "transformer_blocks.12.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
123
- "transformer_blocks.12.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
124
- "transformer_blocks.12.ff.net.2.weight_packed": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
125
  "transformer_blocks.12.ff.net.2.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
126
- "transformer_blocks.12.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
127
  "transformer_blocks.12.norm1.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
128
  "transformer_blocks.12.norm2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
129
  "transformer_blocks.13.adaln_proj.linear.bias": "diffusion_pytorch_model-00002-of-00014.safetensors",
@@ -135,12 +145,14 @@
135
  "transformer_blocks.13.attn.to_out.0.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
136
  "transformer_blocks.13.attn.to_q.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
137
  "transformer_blocks.13.attn.to_v.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
138
- "transformer_blocks.13.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
139
  "transformer_blocks.13.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
140
- "transformer_blocks.13.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
141
- "transformer_blocks.13.ff.net.2.weight_packed": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
142
  "transformer_blocks.13.ff.net.2.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
143
- "transformer_blocks.13.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
144
  "transformer_blocks.13.norm1.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
145
  "transformer_blocks.13.norm2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
146
  "transformer_blocks.14.adaln_proj.linear.bias": "diffusion_pytorch_model-00002-of-00014.safetensors",
@@ -152,12 +164,14 @@
152
  "transformer_blocks.14.attn.to_out.0.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
153
  "transformer_blocks.14.attn.to_q.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
154
  "transformer_blocks.14.attn.to_v.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
155
- "transformer_blocks.14.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
156
  "transformer_blocks.14.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
157
- "transformer_blocks.14.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
158
- "transformer_blocks.14.ff.net.2.weight_packed": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
159
  "transformer_blocks.14.ff.net.2.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
160
- "transformer_blocks.14.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
161
  "transformer_blocks.14.norm1.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
162
  "transformer_blocks.14.norm2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
163
  "transformer_blocks.15.adaln_proj.linear.bias": "diffusion_pytorch_model-00003-of-00014.safetensors",
@@ -169,12 +183,14 @@
169
  "transformer_blocks.15.attn.to_out.0.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
170
  "transformer_blocks.15.attn.to_q.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
171
  "transformer_blocks.15.attn.to_v.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
172
- "transformer_blocks.15.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
173
  "transformer_blocks.15.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
174
- "transformer_blocks.15.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
175
- "transformer_blocks.15.ff.net.2.weight_packed": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
176
  "transformer_blocks.15.ff.net.2.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
177
- "transformer_blocks.15.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
178
  "transformer_blocks.15.norm1.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
179
  "transformer_blocks.15.norm2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
180
  "transformer_blocks.16.adaln_proj.linear.bias": "diffusion_pytorch_model-00003-of-00014.safetensors",
@@ -186,12 +202,14 @@
186
  "transformer_blocks.16.attn.to_out.0.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
187
  "transformer_blocks.16.attn.to_q.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
188
  "transformer_blocks.16.attn.to_v.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
189
- "transformer_blocks.16.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
190
  "transformer_blocks.16.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
191
- "transformer_blocks.16.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
192
- "transformer_blocks.16.ff.net.2.weight_packed": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
193
  "transformer_blocks.16.ff.net.2.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
194
- "transformer_blocks.16.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
195
  "transformer_blocks.16.norm1.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
196
  "transformer_blocks.16.norm2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
197
  "transformer_blocks.17.adaln_proj.linear.bias": "diffusion_pytorch_model-00003-of-00014.safetensors",
@@ -203,12 +221,14 @@
203
  "transformer_blocks.17.attn.to_out.0.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
204
  "transformer_blocks.17.attn.to_q.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
205
  "transformer_blocks.17.attn.to_v.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
206
- "transformer_blocks.17.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
207
  "transformer_blocks.17.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
208
- "transformer_blocks.17.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
209
- "transformer_blocks.17.ff.net.2.weight_packed": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
210
  "transformer_blocks.17.ff.net.2.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
211
- "transformer_blocks.17.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
212
  "transformer_blocks.17.norm1.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
213
  "transformer_blocks.17.norm2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
214
  "transformer_blocks.18.adaln_proj.linear.bias": "diffusion_pytorch_model-00003-of-00014.safetensors",
@@ -220,12 +240,14 @@
220
  "transformer_blocks.18.attn.to_out.0.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
221
  "transformer_blocks.18.attn.to_q.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
222
  "transformer_blocks.18.attn.to_v.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
223
- "transformer_blocks.18.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
224
  "transformer_blocks.18.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
225
- "transformer_blocks.18.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
226
- "transformer_blocks.18.ff.net.2.weight_packed": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
227
  "transformer_blocks.18.ff.net.2.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
228
- "transformer_blocks.18.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
229
  "transformer_blocks.18.norm1.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
230
  "transformer_blocks.18.norm2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
231
  "transformer_blocks.19.adaln_proj.linear.bias": "diffusion_pytorch_model-00004-of-00014.safetensors",
@@ -237,12 +259,14 @@
237
  "transformer_blocks.19.attn.to_out.0.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
238
  "transformer_blocks.19.attn.to_q.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
239
  "transformer_blocks.19.attn.to_v.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
240
- "transformer_blocks.19.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
241
  "transformer_blocks.19.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
242
- "transformer_blocks.19.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
243
- "transformer_blocks.19.ff.net.2.weight_packed": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
244
  "transformer_blocks.19.ff.net.2.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
245
- "transformer_blocks.19.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
246
  "transformer_blocks.19.norm1.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
247
  "transformer_blocks.19.norm2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
248
  "transformer_blocks.2.adaln_proj.linear.bias": "diffusion_pytorch_model-00004-of-00014.safetensors",
@@ -254,12 +278,14 @@
254
  "transformer_blocks.2.attn.to_out.0.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
255
  "transformer_blocks.2.attn.to_q.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
256
  "transformer_blocks.2.attn.to_v.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
257
- "transformer_blocks.2.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
258
  "transformer_blocks.2.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
259
- "transformer_blocks.2.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
260
- "transformer_blocks.2.ff.net.2.weight_packed": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
261
  "transformer_blocks.2.ff.net.2.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
262
- "transformer_blocks.2.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
263
  "transformer_blocks.2.norm1.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
264
  "transformer_blocks.2.norm2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
265
  "transformer_blocks.20.adaln_proj.linear.bias": "diffusion_pytorch_model-00004-of-00014.safetensors",
@@ -271,12 +297,14 @@
271
  "transformer_blocks.20.attn.to_out.0.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
272
  "transformer_blocks.20.attn.to_q.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
273
  "transformer_blocks.20.attn.to_v.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
274
- "transformer_blocks.20.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
275
  "transformer_blocks.20.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
276
- "transformer_blocks.20.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
277
- "transformer_blocks.20.ff.net.2.weight_packed": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
278
  "transformer_blocks.20.ff.net.2.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
279
- "transformer_blocks.20.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
280
  "transformer_blocks.20.norm1.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
281
  "transformer_blocks.20.norm2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
282
  "transformer_blocks.21.adaln_proj.linear.bias": "diffusion_pytorch_model-00004-of-00014.safetensors",
@@ -288,12 +316,14 @@
288
  "transformer_blocks.21.attn.to_out.0.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
289
  "transformer_blocks.21.attn.to_q.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
290
  "transformer_blocks.21.attn.to_v.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
291
- "transformer_blocks.21.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00005-of-00014.safetensors",
 
292
  "transformer_blocks.21.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
293
- "transformer_blocks.21.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
294
- "transformer_blocks.21.ff.net.2.weight_packed": "diffusion_pytorch_model-00005-of-00014.safetensors",
 
295
  "transformer_blocks.21.ff.net.2.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
296
- "transformer_blocks.21.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
297
  "transformer_blocks.21.norm1.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
298
  "transformer_blocks.21.norm2.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
299
  "transformer_blocks.22.adaln_proj.linear.bias": "diffusion_pytorch_model-00005-of-00014.safetensors",
@@ -305,12 +335,14 @@
305
  "transformer_blocks.22.attn.to_out.0.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
306
  "transformer_blocks.22.attn.to_q.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
307
  "transformer_blocks.22.attn.to_v.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
308
- "transformer_blocks.22.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00005-of-00014.safetensors",
 
309
  "transformer_blocks.22.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
310
- "transformer_blocks.22.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
311
- "transformer_blocks.22.ff.net.2.weight_packed": "diffusion_pytorch_model-00005-of-00014.safetensors",
 
312
  "transformer_blocks.22.ff.net.2.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
313
- "transformer_blocks.22.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
314
  "transformer_blocks.22.norm1.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
315
  "transformer_blocks.22.norm2.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
316
  "transformer_blocks.23.adaln_proj.linear.bias": "diffusion_pytorch_model-00005-of-00014.safetensors",
@@ -322,12 +354,14 @@
322
  "transformer_blocks.23.attn.to_out.0.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
323
  "transformer_blocks.23.attn.to_q.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
324
  "transformer_blocks.23.attn.to_v.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
325
- "transformer_blocks.23.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00005-of-00014.safetensors",
 
326
  "transformer_blocks.23.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
327
- "transformer_blocks.23.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
328
- "transformer_blocks.23.ff.net.2.weight_packed": "diffusion_pytorch_model-00005-of-00014.safetensors",
 
329
  "transformer_blocks.23.ff.net.2.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
330
- "transformer_blocks.23.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
331
  "transformer_blocks.23.norm1.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
332
  "transformer_blocks.23.norm2.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
333
  "transformer_blocks.24.adaln_proj.linear.bias": "diffusion_pytorch_model-00005-of-00014.safetensors",
@@ -339,12 +373,14 @@
339
  "transformer_blocks.24.attn.to_out.0.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
340
  "transformer_blocks.24.attn.to_q.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
341
  "transformer_blocks.24.attn.to_v.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
342
- "transformer_blocks.24.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00005-of-00014.safetensors",
 
343
  "transformer_blocks.24.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
344
- "transformer_blocks.24.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
345
- "transformer_blocks.24.ff.net.2.weight_packed": "diffusion_pytorch_model-00006-of-00014.safetensors",
 
346
  "transformer_blocks.24.ff.net.2.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
347
- "transformer_blocks.24.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
348
  "transformer_blocks.24.norm1.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
349
  "transformer_blocks.24.norm2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
350
  "transformer_blocks.25.adaln_proj.linear.bias": "diffusion_pytorch_model-00006-of-00014.safetensors",
@@ -356,12 +392,14 @@
356
  "transformer_blocks.25.attn.to_out.0.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
357
  "transformer_blocks.25.attn.to_q.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
358
  "transformer_blocks.25.attn.to_v.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
359
- "transformer_blocks.25.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00006-of-00014.safetensors",
 
360
  "transformer_blocks.25.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
361
- "transformer_blocks.25.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
362
- "transformer_blocks.25.ff.net.2.weight_packed": "diffusion_pytorch_model-00006-of-00014.safetensors",
 
363
  "transformer_blocks.25.ff.net.2.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
364
- "transformer_blocks.25.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
365
  "transformer_blocks.25.norm1.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
366
  "transformer_blocks.25.norm2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
367
  "transformer_blocks.26.adaln_proj.linear.bias": "diffusion_pytorch_model-00006-of-00014.safetensors",
@@ -373,12 +411,14 @@
373
  "transformer_blocks.26.attn.to_out.0.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
374
  "transformer_blocks.26.attn.to_q.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
375
  "transformer_blocks.26.attn.to_v.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
376
- "transformer_blocks.26.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00006-of-00014.safetensors",
 
377
  "transformer_blocks.26.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
378
- "transformer_blocks.26.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
379
- "transformer_blocks.26.ff.net.2.weight_packed": "diffusion_pytorch_model-00006-of-00014.safetensors",
 
380
  "transformer_blocks.26.ff.net.2.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
381
- "transformer_blocks.26.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
382
  "transformer_blocks.26.norm1.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
383
  "transformer_blocks.26.norm2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
384
  "transformer_blocks.27.adaln_proj.linear.bias": "diffusion_pytorch_model-00006-of-00014.safetensors",
@@ -390,12 +430,14 @@
390
  "transformer_blocks.27.attn.to_out.0.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
391
  "transformer_blocks.27.attn.to_q.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
392
  "transformer_blocks.27.attn.to_v.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
393
- "transformer_blocks.27.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00006-of-00014.safetensors",
 
394
  "transformer_blocks.27.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
395
- "transformer_blocks.27.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
396
- "transformer_blocks.27.ff.net.2.weight_packed": "diffusion_pytorch_model-00006-of-00014.safetensors",
 
397
  "transformer_blocks.27.ff.net.2.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
398
- "transformer_blocks.27.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
399
  "transformer_blocks.27.norm1.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
400
  "transformer_blocks.27.norm2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
401
  "transformer_blocks.28.adaln_proj.linear.bias": "diffusion_pytorch_model-00006-of-00014.safetensors",
@@ -407,12 +449,14 @@
407
  "transformer_blocks.28.attn.to_out.0.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
408
  "transformer_blocks.28.attn.to_q.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
409
  "transformer_blocks.28.attn.to_v.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
410
- "transformer_blocks.28.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
411
  "transformer_blocks.28.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
412
- "transformer_blocks.28.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
413
- "transformer_blocks.28.ff.net.2.weight_packed": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
414
  "transformer_blocks.28.ff.net.2.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
415
- "transformer_blocks.28.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
416
  "transformer_blocks.28.norm1.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
417
  "transformer_blocks.28.norm2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
418
  "transformer_blocks.29.adaln_proj.linear.bias": "diffusion_pytorch_model-00007-of-00014.safetensors",
@@ -424,12 +468,14 @@
424
  "transformer_blocks.29.attn.to_out.0.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
425
  "transformer_blocks.29.attn.to_q.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
426
  "transformer_blocks.29.attn.to_v.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
427
- "transformer_blocks.29.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
428
  "transformer_blocks.29.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
429
- "transformer_blocks.29.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
430
- "transformer_blocks.29.ff.net.2.weight_packed": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
431
  "transformer_blocks.29.ff.net.2.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
432
- "transformer_blocks.29.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
433
  "transformer_blocks.29.norm1.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
434
  "transformer_blocks.29.norm2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
435
  "transformer_blocks.3.adaln_proj.linear.bias": "diffusion_pytorch_model-00007-of-00014.safetensors",
@@ -441,12 +487,14 @@
441
  "transformer_blocks.3.attn.to_out.0.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
442
  "transformer_blocks.3.attn.to_q.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
443
  "transformer_blocks.3.attn.to_v.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
444
- "transformer_blocks.3.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
445
  "transformer_blocks.3.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
446
- "transformer_blocks.3.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
447
- "transformer_blocks.3.ff.net.2.weight_packed": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
448
  "transformer_blocks.3.ff.net.2.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
449
- "transformer_blocks.3.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
450
  "transformer_blocks.3.norm1.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
451
  "transformer_blocks.3.norm2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
452
  "transformer_blocks.30.adaln_proj.linear.bias": "diffusion_pytorch_model-00007-of-00014.safetensors",
@@ -458,12 +506,14 @@
458
  "transformer_blocks.30.attn.to_out.0.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
459
  "transformer_blocks.30.attn.to_q.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
460
  "transformer_blocks.30.attn.to_v.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
461
- "transformer_blocks.30.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
462
  "transformer_blocks.30.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
463
- "transformer_blocks.30.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
464
- "transformer_blocks.30.ff.net.2.weight_packed": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
465
  "transformer_blocks.30.ff.net.2.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
466
- "transformer_blocks.30.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
467
  "transformer_blocks.30.norm1.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
468
  "transformer_blocks.30.norm2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
469
  "transformer_blocks.31.adaln_proj.linear.bias": "diffusion_pytorch_model-00007-of-00014.safetensors",
@@ -475,12 +525,14 @@
475
  "transformer_blocks.31.attn.to_out.0.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
476
  "transformer_blocks.31.attn.to_q.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
477
  "transformer_blocks.31.attn.to_v.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
478
- "transformer_blocks.31.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
479
  "transformer_blocks.31.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
480
- "transformer_blocks.31.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
481
- "transformer_blocks.31.ff.net.2.weight_packed": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
482
  "transformer_blocks.31.ff.net.2.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
483
- "transformer_blocks.31.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
484
  "transformer_blocks.31.norm1.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
485
  "transformer_blocks.31.norm2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
486
  "transformer_blocks.32.adaln_proj.linear.bias": "diffusion_pytorch_model-00008-of-00014.safetensors",
@@ -492,12 +544,14 @@
492
  "transformer_blocks.32.attn.to_out.0.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
493
  "transformer_blocks.32.attn.to_q.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
494
  "transformer_blocks.32.attn.to_v.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
495
- "transformer_blocks.32.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
496
  "transformer_blocks.32.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
497
- "transformer_blocks.32.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
498
- "transformer_blocks.32.ff.net.2.weight_packed": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
499
  "transformer_blocks.32.ff.net.2.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
500
- "transformer_blocks.32.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
501
  "transformer_blocks.32.norm1.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
502
  "transformer_blocks.32.norm2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
503
  "transformer_blocks.33.adaln_proj.linear.bias": "diffusion_pytorch_model-00008-of-00014.safetensors",
@@ -509,12 +563,14 @@
509
  "transformer_blocks.33.attn.to_out.0.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
510
  "transformer_blocks.33.attn.to_q.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
511
  "transformer_blocks.33.attn.to_v.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
512
- "transformer_blocks.33.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
513
  "transformer_blocks.33.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
514
- "transformer_blocks.33.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
515
- "transformer_blocks.33.ff.net.2.weight_packed": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
516
  "transformer_blocks.33.ff.net.2.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
517
- "transformer_blocks.33.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
518
  "transformer_blocks.33.norm1.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
519
  "transformer_blocks.33.norm2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
520
  "transformer_blocks.34.adaln_proj.linear.bias": "diffusion_pytorch_model-00008-of-00014.safetensors",
@@ -526,12 +582,14 @@
526
  "transformer_blocks.34.attn.to_out.0.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
527
  "transformer_blocks.34.attn.to_q.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
528
  "transformer_blocks.34.attn.to_v.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
529
- "transformer_blocks.34.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
530
  "transformer_blocks.34.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
531
- "transformer_blocks.34.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
532
- "transformer_blocks.34.ff.net.2.weight_packed": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
533
  "transformer_blocks.34.ff.net.2.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
534
- "transformer_blocks.34.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
535
  "transformer_blocks.34.norm1.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
536
  "transformer_blocks.34.norm2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
537
  "transformer_blocks.35.adaln_proj.linear.bias": "diffusion_pytorch_model-00008-of-00014.safetensors",
@@ -543,12 +601,14 @@
543
  "transformer_blocks.35.attn.to_out.0.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
544
  "transformer_blocks.35.attn.to_q.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
545
  "transformer_blocks.35.attn.to_v.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
546
- "transformer_blocks.35.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
547
  "transformer_blocks.35.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
548
- "transformer_blocks.35.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
549
- "transformer_blocks.35.ff.net.2.weight_packed": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
550
  "transformer_blocks.35.ff.net.2.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
551
- "transformer_blocks.35.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
552
  "transformer_blocks.35.norm1.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
553
  "transformer_blocks.35.norm2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
554
  "transformer_blocks.36.adaln_proj.linear.bias": "diffusion_pytorch_model-00009-of-00014.safetensors",
@@ -560,12 +620,14 @@
560
  "transformer_blocks.36.attn.to_out.0.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
561
  "transformer_blocks.36.attn.to_q.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
562
  "transformer_blocks.36.attn.to_v.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
563
- "transformer_blocks.36.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
564
  "transformer_blocks.36.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
565
- "transformer_blocks.36.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
566
- "transformer_blocks.36.ff.net.2.weight_packed": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
567
  "transformer_blocks.36.ff.net.2.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
568
- "transformer_blocks.36.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
569
  "transformer_blocks.36.norm1.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
570
  "transformer_blocks.36.norm2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
571
  "transformer_blocks.37.adaln_proj.linear.bias": "diffusion_pytorch_model-00009-of-00014.safetensors",
@@ -577,12 +639,14 @@
577
  "transformer_blocks.37.attn.to_out.0.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
578
  "transformer_blocks.37.attn.to_q.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
579
  "transformer_blocks.37.attn.to_v.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
580
- "transformer_blocks.37.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
581
  "transformer_blocks.37.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
582
- "transformer_blocks.37.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
583
- "transformer_blocks.37.ff.net.2.weight_packed": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
584
  "transformer_blocks.37.ff.net.2.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
585
- "transformer_blocks.37.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
586
  "transformer_blocks.37.norm1.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
587
  "transformer_blocks.37.norm2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
588
  "transformer_blocks.38.adaln_proj.linear.bias": "diffusion_pytorch_model-00009-of-00014.safetensors",
@@ -594,12 +658,14 @@
594
  "transformer_blocks.38.attn.to_out.0.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
595
  "transformer_blocks.38.attn.to_q.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
596
  "transformer_blocks.38.attn.to_v.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
597
- "transformer_blocks.38.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
598
  "transformer_blocks.38.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
599
- "transformer_blocks.38.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
600
- "transformer_blocks.38.ff.net.2.weight_packed": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
601
  "transformer_blocks.38.ff.net.2.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
602
- "transformer_blocks.38.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
603
  "transformer_blocks.38.norm1.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
604
  "transformer_blocks.38.norm2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
605
  "transformer_blocks.39.adaln_proj.linear.bias": "diffusion_pytorch_model-00009-of-00014.safetensors",
@@ -611,12 +677,14 @@
611
  "transformer_blocks.39.attn.to_out.0.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
612
  "transformer_blocks.39.attn.to_q.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
613
  "transformer_blocks.39.attn.to_v.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
614
- "transformer_blocks.39.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00010-of-00014.safetensors",
 
615
  "transformer_blocks.39.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
616
- "transformer_blocks.39.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
617
- "transformer_blocks.39.ff.net.2.weight_packed": "diffusion_pytorch_model-00010-of-00014.safetensors",
 
618
  "transformer_blocks.39.ff.net.2.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
619
- "transformer_blocks.39.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
620
  "transformer_blocks.39.norm1.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
621
  "transformer_blocks.39.norm2.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
622
  "transformer_blocks.4.adaln_proj.linear.bias": "diffusion_pytorch_model-00010-of-00014.safetensors",
@@ -628,12 +696,14 @@
628
  "transformer_blocks.4.attn.to_out.0.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
629
  "transformer_blocks.4.attn.to_q.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
630
  "transformer_blocks.4.attn.to_v.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
631
- "transformer_blocks.4.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00010-of-00014.safetensors",
 
632
  "transformer_blocks.4.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
633
- "transformer_blocks.4.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
634
- "transformer_blocks.4.ff.net.2.weight_packed": "diffusion_pytorch_model-00010-of-00014.safetensors",
 
635
  "transformer_blocks.4.ff.net.2.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
636
- "transformer_blocks.4.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
637
  "transformer_blocks.4.norm1.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
638
  "transformer_blocks.4.norm2.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
639
  "transformer_blocks.40.adaln_proj.linear.bias": "diffusion_pytorch_model-00010-of-00014.safetensors",
@@ -645,12 +715,14 @@
645
  "transformer_blocks.40.attn.to_out.0.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
646
  "transformer_blocks.40.attn.to_q.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
647
  "transformer_blocks.40.attn.to_v.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
648
- "transformer_blocks.40.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00010-of-00014.safetensors",
 
649
  "transformer_blocks.40.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
650
- "transformer_blocks.40.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
651
- "transformer_blocks.40.ff.net.2.weight_packed": "diffusion_pytorch_model-00010-of-00014.safetensors",
 
652
  "transformer_blocks.40.ff.net.2.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
653
- "transformer_blocks.40.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
654
  "transformer_blocks.40.norm1.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
655
  "transformer_blocks.40.norm2.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
656
  "transformer_blocks.41.adaln_proj.linear.bias": "diffusion_pytorch_model-00010-of-00014.safetensors",
@@ -662,12 +734,14 @@
662
  "transformer_blocks.41.attn.to_out.0.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
663
  "transformer_blocks.41.attn.to_q.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
664
  "transformer_blocks.41.attn.to_v.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
665
- "transformer_blocks.41.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00010-of-00014.safetensors",
 
666
  "transformer_blocks.41.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
667
- "transformer_blocks.41.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
668
- "transformer_blocks.41.ff.net.2.weight_packed": "diffusion_pytorch_model-00011-of-00014.safetensors",
 
669
  "transformer_blocks.41.ff.net.2.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
670
- "transformer_blocks.41.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
671
  "transformer_blocks.41.norm1.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
672
  "transformer_blocks.41.norm2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
673
  "transformer_blocks.42.adaln_proj.linear.bias": "diffusion_pytorch_model-00011-of-00014.safetensors",
@@ -679,12 +753,14 @@
679
  "transformer_blocks.42.attn.to_out.0.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
680
  "transformer_blocks.42.attn.to_q.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
681
  "transformer_blocks.42.attn.to_v.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
682
- "transformer_blocks.42.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00011-of-00014.safetensors",
 
683
  "transformer_blocks.42.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
684
- "transformer_blocks.42.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
685
- "transformer_blocks.42.ff.net.2.weight_packed": "diffusion_pytorch_model-00011-of-00014.safetensors",
 
686
  "transformer_blocks.42.ff.net.2.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
687
- "transformer_blocks.42.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
688
  "transformer_blocks.42.norm1.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
689
  "transformer_blocks.42.norm2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
690
  "transformer_blocks.43.adaln_proj.linear.bias": "diffusion_pytorch_model-00011-of-00014.safetensors",
@@ -696,12 +772,14 @@
696
  "transformer_blocks.43.attn.to_out.0.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
697
  "transformer_blocks.43.attn.to_q.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
698
  "transformer_blocks.43.attn.to_v.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
699
- "transformer_blocks.43.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00011-of-00014.safetensors",
 
700
  "transformer_blocks.43.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
701
- "transformer_blocks.43.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
702
- "transformer_blocks.43.ff.net.2.weight_packed": "diffusion_pytorch_model-00011-of-00014.safetensors",
 
703
  "transformer_blocks.43.ff.net.2.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
704
- "transformer_blocks.43.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
705
  "transformer_blocks.43.norm1.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
706
  "transformer_blocks.43.norm2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
707
  "transformer_blocks.44.adaln_proj.linear.bias": "diffusion_pytorch_model-00011-of-00014.safetensors",
@@ -713,12 +791,14 @@
713
  "transformer_blocks.44.attn.to_out.0.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
714
  "transformer_blocks.44.attn.to_q.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
715
  "transformer_blocks.44.attn.to_v.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
716
- "transformer_blocks.44.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00011-of-00014.safetensors",
 
717
  "transformer_blocks.44.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
718
- "transformer_blocks.44.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
719
- "transformer_blocks.44.ff.net.2.weight_packed": "diffusion_pytorch_model-00011-of-00014.safetensors",
 
720
  "transformer_blocks.44.ff.net.2.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
721
- "transformer_blocks.44.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
722
  "transformer_blocks.44.norm1.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
723
  "transformer_blocks.44.norm2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
724
  "transformer_blocks.45.adaln_proj.linear.bias": "diffusion_pytorch_model-00011-of-00014.safetensors",
@@ -730,12 +810,14 @@
730
  "transformer_blocks.45.attn.to_out.0.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
731
  "transformer_blocks.45.attn.to_q.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
732
  "transformer_blocks.45.attn.to_v.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
733
- "transformer_blocks.45.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
734
  "transformer_blocks.45.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
735
- "transformer_blocks.45.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
736
- "transformer_blocks.45.ff.net.2.weight_packed": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
737
  "transformer_blocks.45.ff.net.2.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
738
- "transformer_blocks.45.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
739
  "transformer_blocks.45.norm1.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
740
  "transformer_blocks.45.norm2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
741
  "transformer_blocks.46.adaln_proj.linear.bias": "diffusion_pytorch_model-00012-of-00014.safetensors",
@@ -747,12 +829,14 @@
747
  "transformer_blocks.46.attn.to_out.0.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
748
  "transformer_blocks.46.attn.to_q.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
749
  "transformer_blocks.46.attn.to_v.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
750
- "transformer_blocks.46.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
751
  "transformer_blocks.46.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
752
- "transformer_blocks.46.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
753
- "transformer_blocks.46.ff.net.2.weight_packed": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
754
  "transformer_blocks.46.ff.net.2.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
755
- "transformer_blocks.46.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
756
  "transformer_blocks.46.norm1.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
757
  "transformer_blocks.46.norm2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
758
  "transformer_blocks.47.adaln_proj.linear.bias": "diffusion_pytorch_model-00012-of-00014.safetensors",
@@ -764,12 +848,14 @@
764
  "transformer_blocks.47.attn.to_out.0.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
765
  "transformer_blocks.47.attn.to_q.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
766
  "transformer_blocks.47.attn.to_v.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
767
- "transformer_blocks.47.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
768
  "transformer_blocks.47.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
769
- "transformer_blocks.47.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
770
- "transformer_blocks.47.ff.net.2.weight_packed": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
771
  "transformer_blocks.47.ff.net.2.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
772
- "transformer_blocks.47.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
773
  "transformer_blocks.47.norm1.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
774
  "transformer_blocks.47.norm2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
775
  "transformer_blocks.48.adaln_proj.linear.bias": "diffusion_pytorch_model-00012-of-00014.safetensors",
@@ -781,12 +867,14 @@
781
  "transformer_blocks.48.attn.to_out.0.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
782
  "transformer_blocks.48.attn.to_q.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
783
  "transformer_blocks.48.attn.to_v.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
784
- "transformer_blocks.48.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
785
  "transformer_blocks.48.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
786
- "transformer_blocks.48.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
787
- "transformer_blocks.48.ff.net.2.weight_packed": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
788
  "transformer_blocks.48.ff.net.2.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
789
- "transformer_blocks.48.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
790
  "transformer_blocks.48.norm1.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
791
  "transformer_blocks.48.norm2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
792
  "transformer_blocks.49.adaln_proj.linear.bias": "diffusion_pytorch_model-00012-of-00014.safetensors",
@@ -798,12 +886,14 @@
798
  "transformer_blocks.49.attn.to_out.0.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
799
  "transformer_blocks.49.attn.to_q.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
800
  "transformer_blocks.49.attn.to_v.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
801
- "transformer_blocks.49.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
802
  "transformer_blocks.49.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
803
- "transformer_blocks.49.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
804
- "transformer_blocks.49.ff.net.2.weight_packed": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
805
  "transformer_blocks.49.ff.net.2.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
806
- "transformer_blocks.49.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
807
  "transformer_blocks.49.norm1.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
808
  "transformer_blocks.49.norm2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
809
  "transformer_blocks.5.adaln_proj.linear.bias": "diffusion_pytorch_model-00013-of-00014.safetensors",
@@ -815,12 +905,14 @@
815
  "transformer_blocks.5.attn.to_out.0.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
816
  "transformer_blocks.5.attn.to_q.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
817
  "transformer_blocks.5.attn.to_v.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
818
- "transformer_blocks.5.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
819
  "transformer_blocks.5.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
820
- "transformer_blocks.5.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
821
- "transformer_blocks.5.ff.net.2.weight_packed": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
822
  "transformer_blocks.5.ff.net.2.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
823
- "transformer_blocks.5.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
824
  "transformer_blocks.5.norm1.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
825
  "transformer_blocks.5.norm2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
826
  "transformer_blocks.6.adaln_proj.linear.bias": "diffusion_pytorch_model-00013-of-00014.safetensors",
@@ -832,12 +924,14 @@
832
  "transformer_blocks.6.attn.to_out.0.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
833
  "transformer_blocks.6.attn.to_q.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
834
  "transformer_blocks.6.attn.to_v.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
835
- "transformer_blocks.6.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
836
  "transformer_blocks.6.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
837
- "transformer_blocks.6.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
838
- "transformer_blocks.6.ff.net.2.weight_packed": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
839
  "transformer_blocks.6.ff.net.2.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
840
- "transformer_blocks.6.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
841
  "transformer_blocks.6.norm1.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
842
  "transformer_blocks.6.norm2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
843
  "transformer_blocks.7.adaln_proj.linear.bias": "diffusion_pytorch_model-00013-of-00014.safetensors",
@@ -849,12 +943,14 @@
849
  "transformer_blocks.7.attn.to_out.0.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
850
  "transformer_blocks.7.attn.to_q.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
851
  "transformer_blocks.7.attn.to_v.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
852
- "transformer_blocks.7.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
853
  "transformer_blocks.7.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
854
- "transformer_blocks.7.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
855
- "transformer_blocks.7.ff.net.2.weight_packed": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
856
  "transformer_blocks.7.ff.net.2.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
857
- "transformer_blocks.7.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
858
  "transformer_blocks.7.norm1.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
859
  "transformer_blocks.7.norm2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
860
  "transformer_blocks.8.adaln_proj.linear.bias": "diffusion_pytorch_model-00013-of-00014.safetensors",
@@ -866,12 +962,14 @@
866
  "transformer_blocks.8.attn.to_out.0.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
867
  "transformer_blocks.8.attn.to_q.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
868
  "transformer_blocks.8.attn.to_v.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
869
- "transformer_blocks.8.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00014-of-00014.safetensors",
 
870
  "transformer_blocks.8.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
871
- "transformer_blocks.8.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
872
- "transformer_blocks.8.ff.net.2.weight_packed": "diffusion_pytorch_model-00014-of-00014.safetensors",
 
873
  "transformer_blocks.8.ff.net.2.weight_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
874
- "transformer_blocks.8.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
875
  "transformer_blocks.8.norm1.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
876
  "transformer_blocks.8.norm2.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
877
  "transformer_blocks.9.adaln_proj.linear.bias": "diffusion_pytorch_model-00014-of-00014.safetensors",
@@ -883,12 +981,14 @@
883
  "transformer_blocks.9.attn.to_out.0.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
884
  "transformer_blocks.9.attn.to_q.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
885
  "transformer_blocks.9.attn.to_v.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
886
- "transformer_blocks.9.ff.net.0.proj.weight_packed": "diffusion_pytorch_model-00014-of-00014.safetensors",
 
887
  "transformer_blocks.9.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
888
- "transformer_blocks.9.ff.net.0.proj.weight_tensor_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
889
- "transformer_blocks.9.ff.net.2.weight_packed": "diffusion_pytorch_model-00014-of-00014.safetensors",
 
890
  "transformer_blocks.9.ff.net.2.weight_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
891
- "transformer_blocks.9.ff.net.2.weight_tensor_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
892
  "transformer_blocks.9.norm1.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
893
  "transformer_blocks.9.norm2.weight": "diffusion_pytorch_model-00014-of-00014.safetensors"
894
  }
 
50
  "transformer_blocks.0.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
51
  "transformer_blocks.0.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
52
  "transformer_blocks.0.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
53
+ "transformer_blocks.0.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
54
+ "transformer_blocks.0.ff.net.0.proj.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
55
  "transformer_blocks.0.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
56
+ "transformer_blocks.0.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00001-of-00014.safetensors",
57
+ "transformer_blocks.0.ff.net.2.input_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
58
+ "transformer_blocks.0.ff.net.2.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
59
  "transformer_blocks.0.ff.net.2.weight_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
60
+ "transformer_blocks.0.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00001-of-00014.safetensors",
61
  "transformer_blocks.0.norm1.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
62
  "transformer_blocks.0.norm2.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
63
  "transformer_blocks.1.adaln_proj.linear.bias": "diffusion_pytorch_model-00001-of-00014.safetensors",
 
69
  "transformer_blocks.1.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
70
  "transformer_blocks.1.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
71
  "transformer_blocks.1.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
72
+ "transformer_blocks.1.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
73
+ "transformer_blocks.1.ff.net.0.proj.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
74
  "transformer_blocks.1.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
75
+ "transformer_blocks.1.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00001-of-00014.safetensors",
76
+ "transformer_blocks.1.ff.net.2.input_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
77
+ "transformer_blocks.1.ff.net.2.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
78
  "transformer_blocks.1.ff.net.2.weight_scale": "diffusion_pytorch_model-00001-of-00014.safetensors",
79
+ "transformer_blocks.1.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00001-of-00014.safetensors",
80
  "transformer_blocks.1.norm1.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
81
  "transformer_blocks.1.norm2.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
82
  "transformer_blocks.10.adaln_proj.linear.bias": "diffusion_pytorch_model-00001-of-00014.safetensors",
 
88
  "transformer_blocks.10.attn.to_out.0.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
89
  "transformer_blocks.10.attn.to_q.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
90
  "transformer_blocks.10.attn.to_v.weight": "diffusion_pytorch_model-00001-of-00014.safetensors",
91
+ "transformer_blocks.10.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
92
+ "transformer_blocks.10.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
93
  "transformer_blocks.10.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
94
+ "transformer_blocks.10.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00002-of-00014.safetensors",
95
+ "transformer_blocks.10.ff.net.2.input_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
96
+ "transformer_blocks.10.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
97
  "transformer_blocks.10.ff.net.2.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
98
+ "transformer_blocks.10.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00002-of-00014.safetensors",
99
  "transformer_blocks.10.norm1.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
100
  "transformer_blocks.10.norm2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
101
  "transformer_blocks.11.adaln_proj.linear.bias": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
107
  "transformer_blocks.11.attn.to_out.0.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
108
  "transformer_blocks.11.attn.to_q.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
109
  "transformer_blocks.11.attn.to_v.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
110
+ "transformer_blocks.11.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
111
+ "transformer_blocks.11.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
112
  "transformer_blocks.11.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
113
+ "transformer_blocks.11.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00002-of-00014.safetensors",
114
+ "transformer_blocks.11.ff.net.2.input_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
115
+ "transformer_blocks.11.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
116
  "transformer_blocks.11.ff.net.2.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
117
+ "transformer_blocks.11.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00002-of-00014.safetensors",
118
  "transformer_blocks.11.norm1.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
119
  "transformer_blocks.11.norm2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
120
  "transformer_blocks.12.adaln_proj.linear.bias": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
126
  "transformer_blocks.12.attn.to_out.0.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
127
  "transformer_blocks.12.attn.to_q.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
128
  "transformer_blocks.12.attn.to_v.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
129
+ "transformer_blocks.12.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
130
+ "transformer_blocks.12.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
131
  "transformer_blocks.12.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
132
+ "transformer_blocks.12.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00002-of-00014.safetensors",
133
+ "transformer_blocks.12.ff.net.2.input_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
134
+ "transformer_blocks.12.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
135
  "transformer_blocks.12.ff.net.2.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
136
+ "transformer_blocks.12.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00002-of-00014.safetensors",
137
  "transformer_blocks.12.norm1.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
138
  "transformer_blocks.12.norm2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
139
  "transformer_blocks.13.adaln_proj.linear.bias": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
145
  "transformer_blocks.13.attn.to_out.0.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
146
  "transformer_blocks.13.attn.to_q.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
147
  "transformer_blocks.13.attn.to_v.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
148
+ "transformer_blocks.13.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
149
+ "transformer_blocks.13.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
150
  "transformer_blocks.13.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
151
+ "transformer_blocks.13.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00002-of-00014.safetensors",
152
+ "transformer_blocks.13.ff.net.2.input_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
153
+ "transformer_blocks.13.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
154
  "transformer_blocks.13.ff.net.2.weight_scale": "diffusion_pytorch_model-00002-of-00014.safetensors",
155
+ "transformer_blocks.13.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00002-of-00014.safetensors",
156
  "transformer_blocks.13.norm1.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
157
  "transformer_blocks.13.norm2.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
158
  "transformer_blocks.14.adaln_proj.linear.bias": "diffusion_pytorch_model-00002-of-00014.safetensors",
 
164
  "transformer_blocks.14.attn.to_out.0.weight": "diffusion_pytorch_model-00002-of-00014.safetensors",
165
  "transformer_blocks.14.attn.to_q.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
166
  "transformer_blocks.14.attn.to_v.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
167
+ "transformer_blocks.14.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
168
+ "transformer_blocks.14.ff.net.0.proj.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
169
  "transformer_blocks.14.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
170
+ "transformer_blocks.14.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00003-of-00014.safetensors",
171
+ "transformer_blocks.14.ff.net.2.input_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
172
+ "transformer_blocks.14.ff.net.2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
173
  "transformer_blocks.14.ff.net.2.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
174
+ "transformer_blocks.14.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00003-of-00014.safetensors",
175
  "transformer_blocks.14.norm1.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
176
  "transformer_blocks.14.norm2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
177
  "transformer_blocks.15.adaln_proj.linear.bias": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
183
  "transformer_blocks.15.attn.to_out.0.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
184
  "transformer_blocks.15.attn.to_q.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
185
  "transformer_blocks.15.attn.to_v.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
186
+ "transformer_blocks.15.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
187
+ "transformer_blocks.15.ff.net.0.proj.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
188
  "transformer_blocks.15.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
189
+ "transformer_blocks.15.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00003-of-00014.safetensors",
190
+ "transformer_blocks.15.ff.net.2.input_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
191
+ "transformer_blocks.15.ff.net.2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
192
  "transformer_blocks.15.ff.net.2.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
193
+ "transformer_blocks.15.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00003-of-00014.safetensors",
194
  "transformer_blocks.15.norm1.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
195
  "transformer_blocks.15.norm2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
196
  "transformer_blocks.16.adaln_proj.linear.bias": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
202
  "transformer_blocks.16.attn.to_out.0.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
203
  "transformer_blocks.16.attn.to_q.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
204
  "transformer_blocks.16.attn.to_v.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
205
+ "transformer_blocks.16.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
206
+ "transformer_blocks.16.ff.net.0.proj.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
207
  "transformer_blocks.16.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
208
+ "transformer_blocks.16.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00003-of-00014.safetensors",
209
+ "transformer_blocks.16.ff.net.2.input_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
210
+ "transformer_blocks.16.ff.net.2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
211
  "transformer_blocks.16.ff.net.2.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
212
+ "transformer_blocks.16.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00003-of-00014.safetensors",
213
  "transformer_blocks.16.norm1.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
214
  "transformer_blocks.16.norm2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
215
  "transformer_blocks.17.adaln_proj.linear.bias": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
221
  "transformer_blocks.17.attn.to_out.0.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
222
  "transformer_blocks.17.attn.to_q.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
223
  "transformer_blocks.17.attn.to_v.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
224
+ "transformer_blocks.17.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
225
+ "transformer_blocks.17.ff.net.0.proj.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
226
  "transformer_blocks.17.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
227
+ "transformer_blocks.17.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00003-of-00014.safetensors",
228
+ "transformer_blocks.17.ff.net.2.input_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
229
+ "transformer_blocks.17.ff.net.2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
230
  "transformer_blocks.17.ff.net.2.weight_scale": "diffusion_pytorch_model-00003-of-00014.safetensors",
231
+ "transformer_blocks.17.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00003-of-00014.safetensors",
232
  "transformer_blocks.17.norm1.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
233
  "transformer_blocks.17.norm2.weight": "diffusion_pytorch_model-00003-of-00014.safetensors",
234
  "transformer_blocks.18.adaln_proj.linear.bias": "diffusion_pytorch_model-00003-of-00014.safetensors",
 
240
  "transformer_blocks.18.attn.to_out.0.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
241
  "transformer_blocks.18.attn.to_q.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
242
  "transformer_blocks.18.attn.to_v.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
243
+ "transformer_blocks.18.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
244
+ "transformer_blocks.18.ff.net.0.proj.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
245
  "transformer_blocks.18.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
246
+ "transformer_blocks.18.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00004-of-00014.safetensors",
247
+ "transformer_blocks.18.ff.net.2.input_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
248
+ "transformer_blocks.18.ff.net.2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
249
  "transformer_blocks.18.ff.net.2.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
250
+ "transformer_blocks.18.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00004-of-00014.safetensors",
251
  "transformer_blocks.18.norm1.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
252
  "transformer_blocks.18.norm2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
253
  "transformer_blocks.19.adaln_proj.linear.bias": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
259
  "transformer_blocks.19.attn.to_out.0.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
260
  "transformer_blocks.19.attn.to_q.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
261
  "transformer_blocks.19.attn.to_v.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
262
+ "transformer_blocks.19.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
263
+ "transformer_blocks.19.ff.net.0.proj.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
264
  "transformer_blocks.19.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
265
+ "transformer_blocks.19.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00004-of-00014.safetensors",
266
+ "transformer_blocks.19.ff.net.2.input_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
267
+ "transformer_blocks.19.ff.net.2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
268
  "transformer_blocks.19.ff.net.2.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
269
+ "transformer_blocks.19.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00004-of-00014.safetensors",
270
  "transformer_blocks.19.norm1.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
271
  "transformer_blocks.19.norm2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
272
  "transformer_blocks.2.adaln_proj.linear.bias": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
278
  "transformer_blocks.2.attn.to_out.0.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
279
  "transformer_blocks.2.attn.to_q.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
280
  "transformer_blocks.2.attn.to_v.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
281
+ "transformer_blocks.2.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
282
+ "transformer_blocks.2.ff.net.0.proj.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
283
  "transformer_blocks.2.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
284
+ "transformer_blocks.2.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00004-of-00014.safetensors",
285
+ "transformer_blocks.2.ff.net.2.input_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
286
+ "transformer_blocks.2.ff.net.2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
287
  "transformer_blocks.2.ff.net.2.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
288
+ "transformer_blocks.2.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00004-of-00014.safetensors",
289
  "transformer_blocks.2.norm1.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
290
  "transformer_blocks.2.norm2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
291
  "transformer_blocks.20.adaln_proj.linear.bias": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
297
  "transformer_blocks.20.attn.to_out.0.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
298
  "transformer_blocks.20.attn.to_q.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
299
  "transformer_blocks.20.attn.to_v.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
300
+ "transformer_blocks.20.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
301
+ "transformer_blocks.20.ff.net.0.proj.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
302
  "transformer_blocks.20.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
303
+ "transformer_blocks.20.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00004-of-00014.safetensors",
304
+ "transformer_blocks.20.ff.net.2.input_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
305
+ "transformer_blocks.20.ff.net.2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
306
  "transformer_blocks.20.ff.net.2.weight_scale": "diffusion_pytorch_model-00004-of-00014.safetensors",
307
+ "transformer_blocks.20.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00004-of-00014.safetensors",
308
  "transformer_blocks.20.norm1.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
309
  "transformer_blocks.20.norm2.weight": "diffusion_pytorch_model-00004-of-00014.safetensors",
310
  "transformer_blocks.21.adaln_proj.linear.bias": "diffusion_pytorch_model-00004-of-00014.safetensors",
 
316
  "transformer_blocks.21.attn.to_out.0.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
317
  "transformer_blocks.21.attn.to_q.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
318
  "transformer_blocks.21.attn.to_v.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
319
+ "transformer_blocks.21.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
320
+ "transformer_blocks.21.ff.net.0.proj.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
321
  "transformer_blocks.21.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
322
+ "transformer_blocks.21.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00005-of-00014.safetensors",
323
+ "transformer_blocks.21.ff.net.2.input_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
324
+ "transformer_blocks.21.ff.net.2.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
325
  "transformer_blocks.21.ff.net.2.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
326
+ "transformer_blocks.21.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00005-of-00014.safetensors",
327
  "transformer_blocks.21.norm1.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
328
  "transformer_blocks.21.norm2.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
329
  "transformer_blocks.22.adaln_proj.linear.bias": "diffusion_pytorch_model-00005-of-00014.safetensors",
 
335
  "transformer_blocks.22.attn.to_out.0.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
336
  "transformer_blocks.22.attn.to_q.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
337
  "transformer_blocks.22.attn.to_v.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
338
+ "transformer_blocks.22.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
339
+ "transformer_blocks.22.ff.net.0.proj.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
340
  "transformer_blocks.22.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
341
+ "transformer_blocks.22.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00005-of-00014.safetensors",
342
+ "transformer_blocks.22.ff.net.2.input_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
343
+ "transformer_blocks.22.ff.net.2.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
344
  "transformer_blocks.22.ff.net.2.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
345
+ "transformer_blocks.22.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00005-of-00014.safetensors",
346
  "transformer_blocks.22.norm1.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
347
  "transformer_blocks.22.norm2.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
348
  "transformer_blocks.23.adaln_proj.linear.bias": "diffusion_pytorch_model-00005-of-00014.safetensors",
 
354
  "transformer_blocks.23.attn.to_out.0.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
355
  "transformer_blocks.23.attn.to_q.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
356
  "transformer_blocks.23.attn.to_v.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
357
+ "transformer_blocks.23.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
358
+ "transformer_blocks.23.ff.net.0.proj.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
359
  "transformer_blocks.23.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
360
+ "transformer_blocks.23.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00005-of-00014.safetensors",
361
+ "transformer_blocks.23.ff.net.2.input_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
362
+ "transformer_blocks.23.ff.net.2.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
363
  "transformer_blocks.23.ff.net.2.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
364
+ "transformer_blocks.23.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00005-of-00014.safetensors",
365
  "transformer_blocks.23.norm1.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
366
  "transformer_blocks.23.norm2.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
367
  "transformer_blocks.24.adaln_proj.linear.bias": "diffusion_pytorch_model-00005-of-00014.safetensors",
 
373
  "transformer_blocks.24.attn.to_out.0.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
374
  "transformer_blocks.24.attn.to_q.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
375
  "transformer_blocks.24.attn.to_v.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
376
+ "transformer_blocks.24.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
377
+ "transformer_blocks.24.ff.net.0.proj.weight": "diffusion_pytorch_model-00005-of-00014.safetensors",
378
  "transformer_blocks.24.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00005-of-00014.safetensors",
379
+ "transformer_blocks.24.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00005-of-00014.safetensors",
380
+ "transformer_blocks.24.ff.net.2.input_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
381
+ "transformer_blocks.24.ff.net.2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
382
  "transformer_blocks.24.ff.net.2.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
383
+ "transformer_blocks.24.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00006-of-00014.safetensors",
384
  "transformer_blocks.24.norm1.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
385
  "transformer_blocks.24.norm2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
386
  "transformer_blocks.25.adaln_proj.linear.bias": "diffusion_pytorch_model-00006-of-00014.safetensors",
 
392
  "transformer_blocks.25.attn.to_out.0.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
393
  "transformer_blocks.25.attn.to_q.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
394
  "transformer_blocks.25.attn.to_v.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
395
+ "transformer_blocks.25.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
396
+ "transformer_blocks.25.ff.net.0.proj.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
397
  "transformer_blocks.25.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
398
+ "transformer_blocks.25.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00006-of-00014.safetensors",
399
+ "transformer_blocks.25.ff.net.2.input_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
400
+ "transformer_blocks.25.ff.net.2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
401
  "transformer_blocks.25.ff.net.2.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
402
+ "transformer_blocks.25.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00006-of-00014.safetensors",
403
  "transformer_blocks.25.norm1.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
404
  "transformer_blocks.25.norm2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
405
  "transformer_blocks.26.adaln_proj.linear.bias": "diffusion_pytorch_model-00006-of-00014.safetensors",
 
411
  "transformer_blocks.26.attn.to_out.0.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
412
  "transformer_blocks.26.attn.to_q.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
413
  "transformer_blocks.26.attn.to_v.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
414
+ "transformer_blocks.26.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
415
+ "transformer_blocks.26.ff.net.0.proj.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
416
  "transformer_blocks.26.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
417
+ "transformer_blocks.26.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00006-of-00014.safetensors",
418
+ "transformer_blocks.26.ff.net.2.input_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
419
+ "transformer_blocks.26.ff.net.2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
420
  "transformer_blocks.26.ff.net.2.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
421
+ "transformer_blocks.26.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00006-of-00014.safetensors",
422
  "transformer_blocks.26.norm1.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
423
  "transformer_blocks.26.norm2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
424
  "transformer_blocks.27.adaln_proj.linear.bias": "diffusion_pytorch_model-00006-of-00014.safetensors",
 
430
  "transformer_blocks.27.attn.to_out.0.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
431
  "transformer_blocks.27.attn.to_q.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
432
  "transformer_blocks.27.attn.to_v.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
433
+ "transformer_blocks.27.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
434
+ "transformer_blocks.27.ff.net.0.proj.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
435
  "transformer_blocks.27.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
436
+ "transformer_blocks.27.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00006-of-00014.safetensors",
437
+ "transformer_blocks.27.ff.net.2.input_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
438
+ "transformer_blocks.27.ff.net.2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
439
  "transformer_blocks.27.ff.net.2.weight_scale": "diffusion_pytorch_model-00006-of-00014.safetensors",
440
+ "transformer_blocks.27.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00006-of-00014.safetensors",
441
  "transformer_blocks.27.norm1.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
442
  "transformer_blocks.27.norm2.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
443
  "transformer_blocks.28.adaln_proj.linear.bias": "diffusion_pytorch_model-00006-of-00014.safetensors",
 
449
  "transformer_blocks.28.attn.to_out.0.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
450
  "transformer_blocks.28.attn.to_q.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
451
  "transformer_blocks.28.attn.to_v.weight": "diffusion_pytorch_model-00006-of-00014.safetensors",
452
+ "transformer_blocks.28.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
453
+ "transformer_blocks.28.ff.net.0.proj.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
454
  "transformer_blocks.28.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
455
+ "transformer_blocks.28.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00007-of-00014.safetensors",
456
+ "transformer_blocks.28.ff.net.2.input_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
457
+ "transformer_blocks.28.ff.net.2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
458
  "transformer_blocks.28.ff.net.2.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
459
+ "transformer_blocks.28.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00007-of-00014.safetensors",
460
  "transformer_blocks.28.norm1.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
461
  "transformer_blocks.28.norm2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
462
  "transformer_blocks.29.adaln_proj.linear.bias": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
468
  "transformer_blocks.29.attn.to_out.0.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
469
  "transformer_blocks.29.attn.to_q.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
470
  "transformer_blocks.29.attn.to_v.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
471
+ "transformer_blocks.29.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
472
+ "transformer_blocks.29.ff.net.0.proj.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
473
  "transformer_blocks.29.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
474
+ "transformer_blocks.29.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00007-of-00014.safetensors",
475
+ "transformer_blocks.29.ff.net.2.input_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
476
+ "transformer_blocks.29.ff.net.2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
477
  "transformer_blocks.29.ff.net.2.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
478
+ "transformer_blocks.29.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00007-of-00014.safetensors",
479
  "transformer_blocks.29.norm1.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
480
  "transformer_blocks.29.norm2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
481
  "transformer_blocks.3.adaln_proj.linear.bias": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
487
  "transformer_blocks.3.attn.to_out.0.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
488
  "transformer_blocks.3.attn.to_q.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
489
  "transformer_blocks.3.attn.to_v.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
490
+ "transformer_blocks.3.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
491
+ "transformer_blocks.3.ff.net.0.proj.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
492
  "transformer_blocks.3.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
493
+ "transformer_blocks.3.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00007-of-00014.safetensors",
494
+ "transformer_blocks.3.ff.net.2.input_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
495
+ "transformer_blocks.3.ff.net.2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
496
  "transformer_blocks.3.ff.net.2.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
497
+ "transformer_blocks.3.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00007-of-00014.safetensors",
498
  "transformer_blocks.3.norm1.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
499
  "transformer_blocks.3.norm2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
500
  "transformer_blocks.30.adaln_proj.linear.bias": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
506
  "transformer_blocks.30.attn.to_out.0.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
507
  "transformer_blocks.30.attn.to_q.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
508
  "transformer_blocks.30.attn.to_v.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
509
+ "transformer_blocks.30.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
510
+ "transformer_blocks.30.ff.net.0.proj.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
511
  "transformer_blocks.30.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
512
+ "transformer_blocks.30.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00007-of-00014.safetensors",
513
+ "transformer_blocks.30.ff.net.2.input_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
514
+ "transformer_blocks.30.ff.net.2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
515
  "transformer_blocks.30.ff.net.2.weight_scale": "diffusion_pytorch_model-00007-of-00014.safetensors",
516
+ "transformer_blocks.30.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00007-of-00014.safetensors",
517
  "transformer_blocks.30.norm1.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
518
  "transformer_blocks.30.norm2.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
519
  "transformer_blocks.31.adaln_proj.linear.bias": "diffusion_pytorch_model-00007-of-00014.safetensors",
 
525
  "transformer_blocks.31.attn.to_out.0.weight": "diffusion_pytorch_model-00007-of-00014.safetensors",
526
  "transformer_blocks.31.attn.to_q.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
527
  "transformer_blocks.31.attn.to_v.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
528
+ "transformer_blocks.31.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
529
+ "transformer_blocks.31.ff.net.0.proj.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
530
  "transformer_blocks.31.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
531
+ "transformer_blocks.31.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00008-of-00014.safetensors",
532
+ "transformer_blocks.31.ff.net.2.input_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
533
+ "transformer_blocks.31.ff.net.2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
534
  "transformer_blocks.31.ff.net.2.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
535
+ "transformer_blocks.31.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00008-of-00014.safetensors",
536
  "transformer_blocks.31.norm1.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
537
  "transformer_blocks.31.norm2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
538
  "transformer_blocks.32.adaln_proj.linear.bias": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
544
  "transformer_blocks.32.attn.to_out.0.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
545
  "transformer_blocks.32.attn.to_q.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
546
  "transformer_blocks.32.attn.to_v.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
547
+ "transformer_blocks.32.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
548
+ "transformer_blocks.32.ff.net.0.proj.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
549
  "transformer_blocks.32.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
550
+ "transformer_blocks.32.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00008-of-00014.safetensors",
551
+ "transformer_blocks.32.ff.net.2.input_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
552
+ "transformer_blocks.32.ff.net.2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
553
  "transformer_blocks.32.ff.net.2.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
554
+ "transformer_blocks.32.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00008-of-00014.safetensors",
555
  "transformer_blocks.32.norm1.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
556
  "transformer_blocks.32.norm2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
557
  "transformer_blocks.33.adaln_proj.linear.bias": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
563
  "transformer_blocks.33.attn.to_out.0.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
564
  "transformer_blocks.33.attn.to_q.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
565
  "transformer_blocks.33.attn.to_v.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
566
+ "transformer_blocks.33.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
567
+ "transformer_blocks.33.ff.net.0.proj.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
568
  "transformer_blocks.33.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
569
+ "transformer_blocks.33.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00008-of-00014.safetensors",
570
+ "transformer_blocks.33.ff.net.2.input_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
571
+ "transformer_blocks.33.ff.net.2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
572
  "transformer_blocks.33.ff.net.2.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
573
+ "transformer_blocks.33.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00008-of-00014.safetensors",
574
  "transformer_blocks.33.norm1.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
575
  "transformer_blocks.33.norm2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
576
  "transformer_blocks.34.adaln_proj.linear.bias": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
582
  "transformer_blocks.34.attn.to_out.0.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
583
  "transformer_blocks.34.attn.to_q.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
584
  "transformer_blocks.34.attn.to_v.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
585
+ "transformer_blocks.34.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
586
+ "transformer_blocks.34.ff.net.0.proj.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
587
  "transformer_blocks.34.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
588
+ "transformer_blocks.34.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00008-of-00014.safetensors",
589
+ "transformer_blocks.34.ff.net.2.input_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
590
+ "transformer_blocks.34.ff.net.2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
591
  "transformer_blocks.34.ff.net.2.weight_scale": "diffusion_pytorch_model-00008-of-00014.safetensors",
592
+ "transformer_blocks.34.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00008-of-00014.safetensors",
593
  "transformer_blocks.34.norm1.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
594
  "transformer_blocks.34.norm2.weight": "diffusion_pytorch_model-00008-of-00014.safetensors",
595
  "transformer_blocks.35.adaln_proj.linear.bias": "diffusion_pytorch_model-00008-of-00014.safetensors",
 
601
  "transformer_blocks.35.attn.to_out.0.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
602
  "transformer_blocks.35.attn.to_q.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
603
  "transformer_blocks.35.attn.to_v.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
604
+ "transformer_blocks.35.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
605
+ "transformer_blocks.35.ff.net.0.proj.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
606
  "transformer_blocks.35.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
607
+ "transformer_blocks.35.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00009-of-00014.safetensors",
608
+ "transformer_blocks.35.ff.net.2.input_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
609
+ "transformer_blocks.35.ff.net.2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
610
  "transformer_blocks.35.ff.net.2.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
611
+ "transformer_blocks.35.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00009-of-00014.safetensors",
612
  "transformer_blocks.35.norm1.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
613
  "transformer_blocks.35.norm2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
614
  "transformer_blocks.36.adaln_proj.linear.bias": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
620
  "transformer_blocks.36.attn.to_out.0.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
621
  "transformer_blocks.36.attn.to_q.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
622
  "transformer_blocks.36.attn.to_v.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
623
+ "transformer_blocks.36.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
624
+ "transformer_blocks.36.ff.net.0.proj.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
625
  "transformer_blocks.36.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
626
+ "transformer_blocks.36.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00009-of-00014.safetensors",
627
+ "transformer_blocks.36.ff.net.2.input_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
628
+ "transformer_blocks.36.ff.net.2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
629
  "transformer_blocks.36.ff.net.2.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
630
+ "transformer_blocks.36.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00009-of-00014.safetensors",
631
  "transformer_blocks.36.norm1.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
632
  "transformer_blocks.36.norm2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
633
  "transformer_blocks.37.adaln_proj.linear.bias": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
639
  "transformer_blocks.37.attn.to_out.0.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
640
  "transformer_blocks.37.attn.to_q.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
641
  "transformer_blocks.37.attn.to_v.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
642
+ "transformer_blocks.37.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
643
+ "transformer_blocks.37.ff.net.0.proj.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
644
  "transformer_blocks.37.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
645
+ "transformer_blocks.37.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00009-of-00014.safetensors",
646
+ "transformer_blocks.37.ff.net.2.input_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
647
+ "transformer_blocks.37.ff.net.2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
648
  "transformer_blocks.37.ff.net.2.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
649
+ "transformer_blocks.37.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00009-of-00014.safetensors",
650
  "transformer_blocks.37.norm1.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
651
  "transformer_blocks.37.norm2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
652
  "transformer_blocks.38.adaln_proj.linear.bias": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
658
  "transformer_blocks.38.attn.to_out.0.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
659
  "transformer_blocks.38.attn.to_q.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
660
  "transformer_blocks.38.attn.to_v.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
661
+ "transformer_blocks.38.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
662
+ "transformer_blocks.38.ff.net.0.proj.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
663
  "transformer_blocks.38.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
664
+ "transformer_blocks.38.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00009-of-00014.safetensors",
665
+ "transformer_blocks.38.ff.net.2.input_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
666
+ "transformer_blocks.38.ff.net.2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
667
  "transformer_blocks.38.ff.net.2.weight_scale": "diffusion_pytorch_model-00009-of-00014.safetensors",
668
+ "transformer_blocks.38.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00009-of-00014.safetensors",
669
  "transformer_blocks.38.norm1.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
670
  "transformer_blocks.38.norm2.weight": "diffusion_pytorch_model-00009-of-00014.safetensors",
671
  "transformer_blocks.39.adaln_proj.linear.bias": "diffusion_pytorch_model-00009-of-00014.safetensors",
 
677
  "transformer_blocks.39.attn.to_out.0.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
678
  "transformer_blocks.39.attn.to_q.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
679
  "transformer_blocks.39.attn.to_v.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
680
+ "transformer_blocks.39.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
681
+ "transformer_blocks.39.ff.net.0.proj.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
682
  "transformer_blocks.39.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
683
+ "transformer_blocks.39.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00010-of-00014.safetensors",
684
+ "transformer_blocks.39.ff.net.2.input_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
685
+ "transformer_blocks.39.ff.net.2.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
686
  "transformer_blocks.39.ff.net.2.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
687
+ "transformer_blocks.39.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00010-of-00014.safetensors",
688
  "transformer_blocks.39.norm1.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
689
  "transformer_blocks.39.norm2.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
690
  "transformer_blocks.4.adaln_proj.linear.bias": "diffusion_pytorch_model-00010-of-00014.safetensors",
 
696
  "transformer_blocks.4.attn.to_out.0.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
697
  "transformer_blocks.4.attn.to_q.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
698
  "transformer_blocks.4.attn.to_v.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
699
+ "transformer_blocks.4.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
700
+ "transformer_blocks.4.ff.net.0.proj.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
701
  "transformer_blocks.4.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
702
+ "transformer_blocks.4.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00010-of-00014.safetensors",
703
+ "transformer_blocks.4.ff.net.2.input_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
704
+ "transformer_blocks.4.ff.net.2.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
705
  "transformer_blocks.4.ff.net.2.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
706
+ "transformer_blocks.4.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00010-of-00014.safetensors",
707
  "transformer_blocks.4.norm1.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
708
  "transformer_blocks.4.norm2.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
709
  "transformer_blocks.40.adaln_proj.linear.bias": "diffusion_pytorch_model-00010-of-00014.safetensors",
 
715
  "transformer_blocks.40.attn.to_out.0.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
716
  "transformer_blocks.40.attn.to_q.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
717
  "transformer_blocks.40.attn.to_v.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
718
+ "transformer_blocks.40.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
719
+ "transformer_blocks.40.ff.net.0.proj.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
720
  "transformer_blocks.40.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
721
+ "transformer_blocks.40.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00010-of-00014.safetensors",
722
+ "transformer_blocks.40.ff.net.2.input_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
723
+ "transformer_blocks.40.ff.net.2.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
724
  "transformer_blocks.40.ff.net.2.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
725
+ "transformer_blocks.40.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00010-of-00014.safetensors",
726
  "transformer_blocks.40.norm1.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
727
  "transformer_blocks.40.norm2.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
728
  "transformer_blocks.41.adaln_proj.linear.bias": "diffusion_pytorch_model-00010-of-00014.safetensors",
 
734
  "transformer_blocks.41.attn.to_out.0.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
735
  "transformer_blocks.41.attn.to_q.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
736
  "transformer_blocks.41.attn.to_v.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
737
+ "transformer_blocks.41.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
738
+ "transformer_blocks.41.ff.net.0.proj.weight": "diffusion_pytorch_model-00010-of-00014.safetensors",
739
  "transformer_blocks.41.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00010-of-00014.safetensors",
740
+ "transformer_blocks.41.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00010-of-00014.safetensors",
741
+ "transformer_blocks.41.ff.net.2.input_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
742
+ "transformer_blocks.41.ff.net.2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
743
  "transformer_blocks.41.ff.net.2.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
744
+ "transformer_blocks.41.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00011-of-00014.safetensors",
745
  "transformer_blocks.41.norm1.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
746
  "transformer_blocks.41.norm2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
747
  "transformer_blocks.42.adaln_proj.linear.bias": "diffusion_pytorch_model-00011-of-00014.safetensors",
 
753
  "transformer_blocks.42.attn.to_out.0.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
754
  "transformer_blocks.42.attn.to_q.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
755
  "transformer_blocks.42.attn.to_v.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
756
+ "transformer_blocks.42.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
757
+ "transformer_blocks.42.ff.net.0.proj.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
758
  "transformer_blocks.42.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
759
+ "transformer_blocks.42.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00011-of-00014.safetensors",
760
+ "transformer_blocks.42.ff.net.2.input_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
761
+ "transformer_blocks.42.ff.net.2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
762
  "transformer_blocks.42.ff.net.2.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
763
+ "transformer_blocks.42.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00011-of-00014.safetensors",
764
  "transformer_blocks.42.norm1.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
765
  "transformer_blocks.42.norm2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
766
  "transformer_blocks.43.adaln_proj.linear.bias": "diffusion_pytorch_model-00011-of-00014.safetensors",
 
772
  "transformer_blocks.43.attn.to_out.0.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
773
  "transformer_blocks.43.attn.to_q.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
774
  "transformer_blocks.43.attn.to_v.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
775
+ "transformer_blocks.43.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
776
+ "transformer_blocks.43.ff.net.0.proj.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
777
  "transformer_blocks.43.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
778
+ "transformer_blocks.43.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00011-of-00014.safetensors",
779
+ "transformer_blocks.43.ff.net.2.input_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
780
+ "transformer_blocks.43.ff.net.2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
781
  "transformer_blocks.43.ff.net.2.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
782
+ "transformer_blocks.43.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00011-of-00014.safetensors",
783
  "transformer_blocks.43.norm1.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
784
  "transformer_blocks.43.norm2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
785
  "transformer_blocks.44.adaln_proj.linear.bias": "diffusion_pytorch_model-00011-of-00014.safetensors",
 
791
  "transformer_blocks.44.attn.to_out.0.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
792
  "transformer_blocks.44.attn.to_q.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
793
  "transformer_blocks.44.attn.to_v.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
794
+ "transformer_blocks.44.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
795
+ "transformer_blocks.44.ff.net.0.proj.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
796
  "transformer_blocks.44.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
797
+ "transformer_blocks.44.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00011-of-00014.safetensors",
798
+ "transformer_blocks.44.ff.net.2.input_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
799
+ "transformer_blocks.44.ff.net.2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
800
  "transformer_blocks.44.ff.net.2.weight_scale": "diffusion_pytorch_model-00011-of-00014.safetensors",
801
+ "transformer_blocks.44.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00011-of-00014.safetensors",
802
  "transformer_blocks.44.norm1.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
803
  "transformer_blocks.44.norm2.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
804
  "transformer_blocks.45.adaln_proj.linear.bias": "diffusion_pytorch_model-00011-of-00014.safetensors",
 
810
  "transformer_blocks.45.attn.to_out.0.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
811
  "transformer_blocks.45.attn.to_q.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
812
  "transformer_blocks.45.attn.to_v.weight": "diffusion_pytorch_model-00011-of-00014.safetensors",
813
+ "transformer_blocks.45.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
814
+ "transformer_blocks.45.ff.net.0.proj.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
815
  "transformer_blocks.45.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
816
+ "transformer_blocks.45.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00012-of-00014.safetensors",
817
+ "transformer_blocks.45.ff.net.2.input_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
818
+ "transformer_blocks.45.ff.net.2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
819
  "transformer_blocks.45.ff.net.2.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
820
+ "transformer_blocks.45.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00012-of-00014.safetensors",
821
  "transformer_blocks.45.norm1.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
822
  "transformer_blocks.45.norm2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
823
  "transformer_blocks.46.adaln_proj.linear.bias": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
829
  "transformer_blocks.46.attn.to_out.0.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
830
  "transformer_blocks.46.attn.to_q.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
831
  "transformer_blocks.46.attn.to_v.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
832
+ "transformer_blocks.46.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
833
+ "transformer_blocks.46.ff.net.0.proj.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
834
  "transformer_blocks.46.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
835
+ "transformer_blocks.46.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00012-of-00014.safetensors",
836
+ "transformer_blocks.46.ff.net.2.input_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
837
+ "transformer_blocks.46.ff.net.2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
838
  "transformer_blocks.46.ff.net.2.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
839
+ "transformer_blocks.46.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00012-of-00014.safetensors",
840
  "transformer_blocks.46.norm1.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
841
  "transformer_blocks.46.norm2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
842
  "transformer_blocks.47.adaln_proj.linear.bias": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
848
  "transformer_blocks.47.attn.to_out.0.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
849
  "transformer_blocks.47.attn.to_q.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
850
  "transformer_blocks.47.attn.to_v.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
851
+ "transformer_blocks.47.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
852
+ "transformer_blocks.47.ff.net.0.proj.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
853
  "transformer_blocks.47.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
854
+ "transformer_blocks.47.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00012-of-00014.safetensors",
855
+ "transformer_blocks.47.ff.net.2.input_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
856
+ "transformer_blocks.47.ff.net.2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
857
  "transformer_blocks.47.ff.net.2.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
858
+ "transformer_blocks.47.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00012-of-00014.safetensors",
859
  "transformer_blocks.47.norm1.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
860
  "transformer_blocks.47.norm2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
861
  "transformer_blocks.48.adaln_proj.linear.bias": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
867
  "transformer_blocks.48.attn.to_out.0.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
868
  "transformer_blocks.48.attn.to_q.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
869
  "transformer_blocks.48.attn.to_v.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
870
+ "transformer_blocks.48.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
871
+ "transformer_blocks.48.ff.net.0.proj.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
872
  "transformer_blocks.48.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
873
+ "transformer_blocks.48.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00012-of-00014.safetensors",
874
+ "transformer_blocks.48.ff.net.2.input_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
875
+ "transformer_blocks.48.ff.net.2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
876
  "transformer_blocks.48.ff.net.2.weight_scale": "diffusion_pytorch_model-00012-of-00014.safetensors",
877
+ "transformer_blocks.48.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00012-of-00014.safetensors",
878
  "transformer_blocks.48.norm1.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
879
  "transformer_blocks.48.norm2.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
880
  "transformer_blocks.49.adaln_proj.linear.bias": "diffusion_pytorch_model-00012-of-00014.safetensors",
 
886
  "transformer_blocks.49.attn.to_out.0.weight": "diffusion_pytorch_model-00012-of-00014.safetensors",
887
  "transformer_blocks.49.attn.to_q.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
888
  "transformer_blocks.49.attn.to_v.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
889
+ "transformer_blocks.49.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
890
+ "transformer_blocks.49.ff.net.0.proj.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
891
  "transformer_blocks.49.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
892
+ "transformer_blocks.49.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00013-of-00014.safetensors",
893
+ "transformer_blocks.49.ff.net.2.input_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
894
+ "transformer_blocks.49.ff.net.2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
895
  "transformer_blocks.49.ff.net.2.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
896
+ "transformer_blocks.49.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00013-of-00014.safetensors",
897
  "transformer_blocks.49.norm1.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
898
  "transformer_blocks.49.norm2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
899
  "transformer_blocks.5.adaln_proj.linear.bias": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
905
  "transformer_blocks.5.attn.to_out.0.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
906
  "transformer_blocks.5.attn.to_q.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
907
  "transformer_blocks.5.attn.to_v.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
908
+ "transformer_blocks.5.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
909
+ "transformer_blocks.5.ff.net.0.proj.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
910
  "transformer_blocks.5.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
911
+ "transformer_blocks.5.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00013-of-00014.safetensors",
912
+ "transformer_blocks.5.ff.net.2.input_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
913
+ "transformer_blocks.5.ff.net.2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
914
  "transformer_blocks.5.ff.net.2.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
915
+ "transformer_blocks.5.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00013-of-00014.safetensors",
916
  "transformer_blocks.5.norm1.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
917
  "transformer_blocks.5.norm2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
918
  "transformer_blocks.6.adaln_proj.linear.bias": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
924
  "transformer_blocks.6.attn.to_out.0.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
925
  "transformer_blocks.6.attn.to_q.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
926
  "transformer_blocks.6.attn.to_v.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
927
+ "transformer_blocks.6.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
928
+ "transformer_blocks.6.ff.net.0.proj.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
929
  "transformer_blocks.6.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
930
+ "transformer_blocks.6.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00013-of-00014.safetensors",
931
+ "transformer_blocks.6.ff.net.2.input_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
932
+ "transformer_blocks.6.ff.net.2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
933
  "transformer_blocks.6.ff.net.2.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
934
+ "transformer_blocks.6.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00013-of-00014.safetensors",
935
  "transformer_blocks.6.norm1.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
936
  "transformer_blocks.6.norm2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
937
  "transformer_blocks.7.adaln_proj.linear.bias": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
943
  "transformer_blocks.7.attn.to_out.0.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
944
  "transformer_blocks.7.attn.to_q.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
945
  "transformer_blocks.7.attn.to_v.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
946
+ "transformer_blocks.7.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
947
+ "transformer_blocks.7.ff.net.0.proj.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
948
  "transformer_blocks.7.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
949
+ "transformer_blocks.7.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00013-of-00014.safetensors",
950
+ "transformer_blocks.7.ff.net.2.input_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
951
+ "transformer_blocks.7.ff.net.2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
952
  "transformer_blocks.7.ff.net.2.weight_scale": "diffusion_pytorch_model-00013-of-00014.safetensors",
953
+ "transformer_blocks.7.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00013-of-00014.safetensors",
954
  "transformer_blocks.7.norm1.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
955
  "transformer_blocks.7.norm2.weight": "diffusion_pytorch_model-00013-of-00014.safetensors",
956
  "transformer_blocks.8.adaln_proj.linear.bias": "diffusion_pytorch_model-00013-of-00014.safetensors",
 
962
  "transformer_blocks.8.attn.to_out.0.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
963
  "transformer_blocks.8.attn.to_q.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
964
  "transformer_blocks.8.attn.to_v.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
965
+ "transformer_blocks.8.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
966
+ "transformer_blocks.8.ff.net.0.proj.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
967
  "transformer_blocks.8.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
968
+ "transformer_blocks.8.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00014-of-00014.safetensors",
969
+ "transformer_blocks.8.ff.net.2.input_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
970
+ "transformer_blocks.8.ff.net.2.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
971
  "transformer_blocks.8.ff.net.2.weight_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
972
+ "transformer_blocks.8.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00014-of-00014.safetensors",
973
  "transformer_blocks.8.norm1.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
974
  "transformer_blocks.8.norm2.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
975
  "transformer_blocks.9.adaln_proj.linear.bias": "diffusion_pytorch_model-00014-of-00014.safetensors",
 
981
  "transformer_blocks.9.attn.to_out.0.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
982
  "transformer_blocks.9.attn.to_q.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
983
  "transformer_blocks.9.attn.to_v.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
984
+ "transformer_blocks.9.ff.net.0.proj.input_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
985
+ "transformer_blocks.9.ff.net.0.proj.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
986
  "transformer_blocks.9.ff.net.0.proj.weight_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
987
+ "transformer_blocks.9.ff.net.0.proj.weight_scale_2": "diffusion_pytorch_model-00014-of-00014.safetensors",
988
+ "transformer_blocks.9.ff.net.2.input_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
989
+ "transformer_blocks.9.ff.net.2.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
990
  "transformer_blocks.9.ff.net.2.weight_scale": "diffusion_pytorch_model-00014-of-00014.safetensors",
991
+ "transformer_blocks.9.ff.net.2.weight_scale_2": "diffusion_pytorch_model-00014-of-00014.safetensors",
992
  "transformer_blocks.9.norm1.weight": "diffusion_pytorch_model-00014-of-00014.safetensors",
993
  "transformer_blocks.9.norm2.weight": "diffusion_pytorch_model-00014-of-00014.safetensors"
994
  }
transformer/modelopt-component-manifest.json DELETED
@@ -1,1483 +0,0 @@
1
- {
2
- "component": "denoiser",
3
- "files": {
4
- "config.json": {
5
- "bytes": 546,
6
- "sha256": "74c11bff524336576096993cbfcdcdc2ef4fa2fa4409df693bdcbc6c666282ae"
7
- },
8
- "diffusion_pytorch_model-00001-of-00014.safetensors": {
9
- "bytes": 4666277856,
10
- "sha256": "66dd07f1629c38e19c4c18384405cef84cc9b8dd424bedd1de5f953c05991e41"
11
- },
12
- "diffusion_pytorch_model-00002-of-00014.safetensors": {
13
- "bytes": 3989259848,
14
- "sha256": "2a11227cb6c67446498fc44a92bf45683425da4e06ca09282a2707a260e90584"
15
- },
16
- "diffusion_pytorch_model-00003-of-00014.safetensors": {
17
- "bytes": 3989259864,
18
- "sha256": "4aafaa4c41e4022d58f236b990672545e829c9629a678a0dc4e220a4d29a1446"
19
- },
20
- "diffusion_pytorch_model-00004-of-00014.safetensors": {
21
- "bytes": 3546104664,
22
- "sha256": "072ba683a4dbb184ca4c0d9a561563b1a7259957c342475f3536d7c7f5712c48"
23
- },
24
- "diffusion_pytorch_model-00005-of-00014.safetensors": {
25
- "bytes": 4099832988,
26
- "sha256": "1762c23119da1924ed7a2be99eba18b05f7ee25be320c791067bea99e8f281bc"
27
- },
28
- "diffusion_pytorch_model-00006-of-00014.safetensors": {
29
- "bytes": 4056696260,
30
- "sha256": "6faddc05d3478d965faea9948f65443a3e908f2528a7de25af795615c84ee640"
31
- },
32
- "diffusion_pytorch_model-00007-of-00014.safetensors": {
33
- "bytes": 3989259832,
34
- "sha256": "d8ceaad5fdfd8bb70ffff4c4c5d0cb164f05e609fc3a428b1e171d2ad519c6a6"
35
- },
36
- "diffusion_pytorch_model-00008-of-00014.safetensors": {
37
- "bytes": 3989259864,
38
- "sha256": "f29d771bad140fffc723d3c9fa4910cf3d450058387c83f74b974ee61b0dee5e"
39
- },
40
- "diffusion_pytorch_model-00009-of-00014.safetensors": {
41
- "bytes": 3546104680,
42
- "sha256": "49da116f9b79e3a1b50b3085cbaa62ed1511f498ec1e209544a1567183456dab"
43
- },
44
- "diffusion_pytorch_model-00010-of-00014.safetensors": {
45
- "bytes": 4099832972,
46
- "sha256": "550290b051f89ca10a40209fd3d79334e91110088a522cc9544776b43aa17da8"
47
- },
48
- "diffusion_pytorch_model-00011-of-00014.safetensors": {
49
- "bytes": 4056696260,
50
- "sha256": "a3626720879942a3afb460a00b98a375911d3e1d07b52a885b5ba4843f5def55"
51
- },
52
- "diffusion_pytorch_model-00012-of-00014.safetensors": {
53
- "bytes": 3989259848,
54
- "sha256": "cd2be36a2032ee027083be015675949970732d4d579ada810a34b854fe557731"
55
- },
56
- "diffusion_pytorch_model-00013-of-00014.safetensors": {
57
- "bytes": 3989259808,
58
- "sha256": "a63129e407c505b446b434cbb8375874156ab27a6031d440e48eac1eb32db981"
59
- },
60
- "diffusion_pytorch_model-00014-of-00014.safetensors": {
61
- "bytes": 1474210592,
62
- "sha256": "aa1081318d90d28644ee9250314298639bcf323e770e5e758d527886cc246bb3"
63
- },
64
- "diffusion_pytorch_model.safetensors.index.json": {
65
- "bytes": 92939,
66
- "sha256": "af2cfa36eeae8bc371c908b658526d9599a548663e6a87eb3495a1f32377aa98"
67
- }
68
- },
69
- "modelopt_commit": "6a4b3f147e14a6fec690fedbced8df402344085d",
70
- "numerical_format": {
71
- "activation": "a4",
72
- "block_scale": "fp8_e4m3",
73
- "block_size": 16,
74
- "output": "bf16",
75
- "tensor_scale": "fp32",
76
- "values": "e2m1",
77
- "weight": "w4"
78
- },
79
- "source": "/workspace/models/FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree/transformer",
80
- "target_weights": 100,
81
- "weights": {
82
- "transformer_blocks.0.ff.net.0.proj.weight": {
83
- "block_scale_max": 448.0,
84
- "block_scale_mean": 16.472774505615234,
85
- "block_scale_min": 0.375,
86
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
87
- "shape": [
88
- 28672,
89
- 5376
90
- ],
91
- "source_dtype": "bfloat16",
92
- "weight_global_amax": 8.375,
93
- "weight_saturation_rate": 0.06355322897434235,
94
- "weight_tensor_scale": 0.0031156993936747313
95
- },
96
- "transformer_blocks.0.ff.net.2.weight": {
97
- "block_scale_max": 448.0,
98
- "block_scale_mean": 33.925846099853516,
99
- "block_scale_min": 5.0,
100
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
101
- "shape": [
102
- 5376,
103
- 14336
104
- ],
105
- "source_dtype": "bfloat16",
106
- "weight_global_amax": 4.8125,
107
- "weight_saturation_rate": 0.06495451927185059,
108
- "weight_tensor_scale": 0.0017903646221384406
109
- },
110
- "transformer_blocks.1.ff.net.0.proj.weight": {
111
- "block_scale_max": 448.0,
112
- "block_scale_mean": 28.016803741455078,
113
- "block_scale_min": 0.6875,
114
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
115
- "shape": [
116
- 28672,
117
- 5376
118
- ],
119
- "source_dtype": "bfloat16",
120
- "weight_global_amax": 4.40625,
121
- "weight_saturation_rate": 0.06361088156700134,
122
- "weight_tensor_scale": 0.0016392299439758062
123
- },
124
- "transformer_blocks.1.ff.net.2.weight": {
125
- "block_scale_max": 448.0,
126
- "block_scale_mean": 17.470399856567383,
127
- "block_scale_min": 2.25,
128
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
129
- "shape": [
130
- 5376,
131
- 14336
132
- ],
133
- "source_dtype": "bfloat16",
134
- "weight_global_amax": 8.4375,
135
- "weight_saturation_rate": 0.06421976536512375,
136
- "weight_tensor_scale": 0.003138951025903225
137
- },
138
- "transformer_blocks.10.ff.net.0.proj.weight": {
139
- "block_scale_max": 448.0,
140
- "block_scale_mean": 44.30687713623047,
141
- "block_scale_min": 1.375,
142
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
143
- "shape": [
144
- 28672,
145
- 5376
146
- ],
147
- "source_dtype": "bfloat16",
148
- "weight_global_amax": 3.609375,
149
- "weight_saturation_rate": 0.06342260539531708,
150
- "weight_tensor_scale": 0.0013427734375
151
- },
152
- "transformer_blocks.10.ff.net.2.weight": {
153
- "block_scale_max": 448.0,
154
- "block_scale_mean": 70.12530517578125,
155
- "block_scale_min": 8.0,
156
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
157
- "shape": [
158
- 5376,
159
- 14336
160
- ],
161
- "source_dtype": "bfloat16",
162
- "weight_global_amax": 2.40625,
163
- "weight_saturation_rate": 0.06392619013786316,
164
- "weight_tensor_scale": 0.0008951823110692203
165
- },
166
- "transformer_blocks.11.ff.net.0.proj.weight": {
167
- "block_scale_max": 448.0,
168
- "block_scale_mean": 26.485719680786133,
169
- "block_scale_min": 0.75,
170
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
171
- "shape": [
172
- 28672,
173
- 5376
174
- ],
175
- "source_dtype": "bfloat16",
176
- "weight_global_amax": 5.9375,
177
- "weight_saturation_rate": 0.0634349137544632,
178
- "weight_tensor_scale": 0.002208891324698925
179
- },
180
- "transformer_blocks.11.ff.net.2.weight": {
181
- "block_scale_max": 448.0,
182
- "block_scale_mean": 36.138858795166016,
183
- "block_scale_min": 3.0,
184
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
185
- "shape": [
186
- 5376,
187
- 14336
188
- ],
189
- "source_dtype": "bfloat16",
190
- "weight_global_amax": 4.5625,
191
- "weight_saturation_rate": 0.06391075998544693,
192
- "weight_tensor_scale": 0.001697358675301075
193
- },
194
- "transformer_blocks.12.ff.net.0.proj.weight": {
195
- "block_scale_max": 448.0,
196
- "block_scale_mean": 38.985008239746094,
197
- "block_scale_min": 1.125,
198
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
199
- "shape": [
200
- 28672,
201
- 5376
202
- ],
203
- "source_dtype": "bfloat16",
204
- "weight_global_amax": 3.953125,
205
- "weight_saturation_rate": 0.06354819983243942,
206
- "weight_tensor_scale": 0.0014706565998494625
207
- },
208
- "transformer_blocks.12.ff.net.2.weight": {
209
- "block_scale_max": 448.0,
210
- "block_scale_mean": 41.339073181152344,
211
- "block_scale_min": 3.25,
212
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
213
- "shape": [
214
- 5376,
215
- 14336
216
- ],
217
- "source_dtype": "bfloat16",
218
- "weight_global_amax": 3.921875,
219
- "weight_saturation_rate": 0.06375246495008469,
220
- "weight_tensor_scale": 0.0014590309001505375
221
- },
222
- "transformer_blocks.13.ff.net.0.proj.weight": {
223
- "block_scale_max": 448.0,
224
- "block_scale_mean": 12.657382011413574,
225
- "block_scale_min": 0.375,
226
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
227
- "shape": [
228
- 28672,
229
- 5376
230
- ],
231
- "source_dtype": "bfloat16",
232
- "weight_global_amax": 12.125,
233
- "weight_saturation_rate": 0.06352914869785309,
234
- "weight_tensor_scale": 0.0045107887126505375
235
- },
236
- "transformer_blocks.13.ff.net.2.weight": {
237
- "block_scale_max": 448.0,
238
- "block_scale_mean": 10.395652770996094,
239
- "block_scale_min": 0.625,
240
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
241
- "shape": [
242
- 5376,
243
- 14336
244
- ],
245
- "source_dtype": "bfloat16",
246
- "weight_global_amax": 15.375,
247
- "weight_saturation_rate": 0.06361445039510727,
248
- "weight_tensor_scale": 0.0057198661379516125
249
- },
250
- "transformer_blocks.14.ff.net.0.proj.weight": {
251
- "block_scale_max": 448.0,
252
- "block_scale_mean": 32.547725677490234,
253
- "block_scale_min": 0.9375,
254
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
255
- "shape": [
256
- 28672,
257
- 5376
258
- ],
259
- "source_dtype": "bfloat16",
260
- "weight_global_amax": 4.75,
261
- "weight_saturation_rate": 0.06362791359424591,
262
- "weight_tensor_scale": 0.0017671131063252687
263
- },
264
- "transformer_blocks.14.ff.net.2.weight": {
265
- "block_scale_max": 448.0,
266
- "block_scale_mean": 40.63711929321289,
267
- "block_scale_min": 2.5,
268
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
269
- "shape": [
270
- 5376,
271
- 14336
272
- ],
273
- "source_dtype": "bfloat16",
274
- "weight_global_amax": 3.921875,
275
- "weight_saturation_rate": 0.06358253955841064,
276
- "weight_tensor_scale": 0.0014590309001505375
277
- },
278
- "transformer_blocks.15.ff.net.0.proj.weight": {
279
- "block_scale_max": 448.0,
280
- "block_scale_mean": 26.8400821685791,
281
- "block_scale_min": 0.8125,
282
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
283
- "shape": [
284
- 28672,
285
- 5376
286
- ],
287
- "source_dtype": "bfloat16",
288
- "weight_global_amax": 5.71875,
289
- "weight_saturation_rate": 0.06362185627222061,
290
- "weight_tensor_scale": 0.0021275111939758062
291
- },
292
- "transformer_blocks.15.ff.net.2.weight": {
293
- "block_scale_max": 448.0,
294
- "block_scale_mean": 30.485517501831055,
295
- "block_scale_min": 1.375,
296
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
297
- "shape": [
298
- 5376,
299
- 14336
300
- ],
301
- "source_dtype": "bfloat16",
302
- "weight_global_amax": 5.0625,
303
- "weight_saturation_rate": 0.06376423686742783,
304
- "weight_tensor_scale": 0.0018833705689758062
305
- },
306
- "transformer_blocks.16.ff.net.0.proj.weight": {
307
- "block_scale_max": 448.0,
308
- "block_scale_mean": 29.4029541015625,
309
- "block_scale_min": 0.9375,
310
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
311
- "shape": [
312
- 28672,
313
- 5376
314
- ],
315
- "source_dtype": "bfloat16",
316
- "weight_global_amax": 5.21875,
317
- "weight_saturation_rate": 0.0635925754904747,
318
- "weight_tensor_scale": 0.001941499300301075
319
- },
320
- "transformer_blocks.16.ff.net.2.weight": {
321
- "block_scale_max": 448.0,
322
- "block_scale_mean": 41.52595901489258,
323
- "block_scale_min": 2.5,
324
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
325
- "shape": [
326
- 5376,
327
- 14336
328
- ],
329
- "source_dtype": "bfloat16",
330
- "weight_global_amax": 3.671875,
331
- "weight_saturation_rate": 0.06345658004283905,
332
- "weight_tensor_scale": 0.0013660249533131719
333
- },
334
- "transformer_blocks.17.ff.net.0.proj.weight": {
335
- "block_scale_max": 448.0,
336
- "block_scale_mean": 35.89394760131836,
337
- "block_scale_min": 0.9375,
338
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
339
- "shape": [
340
- 28672,
341
- 5376
342
- ],
343
- "source_dtype": "bfloat16",
344
- "weight_global_amax": 4.34375,
345
- "weight_saturation_rate": 0.063621386885643,
346
- "weight_tensor_scale": 0.0016159784281626344
347
- },
348
- "transformer_blocks.17.ff.net.2.weight": {
349
- "block_scale_max": 448.0,
350
- "block_scale_mean": 34.49262237548828,
351
- "block_scale_min": 1.875,
352
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
353
- "shape": [
354
- 5376,
355
- 14336
356
- ],
357
- "source_dtype": "bfloat16",
358
- "weight_global_amax": 4.46875,
359
- "weight_saturation_rate": 0.06368372589349747,
360
- "weight_tensor_scale": 0.001662481459788978
361
- },
362
- "transformer_blocks.18.ff.net.0.proj.weight": {
363
- "block_scale_max": 448.0,
364
- "block_scale_mean": 37.41992950439453,
365
- "block_scale_min": 1.0,
366
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
367
- "shape": [
368
- 28672,
369
- 5376
370
- ],
371
- "source_dtype": "bfloat16",
372
- "weight_global_amax": 4.28125,
373
- "weight_saturation_rate": 0.0636681541800499,
374
- "weight_tensor_scale": 0.0015927269123494625
375
- },
376
- "transformer_blocks.18.ff.net.2.weight": {
377
- "block_scale_max": 448.0,
378
- "block_scale_mean": 17.0745906829834,
379
- "block_scale_min": 1.125,
380
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
381
- "shape": [
382
- 5376,
383
- 14336
384
- ],
385
- "source_dtype": "bfloat16",
386
- "weight_global_amax": 9.3125,
387
- "weight_saturation_rate": 0.06357219815254211,
388
- "weight_tensor_scale": 0.0034644717816263437
389
- },
390
- "transformer_blocks.19.ff.net.0.proj.weight": {
391
- "block_scale_max": 448.0,
392
- "block_scale_mean": 53.710105895996094,
393
- "block_scale_min": 1.625,
394
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
395
- "shape": [
396
- 28672,
397
- 5376
398
- ],
399
- "source_dtype": "bfloat16",
400
- "weight_global_amax": 3.03125,
401
- "weight_saturation_rate": 0.06346426159143448,
402
- "weight_tensor_scale": 0.0011276971781626344
403
- },
404
- "transformer_blocks.19.ff.net.2.weight": {
405
- "block_scale_max": 448.0,
406
- "block_scale_mean": 44.2147102355957,
407
- "block_scale_min": 3.25,
408
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
409
- "shape": [
410
- 5376,
411
- 14336
412
- ],
413
- "source_dtype": "bfloat16",
414
- "weight_global_amax": 3.640625,
415
- "weight_saturation_rate": 0.06315502524375916,
416
- "weight_tensor_scale": 0.0013543992536142468
417
- },
418
- "transformer_blocks.2.ff.net.0.proj.weight": {
419
- "block_scale_max": 448.0,
420
- "block_scale_mean": 33.50981140136719,
421
- "block_scale_min": 0.875,
422
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
423
- "shape": [
424
- 28672,
425
- 5376
426
- ],
427
- "source_dtype": "bfloat16",
428
- "weight_global_amax": 4.34375,
429
- "weight_saturation_rate": 0.06351428478956223,
430
- "weight_tensor_scale": 0.0016159784281626344
431
- },
432
- "transformer_blocks.2.ff.net.2.weight": {
433
- "block_scale_max": 448.0,
434
- "block_scale_mean": 23.832834243774414,
435
- "block_scale_min": 4.0,
436
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
437
- "shape": [
438
- 5376,
439
- 14336
440
- ],
441
- "source_dtype": "bfloat16",
442
- "weight_global_amax": 7.21875,
443
- "weight_saturation_rate": 0.06375253200531006,
444
- "weight_tensor_scale": 0.002685546875
445
- },
446
- "transformer_blocks.20.ff.net.0.proj.weight": {
447
- "block_scale_max": 448.0,
448
- "block_scale_mean": 36.75452423095703,
449
- "block_scale_min": 1.5,
450
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
451
- "shape": [
452
- 28672,
453
- 5376
454
- ],
455
- "source_dtype": "bfloat16",
456
- "weight_global_amax": 4.46875,
457
- "weight_saturation_rate": 0.0635710060596466,
458
- "weight_tensor_scale": 0.001662481459788978
459
- },
460
- "transformer_blocks.20.ff.net.2.weight": {
461
- "block_scale_max": 448.0,
462
- "block_scale_mean": 40.204925537109375,
463
- "block_scale_min": 4.5,
464
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
465
- "shape": [
466
- 5376,
467
- 14336
468
- ],
469
- "source_dtype": "bfloat16",
470
- "weight_global_amax": 4.03125,
471
- "weight_saturation_rate": 0.06329218298196793,
472
- "weight_tensor_scale": 0.0014997209655120969
473
- },
474
- "transformer_blocks.21.ff.net.0.proj.weight": {
475
- "block_scale_max": 448.0,
476
- "block_scale_mean": 28.32771873474121,
477
- "block_scale_min": 0.875,
478
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
479
- "shape": [
480
- 28672,
481
- 5376
482
- ],
483
- "source_dtype": "bfloat16",
484
- "weight_global_amax": 5.90625,
485
- "weight_saturation_rate": 0.06339206546545029,
486
- "weight_tensor_scale": 0.002197265625
487
- },
488
- "transformer_blocks.21.ff.net.2.weight": {
489
- "block_scale_max": 448.0,
490
- "block_scale_mean": 34.726619720458984,
491
- "block_scale_min": 3.5,
492
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
493
- "shape": [
494
- 5376,
495
- 14336
496
- ],
497
- "source_dtype": "bfloat16",
498
- "weight_global_amax": 4.6875,
499
- "weight_saturation_rate": 0.06354400515556335,
500
- "weight_tensor_scale": 0.0017438615905120969
501
- },
502
- "transformer_blocks.22.ff.net.0.proj.weight": {
503
- "block_scale_max": 448.0,
504
- "block_scale_mean": 46.91810607910156,
505
- "block_scale_min": 1.375,
506
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
507
- "shape": [
508
- 28672,
509
- 5376
510
- ],
511
- "source_dtype": "bfloat16",
512
- "weight_global_amax": 3.640625,
513
- "weight_saturation_rate": 0.06337128579616547,
514
- "weight_tensor_scale": 0.0013543992536142468
515
- },
516
- "transformer_blocks.22.ff.net.2.weight": {
517
- "block_scale_max": 448.0,
518
- "block_scale_mean": 54.79726028442383,
519
- "block_scale_min": 5.5,
520
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
521
- "shape": [
522
- 5376,
523
- 14336
524
- ],
525
- "source_dtype": "bfloat16",
526
- "weight_global_amax": 3.015625,
527
- "weight_saturation_rate": 0.06334979087114334,
528
- "weight_tensor_scale": 0.0011218843283131719
529
- },
530
- "transformer_blocks.23.ff.net.0.proj.weight": {
531
- "block_scale_max": 448.0,
532
- "block_scale_mean": 32.758995056152344,
533
- "block_scale_min": 2.25,
534
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
535
- "shape": [
536
- 28672,
537
- 5376
538
- ],
539
- "source_dtype": "bfloat16",
540
- "weight_global_amax": 5.3125,
541
- "weight_saturation_rate": 0.06352967768907547,
542
- "weight_tensor_scale": 0.0019763766322284937
543
- },
544
- "transformer_blocks.23.ff.net.2.weight": {
545
- "block_scale_max": 448.0,
546
- "block_scale_mean": 61.83660125732422,
547
- "block_scale_min": 7.0,
548
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
549
- "shape": [
550
- 5376,
551
- 14336
552
- ],
553
- "source_dtype": "bfloat16",
554
- "weight_global_amax": 2.71875,
555
- "weight_saturation_rate": 0.06354644894599915,
556
- "weight_tensor_scale": 0.0010114397155120969
557
- },
558
- "transformer_blocks.24.ff.net.0.proj.weight": {
559
- "block_scale_max": 448.0,
560
- "block_scale_mean": 27.76451873779297,
561
- "block_scale_min": 2.5,
562
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
563
- "shape": [
564
- 28672,
565
- 5376
566
- ],
567
- "source_dtype": "bfloat16",
568
- "weight_global_amax": 6.375,
569
- "weight_saturation_rate": 0.06331139802932739,
570
- "weight_tensor_scale": 0.0023716518189758062
571
- },
572
- "transformer_blocks.24.ff.net.2.weight": {
573
- "block_scale_max": 448.0,
574
- "block_scale_mean": 55.93115997314453,
575
- "block_scale_min": 5.5,
576
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
577
- "shape": [
578
- 5376,
579
- 14336
580
- ],
581
- "source_dtype": "bfloat16",
582
- "weight_global_amax": 3.03125,
583
- "weight_saturation_rate": 0.06340566277503967,
584
- "weight_tensor_scale": 0.0011276971781626344
585
- },
586
- "transformer_blocks.25.ff.net.0.proj.weight": {
587
- "block_scale_max": 448.0,
588
- "block_scale_mean": 39.748226165771484,
589
- "block_scale_min": 4.5,
590
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
591
- "shape": [
592
- 28672,
593
- 5376
594
- ],
595
- "source_dtype": "bfloat16",
596
- "weight_global_amax": 4.53125,
597
- "weight_saturation_rate": 0.06359761208295822,
598
- "weight_tensor_scale": 0.00168573297560215
599
- },
600
- "transformer_blocks.25.ff.net.2.weight": {
601
- "block_scale_max": 448.0,
602
- "block_scale_mean": 53.710121154785156,
603
- "block_scale_min": 5.0,
604
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
605
- "shape": [
606
- 5376,
607
- 14336
608
- ],
609
- "source_dtype": "bfloat16",
610
- "weight_global_amax": 3.21875,
611
- "weight_saturation_rate": 0.0632086992263794,
612
- "weight_tensor_scale": 0.0011974516091868281
613
- },
614
- "transformer_blocks.26.ff.net.0.proj.weight": {
615
- "block_scale_max": 448.0,
616
- "block_scale_mean": 41.97421646118164,
617
- "block_scale_min": 1.125,
618
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
619
- "shape": [
620
- 28672,
621
- 5376
622
- ],
623
- "source_dtype": "bfloat16",
624
- "weight_global_amax": 4.375,
625
- "weight_saturation_rate": 0.06354834884405136,
626
- "weight_tensor_scale": 0.0016276042442768812
627
- },
628
- "transformer_blocks.26.ff.net.2.weight": {
629
- "block_scale_max": 448.0,
630
- "block_scale_mean": 31.09628677368164,
631
- "block_scale_min": 3.75,
632
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
633
- "shape": [
634
- 5376,
635
- 14336
636
- ],
637
- "source_dtype": "bfloat16",
638
- "weight_global_amax": 5.6875,
639
- "weight_saturation_rate": 0.06352340430021286,
640
- "weight_tensor_scale": 0.0021158854942768812
641
- },
642
- "transformer_blocks.27.ff.net.0.proj.weight": {
643
- "block_scale_max": 448.0,
644
- "block_scale_mean": 38.58533477783203,
645
- "block_scale_min": 1.75,
646
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
647
- "shape": [
648
- 28672,
649
- 5376
650
- ],
651
- "source_dtype": "bfloat16",
652
- "weight_global_amax": 4.8125,
653
- "weight_saturation_rate": 0.06368701159954071,
654
- "weight_tensor_scale": 0.0017903646221384406
655
- },
656
- "transformer_blocks.27.ff.net.2.weight": {
657
- "block_scale_max": 448.0,
658
- "block_scale_mean": 17.185806274414062,
659
- "block_scale_min": 2.25,
660
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
661
- "shape": [
662
- 5376,
663
- 14336
664
- ],
665
- "source_dtype": "bfloat16",
666
- "weight_global_amax": 10.5,
667
- "weight_saturation_rate": 0.06366410106420517,
668
- "weight_tensor_scale": 0.00390625
669
- },
670
- "transformer_blocks.28.ff.net.0.proj.weight": {
671
- "block_scale_max": 448.0,
672
- "block_scale_mean": 26.904735565185547,
673
- "block_scale_min": 1.5,
674
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
675
- "shape": [
676
- 28672,
677
- 5376
678
- ],
679
- "source_dtype": "bfloat16",
680
- "weight_global_amax": 7.09375,
681
- "weight_saturation_rate": 0.06313972920179367,
682
- "weight_tensor_scale": 0.0026390438433736563
683
- },
684
- "transformer_blocks.28.ff.net.2.weight": {
685
- "block_scale_max": 448.0,
686
- "block_scale_mean": 25.52775764465332,
687
- "block_scale_min": 3.5,
688
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
689
- "shape": [
690
- 5376,
691
- 14336
692
- ],
693
- "source_dtype": "bfloat16",
694
- "weight_global_amax": 7.3125,
695
- "weight_saturation_rate": 0.06288973242044449,
696
- "weight_tensor_scale": 0.0027204242069274187
697
- },
698
- "transformer_blocks.29.ff.net.0.proj.weight": {
699
- "block_scale_max": 448.0,
700
- "block_scale_mean": 27.099672317504883,
701
- "block_scale_min": 4.5,
702
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
703
- "shape": [
704
- 28672,
705
- 5376
706
- ],
707
- "source_dtype": "bfloat16",
708
- "weight_global_amax": 7.0625,
709
- "weight_saturation_rate": 0.06317251175642014,
710
- "weight_tensor_scale": 0.0026274181436747313
711
- },
712
- "transformer_blocks.29.ff.net.2.weight": {
713
- "block_scale_max": 448.0,
714
- "block_scale_mean": 39.744483947753906,
715
- "block_scale_min": 6.0,
716
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
717
- "shape": [
718
- 5376,
719
- 14336
720
- ],
721
- "source_dtype": "bfloat16",
722
- "weight_global_amax": 4.78125,
723
- "weight_saturation_rate": 0.06351472437381744,
724
- "weight_tensor_scale": 0.0017787389224395156
725
- },
726
- "transformer_blocks.3.ff.net.0.proj.weight": {
727
- "block_scale_max": 448.0,
728
- "block_scale_mean": 37.328556060791016,
729
- "block_scale_min": 1.125,
730
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
731
- "shape": [
732
- 28672,
733
- 5376
734
- ],
735
- "source_dtype": "bfloat16",
736
- "weight_global_amax": 4.3125,
737
- "weight_saturation_rate": 0.0635688379406929,
738
- "weight_tensor_scale": 0.0016043527284637094
739
- },
740
- "transformer_blocks.3.ff.net.2.weight": {
741
- "block_scale_max": 448.0,
742
- "block_scale_mean": 61.82960510253906,
743
- "block_scale_min": 11.0,
744
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
745
- "shape": [
746
- 5376,
747
- 14336
748
- ],
749
- "source_dtype": "bfloat16",
750
- "weight_global_amax": 2.859375,
751
- "weight_saturation_rate": 0.06367281079292297,
752
- "weight_tensor_scale": 0.0010637555969879031
753
- },
754
- "transformer_blocks.30.ff.net.0.proj.weight": {
755
- "block_scale_max": 448.0,
756
- "block_scale_mean": 30.473941802978516,
757
- "block_scale_min": 5.0,
758
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
759
- "shape": [
760
- 28672,
761
- 5376
762
- ],
763
- "source_dtype": "bfloat16",
764
- "weight_global_amax": 6.3125,
765
- "weight_saturation_rate": 0.06351669132709503,
766
- "weight_tensor_scale": 0.002348400419577956
767
- },
768
- "transformer_blocks.30.ff.net.2.weight": {
769
- "block_scale_max": 448.0,
770
- "block_scale_mean": 15.345988273620605,
771
- "block_scale_min": 2.25,
772
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
773
- "shape": [
774
- 5376,
775
- 14336
776
- ],
777
- "source_dtype": "bfloat16",
778
- "weight_global_amax": 12.3125,
779
- "weight_saturation_rate": 0.06355677545070648,
780
- "weight_tensor_scale": 0.004580543376505375
781
- },
782
- "transformer_blocks.31.ff.net.0.proj.weight": {
783
- "block_scale_max": 448.0,
784
- "block_scale_mean": 27.60163688659668,
785
- "block_scale_min": 4.5,
786
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
787
- "shape": [
788
- 28672,
789
- 5376
790
- ],
791
- "source_dtype": "bfloat16",
792
- "weight_global_amax": 7.0,
793
- "weight_saturation_rate": 0.06324271112680435,
794
- "weight_tensor_scale": 0.0026041667442768812
795
- },
796
- "transformer_blocks.31.ff.net.2.weight": {
797
- "block_scale_max": 448.0,
798
- "block_scale_mean": 25.32241439819336,
799
- "block_scale_min": 3.5,
800
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
801
- "shape": [
802
- 5376,
803
- 14336
804
- ],
805
- "source_dtype": "bfloat16",
806
- "weight_global_amax": 7.5,
807
- "weight_saturation_rate": 0.06294678896665573,
808
- "weight_tensor_scale": 0.0027901786379516125
809
- },
810
- "transformer_blocks.32.ff.net.0.proj.weight": {
811
- "block_scale_max": 448.0,
812
- "block_scale_mean": 15.447080612182617,
813
- "block_scale_min": 2.75,
814
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
815
- "shape": [
816
- 28672,
817
- 5376
818
- ],
819
- "source_dtype": "bfloat16",
820
- "weight_global_amax": 12.5,
821
- "weight_saturation_rate": 0.06354077160358429,
822
- "weight_tensor_scale": 0.004650297574698925
823
- },
824
- "transformer_blocks.32.ff.net.2.weight": {
825
- "block_scale_max": 448.0,
826
- "block_scale_mean": 24.05265998840332,
827
- "block_scale_min": 4.0,
828
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
829
- "shape": [
830
- 5376,
831
- 14336
832
- ],
833
- "source_dtype": "bfloat16",
834
- "weight_global_amax": 8.0,
835
- "weight_saturation_rate": 0.06287482380867004,
836
- "weight_tensor_scale": 0.0029761905316263437
837
- },
838
- "transformer_blocks.33.ff.net.0.proj.weight": {
839
- "block_scale_max": 448.0,
840
- "block_scale_mean": 23.43458366394043,
841
- "block_scale_min": 4.5,
842
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
843
- "shape": [
844
- 28672,
845
- 5376
846
- ],
847
- "source_dtype": "bfloat16",
848
- "weight_global_amax": 8.3125,
849
- "weight_saturation_rate": 0.06324451416730881,
850
- "weight_tensor_scale": 0.0030924479942768812
851
- },
852
- "transformer_blocks.33.ff.net.2.weight": {
853
- "block_scale_max": 448.0,
854
- "block_scale_mean": 17.67997932434082,
855
- "block_scale_min": 3.25,
856
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
857
- "shape": [
858
- 5376,
859
- 14336
860
- ],
861
- "source_dtype": "bfloat16",
862
- "weight_global_amax": 10.875,
863
- "weight_saturation_rate": 0.06367780268192291,
864
- "weight_tensor_scale": 0.0040457588620483875
865
- },
866
- "transformer_blocks.34.ff.net.0.proj.weight": {
867
- "block_scale_max": 448.0,
868
- "block_scale_mean": 23.52635383605957,
869
- "block_scale_min": 3.75,
870
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
871
- "shape": [
872
- 28672,
873
- 5376
874
- ],
875
- "source_dtype": "bfloat16",
876
- "weight_global_amax": 8.3125,
877
- "weight_saturation_rate": 0.06323336064815521,
878
- "weight_tensor_scale": 0.0030924479942768812
879
- },
880
- "transformer_blocks.34.ff.net.2.weight": {
881
- "block_scale_max": 448.0,
882
- "block_scale_mean": 30.291606903076172,
883
- "block_scale_min": 5.5,
884
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
885
- "shape": [
886
- 5376,
887
- 14336
888
- ],
889
- "source_dtype": "bfloat16",
890
- "weight_global_amax": 6.375,
891
- "weight_saturation_rate": 0.06337101757526398,
892
- "weight_tensor_scale": 0.0023716518189758062
893
- },
894
- "transformer_blocks.35.ff.net.0.proj.weight": {
895
- "block_scale_max": 448.0,
896
- "block_scale_mean": 25.029054641723633,
897
- "block_scale_min": 4.0,
898
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
899
- "shape": [
900
- 28672,
901
- 5376
902
- ],
903
- "source_dtype": "bfloat16",
904
- "weight_global_amax": 7.875,
905
- "weight_saturation_rate": 0.06321961432695389,
906
- "weight_tensor_scale": 0.0029296875
907
- },
908
- "transformer_blocks.35.ff.net.2.weight": {
909
- "block_scale_max": 448.0,
910
- "block_scale_mean": 29.031099319458008,
911
- "block_scale_min": 5.0,
912
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
913
- "shape": [
914
- 5376,
915
- 14336
916
- ],
917
- "source_dtype": "bfloat16",
918
- "weight_global_amax": 6.75,
919
- "weight_saturation_rate": 0.06319305300712585,
920
- "weight_tensor_scale": 0.0025111606810241938
921
- },
922
- "transformer_blocks.36.ff.net.0.proj.weight": {
923
- "block_scale_max": 448.0,
924
- "block_scale_mean": 16.119068145751953,
925
- "block_scale_min": 2.25,
926
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
927
- "shape": [
928
- 28672,
929
- 5376
930
- ],
931
- "source_dtype": "bfloat16",
932
- "weight_global_amax": 12.1875,
933
- "weight_saturation_rate": 0.06348860263824463,
934
- "weight_tensor_scale": 0.0045340401120483875
935
- },
936
- "transformer_blocks.36.ff.net.2.weight": {
937
- "block_scale_max": 448.0,
938
- "block_scale_mean": 13.287528991699219,
939
- "block_scale_min": 2.25,
940
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
941
- "shape": [
942
- 5376,
943
- 14336
944
- ],
945
- "source_dtype": "bfloat16",
946
- "weight_global_amax": 15.0,
947
- "weight_saturation_rate": 0.06300818175077438,
948
- "weight_tensor_scale": 0.005580357275903225
949
- },
950
- "transformer_blocks.37.ff.net.0.proj.weight": {
951
- "block_scale_max": 448.0,
952
- "block_scale_mean": 22.18779945373535,
953
- "block_scale_min": 2.75,
954
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
955
- "shape": [
956
- 28672,
957
- 5376
958
- ],
959
- "source_dtype": "bfloat16",
960
- "weight_global_amax": 8.875,
961
- "weight_saturation_rate": 0.06334257870912552,
962
- "weight_tensor_scale": 0.0033017112873494625
963
- },
964
- "transformer_blocks.37.ff.net.2.weight": {
965
- "block_scale_max": 448.0,
966
- "block_scale_mean": 20.6065673828125,
967
- "block_scale_min": 3.5,
968
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
969
- "shape": [
970
- 5376,
971
- 14336
972
- ],
973
- "source_dtype": "bfloat16",
974
- "weight_global_amax": 9.875,
975
- "weight_saturation_rate": 0.06330486387014389,
976
- "weight_tensor_scale": 0.003673735074698925
977
- },
978
- "transformer_blocks.38.ff.net.0.proj.weight": {
979
- "block_scale_max": 448.0,
980
- "block_scale_mean": 25.0240478515625,
981
- "block_scale_min": 3.5,
982
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
983
- "shape": [
984
- 28672,
985
- 5376
986
- ],
987
- "source_dtype": "bfloat16",
988
- "weight_global_amax": 7.84375,
989
- "weight_saturation_rate": 0.06324655562639236,
990
- "weight_tensor_scale": 0.002918061800301075
991
- },
992
- "transformer_blocks.38.ff.net.2.weight": {
993
- "block_scale_max": 448.0,
994
- "block_scale_mean": 20.838619232177734,
995
- "block_scale_min": 3.5,
996
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
997
- "shape": [
998
- 5376,
999
- 14336
1000
- ],
1001
- "source_dtype": "bfloat16",
1002
- "weight_global_amax": 9.6875,
1003
- "weight_saturation_rate": 0.06329601258039474,
1004
- "weight_tensor_scale": 0.0036039806436747313
1005
- },
1006
- "transformer_blocks.39.ff.net.0.proj.weight": {
1007
- "block_scale_max": 448.0,
1008
- "block_scale_mean": 16.79216194152832,
1009
- "block_scale_min": 2.5,
1010
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1011
- "shape": [
1012
- 28672,
1013
- 5376
1014
- ],
1015
- "source_dtype": "bfloat16",
1016
- "weight_global_amax": 11.75,
1017
- "weight_saturation_rate": 0.06354103237390518,
1018
- "weight_tensor_scale": 0.00437127985060215
1019
- },
1020
- "transformer_blocks.39.ff.net.2.weight": {
1021
- "block_scale_max": 448.0,
1022
- "block_scale_mean": 12.181177139282227,
1023
- "block_scale_min": 1.875,
1024
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1025
- "shape": [
1026
- 5376,
1027
- 14336
1028
- ],
1029
- "source_dtype": "bfloat16",
1030
- "weight_global_amax": 16.75,
1031
- "weight_saturation_rate": 0.06287278980016708,
1032
- "weight_tensor_scale": 0.0062313987873494625
1033
- },
1034
- "transformer_blocks.4.ff.net.0.proj.weight": {
1035
- "block_scale_max": 448.0,
1036
- "block_scale_mean": 52.58255386352539,
1037
- "block_scale_min": 1.5,
1038
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1039
- "shape": [
1040
- 28672,
1041
- 5376
1042
- ],
1043
- "source_dtype": "bfloat16",
1044
- "weight_global_amax": 3.140625,
1045
- "weight_saturation_rate": 0.06332672387361526,
1046
- "weight_tensor_scale": 0.0011683872435241938
1047
- },
1048
- "transformer_blocks.4.ff.net.2.weight": {
1049
- "block_scale_max": 448.0,
1050
- "block_scale_mean": 66.93839263916016,
1051
- "block_scale_min": 10.0,
1052
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1053
- "shape": [
1054
- 5376,
1055
- 14336
1056
- ],
1057
- "source_dtype": "bfloat16",
1058
- "weight_global_amax": 2.640625,
1059
- "weight_saturation_rate": 0.06381126493215561,
1060
- "weight_tensor_scale": 0.0009823753498494625
1061
- },
1062
- "transformer_blocks.40.ff.net.0.proj.weight": {
1063
- "block_scale_max": 448.0,
1064
- "block_scale_mean": 18.87799644470215,
1065
- "block_scale_min": 2.75,
1066
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1067
- "shape": [
1068
- 28672,
1069
- 5376
1070
- ],
1071
- "source_dtype": "bfloat16",
1072
- "weight_global_amax": 10.4375,
1073
- "weight_saturation_rate": 0.06349590420722961,
1074
- "weight_tensor_scale": 0.00388299860060215
1075
- },
1076
- "transformer_blocks.40.ff.net.2.weight": {
1077
- "block_scale_max": 448.0,
1078
- "block_scale_mean": 21.494430541992188,
1079
- "block_scale_min": 3.75,
1080
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1081
- "shape": [
1082
- 5376,
1083
- 14336
1084
- ],
1085
- "source_dtype": "bfloat16",
1086
- "weight_global_amax": 9.4375,
1087
- "weight_saturation_rate": 0.06319945305585861,
1088
- "weight_tensor_scale": 0.0035109748132526875
1089
- },
1090
- "transformer_blocks.41.ff.net.0.proj.weight": {
1091
- "block_scale_max": 448.0,
1092
- "block_scale_mean": 13.759907722473145,
1093
- "block_scale_min": 2.0,
1094
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1095
- "shape": [
1096
- 28672,
1097
- 5376
1098
- ],
1099
- "source_dtype": "bfloat16",
1100
- "weight_global_amax": 14.3125,
1101
- "weight_saturation_rate": 0.0633372962474823,
1102
- "weight_tensor_scale": 0.0053245909512043
1103
- },
1104
- "transformer_blocks.41.ff.net.2.weight": {
1105
- "block_scale_max": 448.0,
1106
- "block_scale_mean": 18.748844146728516,
1107
- "block_scale_min": 2.0,
1108
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1109
- "shape": [
1110
- 5376,
1111
- 14336
1112
- ],
1113
- "source_dtype": "bfloat16",
1114
- "weight_global_amax": 10.8125,
1115
- "weight_saturation_rate": 0.06360166519880295,
1116
- "weight_tensor_scale": 0.0040225074626505375
1117
- },
1118
- "transformer_blocks.42.ff.net.0.proj.weight": {
1119
- "block_scale_max": 448.0,
1120
- "block_scale_mean": 14.95506763458252,
1121
- "block_scale_min": 0.875,
1122
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1123
- "shape": [
1124
- 28672,
1125
- 5376
1126
- ],
1127
- "source_dtype": "bfloat16",
1128
- "weight_global_amax": 12.9375,
1129
- "weight_saturation_rate": 0.06348538398742676,
1130
- "weight_tensor_scale": 0.00481305830180645
1131
- },
1132
- "transformer_blocks.42.ff.net.2.weight": {
1133
- "block_scale_max": 448.0,
1134
- "block_scale_mean": 20.942319869995117,
1135
- "block_scale_min": 3.0,
1136
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1137
- "shape": [
1138
- 5376,
1139
- 14336
1140
- ],
1141
- "source_dtype": "bfloat16",
1142
- "weight_global_amax": 9.8125,
1143
- "weight_saturation_rate": 0.06326501071453094,
1144
- "weight_tensor_scale": 0.003650483675301075
1145
- },
1146
- "transformer_blocks.43.ff.net.0.proj.weight": {
1147
- "block_scale_max": 448.0,
1148
- "block_scale_mean": 9.634346008300781,
1149
- "block_scale_min": 0.625,
1150
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1151
- "shape": [
1152
- 28672,
1153
- 5376
1154
- ],
1155
- "source_dtype": "bfloat16",
1156
- "weight_global_amax": 20.0,
1157
- "weight_saturation_rate": 0.06337258219718933,
1158
- "weight_tensor_scale": 0.0074404762126505375
1159
- },
1160
- "transformer_blocks.43.ff.net.2.weight": {
1161
- "block_scale_max": 448.0,
1162
- "block_scale_mean": 11.070359230041504,
1163
- "block_scale_min": 1.375,
1164
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1165
- "shape": [
1166
- 5376,
1167
- 14336
1168
- ],
1169
- "source_dtype": "bfloat16",
1170
- "weight_global_amax": 19.0,
1171
- "weight_saturation_rate": 0.06310183554887772,
1172
- "weight_tensor_scale": 0.007068452425301075
1173
- },
1174
- "transformer_blocks.44.ff.net.0.proj.weight": {
1175
- "block_scale_max": 448.0,
1176
- "block_scale_mean": 24.597654342651367,
1177
- "block_scale_min": 1.25,
1178
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1179
- "shape": [
1180
- 28672,
1181
- 5376
1182
- ],
1183
- "source_dtype": "bfloat16",
1184
- "weight_global_amax": 7.75,
1185
- "weight_saturation_rate": 0.06331823021173477,
1186
- "weight_tensor_scale": 0.0028831844683736563
1187
- },
1188
- "transformer_blocks.44.ff.net.2.weight": {
1189
- "block_scale_max": 448.0,
1190
- "block_scale_mean": 14.781774520874023,
1191
- "block_scale_min": 1.75,
1192
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1193
- "shape": [
1194
- 5376,
1195
- 14336
1196
- ],
1197
- "source_dtype": "bfloat16",
1198
- "weight_global_amax": 14.3125,
1199
- "weight_saturation_rate": 0.06344424933195114,
1200
- "weight_tensor_scale": 0.0053245909512043
1201
- },
1202
- "transformer_blocks.45.ff.net.0.proj.weight": {
1203
- "block_scale_max": 448.0,
1204
- "block_scale_mean": 23.468605041503906,
1205
- "block_scale_min": 1.625,
1206
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1207
- "shape": [
1208
- 28672,
1209
- 5376
1210
- ],
1211
- "source_dtype": "bfloat16",
1212
- "weight_global_amax": 8.0625,
1213
- "weight_saturation_rate": 0.06321755051612854,
1214
- "weight_tensor_scale": 0.0029994419310241938
1215
- },
1216
- "transformer_blocks.45.ff.net.2.weight": {
1217
- "block_scale_max": 448.0,
1218
- "block_scale_mean": 12.22935962677002,
1219
- "block_scale_min": 1.625,
1220
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1221
- "shape": [
1222
- 5376,
1223
- 14336
1224
- ],
1225
- "source_dtype": "bfloat16",
1226
- "weight_global_amax": 17.375,
1227
- "weight_saturation_rate": 0.06304015219211578,
1228
- "weight_tensor_scale": 0.0064639137126505375
1229
- },
1230
- "transformer_blocks.46.ff.net.0.proj.weight": {
1231
- "block_scale_max": 448.0,
1232
- "block_scale_mean": 18.22333335876465,
1233
- "block_scale_min": 1.0,
1234
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1235
- "shape": [
1236
- 28672,
1237
- 5376
1238
- ],
1239
- "source_dtype": "bfloat16",
1240
- "weight_global_amax": 10.25,
1241
- "weight_saturation_rate": 0.06341880559921265,
1242
- "weight_tensor_scale": 0.003813244169577956
1243
- },
1244
- "transformer_blocks.46.ff.net.2.weight": {
1245
- "block_scale_max": 448.0,
1246
- "block_scale_mean": 15.134147644042969,
1247
- "block_scale_min": 1.875,
1248
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1249
- "shape": [
1250
- 5376,
1251
- 14336
1252
- ],
1253
- "source_dtype": "bfloat16",
1254
- "weight_global_amax": 13.75,
1255
- "weight_saturation_rate": 0.0635608360171318,
1256
- "weight_tensor_scale": 0.005115327425301075
1257
- },
1258
- "transformer_blocks.47.ff.net.0.proj.weight": {
1259
- "block_scale_max": 448.0,
1260
- "block_scale_mean": 12.564708709716797,
1261
- "block_scale_min": 0.8125,
1262
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1263
- "shape": [
1264
- 28672,
1265
- 5376
1266
- ],
1267
- "source_dtype": "bfloat16",
1268
- "weight_global_amax": 14.375,
1269
- "weight_saturation_rate": 0.06328296661376953,
1270
- "weight_tensor_scale": 0.00534784235060215
1271
- },
1272
- "transformer_blocks.47.ff.net.2.weight": {
1273
- "block_scale_max": 448.0,
1274
- "block_scale_mean": 10.440210342407227,
1275
- "block_scale_min": 1.0,
1276
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1277
- "shape": [
1278
- 5376,
1279
- 14336
1280
- ],
1281
- "source_dtype": "bfloat16",
1282
- "weight_global_amax": 19.5,
1283
- "weight_saturation_rate": 0.06382088363170624,
1284
- "weight_tensor_scale": 0.00725446455180645
1285
- },
1286
- "transformer_blocks.48.ff.net.0.proj.weight": {
1287
- "block_scale_max": 448.0,
1288
- "block_scale_mean": 16.887142181396484,
1289
- "block_scale_min": 0.9375,
1290
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1291
- "shape": [
1292
- 28672,
1293
- 5376
1294
- ],
1295
- "source_dtype": "bfloat16",
1296
- "weight_global_amax": 10.25,
1297
- "weight_saturation_rate": 0.06341958791017532,
1298
- "weight_tensor_scale": 0.003813244169577956
1299
- },
1300
- "transformer_blocks.48.ff.net.2.weight": {
1301
- "block_scale_max": 448.0,
1302
- "block_scale_mean": 13.37950611114502,
1303
- "block_scale_min": 0.8125,
1304
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1305
- "shape": [
1306
- 5376,
1307
- 14336
1308
- ],
1309
- "source_dtype": "bfloat16",
1310
- "weight_global_amax": 14.125,
1311
- "weight_saturation_rate": 0.06343883275985718,
1312
- "weight_tensor_scale": 0.0052548362873494625
1313
- },
1314
- "transformer_blocks.49.ff.net.0.proj.weight": {
1315
- "block_scale_max": 448.0,
1316
- "block_scale_mean": 11.566964149475098,
1317
- "block_scale_min": 0.875,
1318
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1319
- "shape": [
1320
- 28672,
1321
- 5376
1322
- ],
1323
- "source_dtype": "bfloat16",
1324
- "weight_global_amax": 13.1875,
1325
- "weight_saturation_rate": 0.06327934563159943,
1326
- "weight_tensor_scale": 0.00490606389939785
1327
- },
1328
- "transformer_blocks.49.ff.net.2.weight": {
1329
- "block_scale_max": 448.0,
1330
- "block_scale_mean": 7.344207286834717,
1331
- "block_scale_min": 0.5,
1332
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1333
- "shape": [
1334
- 5376,
1335
- 14336
1336
- ],
1337
- "source_dtype": "bfloat16",
1338
- "weight_global_amax": 7.8125,
1339
- "weight_saturation_rate": 0.06335245072841644,
1340
- "weight_tensor_scale": 0.00290643610060215
1341
- },
1342
- "transformer_blocks.5.ff.net.0.proj.weight": {
1343
- "block_scale_max": 448.0,
1344
- "block_scale_mean": 40.42047882080078,
1345
- "block_scale_min": 1.25,
1346
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1347
- "shape": [
1348
- 28672,
1349
- 5376
1350
- ],
1351
- "source_dtype": "bfloat16",
1352
- "weight_global_amax": 4.15625,
1353
- "weight_saturation_rate": 0.063423752784729,
1354
- "weight_tensor_scale": 0.0015462239971384406
1355
- },
1356
- "transformer_blocks.5.ff.net.2.weight": {
1357
- "block_scale_max": 448.0,
1358
- "block_scale_mean": 50.378971099853516,
1359
- "block_scale_min": 7.0,
1360
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1361
- "shape": [
1362
- 5376,
1363
- 14336
1364
- ],
1365
- "source_dtype": "bfloat16",
1366
- "weight_global_amax": 3.515625,
1367
- "weight_saturation_rate": 0.06323359906673431,
1368
- "weight_tensor_scale": 0.0013078962219879031
1369
- },
1370
- "transformer_blocks.6.ff.net.0.proj.weight": {
1371
- "block_scale_max": 448.0,
1372
- "block_scale_mean": 44.86921310424805,
1373
- "block_scale_min": 1.5,
1374
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1375
- "shape": [
1376
- 28672,
1377
- 5376
1378
- ],
1379
- "source_dtype": "bfloat16",
1380
- "weight_global_amax": 3.734375,
1381
- "weight_saturation_rate": 0.06323440372943878,
1382
- "weight_tensor_scale": 0.0013892764691263437
1383
- },
1384
- "transformer_blocks.6.ff.net.2.weight": {
1385
- "block_scale_max": 448.0,
1386
- "block_scale_mean": 34.11458969116211,
1387
- "block_scale_min": 3.75,
1388
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1389
- "shape": [
1390
- 5376,
1391
- 14336
1392
- ],
1393
- "source_dtype": "bfloat16",
1394
- "weight_global_amax": 5.15625,
1395
- "weight_saturation_rate": 0.06395266950130463,
1396
- "weight_tensor_scale": 0.0019182477844879031
1397
- },
1398
- "transformer_blocks.7.ff.net.0.proj.weight": {
1399
- "block_scale_max": 448.0,
1400
- "block_scale_mean": 36.13467025756836,
1401
- "block_scale_min": 1.0,
1402
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1403
- "shape": [
1404
- 28672,
1405
- 5376
1406
- ],
1407
- "source_dtype": "bfloat16",
1408
- "weight_global_amax": 4.625,
1409
- "weight_saturation_rate": 0.06365388631820679,
1410
- "weight_tensor_scale": 0.0017206101911142468
1411
- },
1412
- "transformer_blocks.7.ff.net.2.weight": {
1413
- "block_scale_max": 448.0,
1414
- "block_scale_mean": 36.49232864379883,
1415
- "block_scale_min": 6.0,
1416
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1417
- "shape": [
1418
- 5376,
1419
- 14336
1420
- ],
1421
- "source_dtype": "bfloat16",
1422
- "weight_global_amax": 4.8125,
1423
- "weight_saturation_rate": 0.06392229348421097,
1424
- "weight_tensor_scale": 0.0017903646221384406
1425
- },
1426
- "transformer_blocks.8.ff.net.0.proj.weight": {
1427
- "block_scale_max": 448.0,
1428
- "block_scale_mean": 59.098426818847656,
1429
- "block_scale_min": 1.375,
1430
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1431
- "shape": [
1432
- 28672,
1433
- 5376
1434
- ],
1435
- "source_dtype": "bfloat16",
1436
- "weight_global_amax": 2.796875,
1437
- "weight_saturation_rate": 0.06343978643417358,
1438
- "weight_tensor_scale": 0.0010405040811747313
1439
- },
1440
- "transformer_blocks.8.ff.net.2.weight": {
1441
- "block_scale_max": 448.0,
1442
- "block_scale_mean": 73.22038269042969,
1443
- "block_scale_min": 9.0,
1444
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1445
- "shape": [
1446
- 5376,
1447
- 14336
1448
- ],
1449
- "source_dtype": "bfloat16",
1450
- "weight_global_amax": 2.359375,
1451
- "weight_saturation_rate": 0.06389093399047852,
1452
- "weight_tensor_scale": 0.0008777437033131719
1453
- },
1454
- "transformer_blocks.9.ff.net.0.proj.weight": {
1455
- "block_scale_max": 448.0,
1456
- "block_scale_mean": 34.58193588256836,
1457
- "block_scale_min": 0.875,
1458
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1459
- "shape": [
1460
- 28672,
1461
- 5376
1462
- ],
1463
- "source_dtype": "bfloat16",
1464
- "weight_global_amax": 4.65625,
1465
- "weight_saturation_rate": 0.06360305845737457,
1466
- "weight_tensor_scale": 0.0017322358908131719
1467
- },
1468
- "transformer_blocks.9.ff.net.2.weight": {
1469
- "block_scale_max": 448.0,
1470
- "block_scale_mean": 50.91960144042969,
1471
- "block_scale_min": 6.0,
1472
- "calibration": "static_mse_fp8_e4m3_scale_sweep_126_candidates",
1473
- "shape": [
1474
- 5376,
1475
- 14336
1476
- ],
1477
- "source_dtype": "bfloat16",
1478
- "weight_global_amax": 3.3125,
1479
- "weight_saturation_rate": 0.06345276534557343,
1480
- "weight_tensor_scale": 0.0012323289411142468
1481
- }
1482
- }
1483
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
vae/config.json CHANGED
@@ -93,5 +93,36 @@
93
  3.1627357006073,
94
  2.2816812992095947,
95
  2.6127843856811523
96
- ]
97
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
93
  3.1627357006073,
94
  2.2816812992095947,
95
  2.6127843856811523
96
+ ],
97
+ "quantization_config": {
98
+ "config_groups": {
99
+ "group_0": {
100
+ "input_activations": {
101
+ "dynamic": false,
102
+ "num_bits": 4,
103
+ "type": "float",
104
+ "group_size": 16
105
+ },
106
+ "weights": {
107
+ "dynamic": false,
108
+ "num_bits": 4,
109
+ "type": "float",
110
+ "group_size": 16
111
+ },
112
+ "targets": [
113
+ "Linear"
114
+ ]
115
+ }
116
+ },
117
+ "ignore": [
118
+ "decoder.proj_in",
119
+ "decoder.proj_out"
120
+ ],
121
+ "quant_algo": "NVFP4",
122
+ "producer": {
123
+ "name": "modelopt",
124
+ "version": "0.47.0rc0-63-g6a4b3f14"
125
+ },
126
+ "quant_method": "modelopt"
127
+ }
128
+ }
vae/diffusion_pytorch_model-00001-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:45c80b94a721ec95956775054b3abbf3ae7c61bae695136c4a6e32be5884d3e1
3
- size 735686172
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ebd5775d1b4f1f9661699a26e57cad8a16dc6e56a7fee790962f757c3e64955
3
+ size 735698496
vae/diffusion_pytorch_model-00002-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c8ffb841645a7ac8a30be90ed957cb4d61041c4efb8b95509dc8fb18abe8af89
3
- size 976666252
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:066ac79f7b3fbd5bd0a36fdbc4b724da83742b8b8a27543e1c4eeb50fe3c8a07
3
+ size 976677176
vae/diffusion_pytorch_model.safetensors.index.json CHANGED
The diff for this file is too large to render. See raw diff
 
vae/modelopt-component-manifest.json DELETED
The diff for this file is too large to render. See raw diff