qshou-coder commited on
Commit
426aa5b
·
verified ·
1 Parent(s): d62be2c

Add complete pi05 and GR00T stack-cups-148 final checkpoints

Browse files

Preserve all native policy/configuration/optimizer files of the final 10000-step pi05 and 21100-step GR00T checkpoints, with per-file SHA256 manifests.

This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +8 -0
  2. baseline/gr00t_stack_cups_148/MANIFEST.json +193 -0
  3. baseline/gr00t_stack_cups_148/README.md +13 -0
  4. baseline/gr00t_stack_cups_148/checkpoint-21100/config.json +89 -0
  5. baseline/gr00t_stack_cups_148/checkpoint-21100/embodiment_id.json +58 -0
  6. baseline/gr00t_stack_cups_148/checkpoint-21100/experiment_cfg/conf.yaml +252 -0
  7. baseline/gr00t_stack_cups_148/checkpoint-21100/experiment_cfg/config.yaml +250 -0
  8. baseline/gr00t_stack_cups_148/checkpoint-21100/experiment_cfg/dataset_statistics.json +288 -0
  9. baseline/gr00t_stack_cups_148/checkpoint-21100/experiment_cfg/final_model_config.json +56 -0
  10. baseline/gr00t_stack_cups_148/checkpoint-21100/experiment_cfg/final_processor_config.json +0 -0
  11. baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +3 -0
  12. baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +3 -0
  13. baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt +3 -0
  14. baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt +3 -0
  15. baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt +3 -0
  16. baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt +3 -0
  17. baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt +3 -0
  18. baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt +3 -0
  19. baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/mp_rank_00_model_states.pt +3 -0
  20. baseline/gr00t_stack_cups_148/checkpoint-21100/latest +1 -0
  21. baseline/gr00t_stack_cups_148/checkpoint-21100/model-00001-of-00002.safetensors +3 -0
  22. baseline/gr00t_stack_cups_148/checkpoint-21100/model-00002-of-00002.safetensors +3 -0
  23. baseline/gr00t_stack_cups_148/checkpoint-21100/model.safetensors.index.json +0 -0
  24. baseline/gr00t_stack_cups_148/checkpoint-21100/processor_config.json +1164 -0
  25. baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_0.pth +3 -0
  26. baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_1.pth +3 -0
  27. baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_2.pth +3 -0
  28. baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_3.pth +3 -0
  29. baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_4.pth +3 -0
  30. baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_5.pth +3 -0
  31. baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_6.pth +3 -0
  32. baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_7.pth +3 -0
  33. baseline/gr00t_stack_cups_148/checkpoint-21100/scheduler.pt +3 -0
  34. baseline/gr00t_stack_cups_148/checkpoint-21100/statistics.json +0 -0
  35. baseline/gr00t_stack_cups_148/checkpoint-21100/trainer_state.json +0 -0
  36. baseline/gr00t_stack_cups_148/checkpoint-21100/training_args.bin +3 -0
  37. baseline/gr00t_stack_cups_148/checkpoint-21100/wandb_config.json +1 -0
  38. baseline/gr00t_stack_cups_148/checkpoint-21100/zero_to_fp32.py +760 -0
  39. baseline/pi05_stack_cups_148/10000/_CHECKPOINT_METADATA +1 -0
  40. baseline/pi05_stack_cups_148/10000/assets/galaxea_r1_lite_stack_cups/norm_stats.json +136 -0
  41. baseline/pi05_stack_cups_148/10000/params/_METADATA +1 -0
  42. baseline/pi05_stack_cups_148/10000/params/_sharding +1 -0
  43. baseline/pi05_stack_cups_148/10000/params/array_metadatas/process_0 +1 -0
  44. baseline/pi05_stack_cups_148/10000/params/d/4c1273032c55c08244f3a733657fcac0 +0 -0
  45. baseline/pi05_stack_cups_148/10000/params/manifest.ocdbt +0 -0
  46. baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/319e3d0585c3de362e83981c20d3fe89 +3 -0
  47. baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/3c032c830f6cc7175afe2811c1a18827 +3 -0
  48. baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/58f68e6e347aedcae64dc9ca8caa082b +3 -0
  49. baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/5b8109f6166eafdac21828028a913405 +3 -0
  50. baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/6bd188f310ebefce038484a33ca74dc0 +0 -0
.gitattributes CHANGED
@@ -229,3 +229,11 @@ baseline/pi05-zipper-all/14000/train_state/ocdbt.process_0/d/88a6baac8507976c42b
229
  baseline/pi05-zipper-all/14000/train_state/ocdbt.process_0/d/97a05fcaeb97ea72c8b777a814e547de filter=lfs diff=lfs merge=lfs -text
230
  baseline/pi05-zipper-all/14000/train_state/ocdbt.process_0/d/a31ad1697a0feddf0d914d1147991fd6 filter=lfs diff=lfs merge=lfs -text
231
  spacedreamer_20epochs/assets/Wan2.2-TI2V-5B/google/umt5-xxl/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
229
  baseline/pi05-zipper-all/14000/train_state/ocdbt.process_0/d/97a05fcaeb97ea72c8b777a814e547de filter=lfs diff=lfs merge=lfs -text
230
  baseline/pi05-zipper-all/14000/train_state/ocdbt.process_0/d/a31ad1697a0feddf0d914d1147991fd6 filter=lfs diff=lfs merge=lfs -text
231
  spacedreamer_20epochs/assets/Wan2.2-TI2V-5B/google/umt5-xxl/tokenizer.json filter=lfs diff=lfs merge=lfs -text
232
+ baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/319e3d0585c3de362e83981c20d3fe89 filter=lfs diff=lfs merge=lfs -text
233
+ baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/3c032c830f6cc7175afe2811c1a18827 filter=lfs diff=lfs merge=lfs -text
234
+ baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/5b8109f6166eafdac21828028a913405 filter=lfs diff=lfs merge=lfs -text
235
+ baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/58f68e6e347aedcae64dc9ca8caa082b filter=lfs diff=lfs merge=lfs -text
236
+ baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/bbc98084709472d677f45f8c04bcd7e0 filter=lfs diff=lfs merge=lfs -text
237
+ baseline/pi05_stack_cups_148/10000/train_state/ocdbt.process_0/d/62d19b5a1a2b0a593ba8aa0b11135d26 filter=lfs diff=lfs merge=lfs -text
238
+ baseline/pi05_stack_cups_148/10000/train_state/ocdbt.process_0/d/a9869fb978e9a74072b16455e57b1bb7 filter=lfs diff=lfs merge=lfs -text
239
+ baseline/pi05_stack_cups_148/10000/train_state/ocdbt.process_0/d/8e53ac7baf9db1a070b2ac54307949c1 filter=lfs diff=lfs merge=lfs -text
baseline/gr00t_stack_cups_148/MANIFEST.json ADDED
@@ -0,0 +1,193 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema_version": 1,
3
+ "task": "stack_cups",
4
+ "dataset": {
5
+ "episodes": 148,
6
+ "old_episodes": 81,
7
+ "new_episodes": 67,
8
+ "task": "stack_cups"
9
+ },
10
+ "optimizer_steps": 21100,
11
+ "epochs": 50,
12
+ "action_horizon": 40,
13
+ "training_job": "bifrost-2026110617482401-shouqx",
14
+ "checkpoint_directory": "checkpoint-21100",
15
+ "files": [
16
+ {
17
+ "path": "checkpoint-21100/config.json",
18
+ "size": 2210,
19
+ "sha256": "a9218c8da076839cc56572cd0d732908c691c6162c4f820bae20a3fbe3de7ce4"
20
+ },
21
+ {
22
+ "path": "checkpoint-21100/embodiment_id.json",
23
+ "size": 2278,
24
+ "sha256": "c2620886f99abb1ded1e650539dbf37ce36d74c8709a9d5b137053b4127284b6"
25
+ },
26
+ {
27
+ "path": "checkpoint-21100/experiment_cfg/conf.yaml",
28
+ "size": 5993,
29
+ "sha256": "9b78ff7d1a57fa8ade10da9c80057237e58fc44e9877a4adf9c39a3d61d87af3"
30
+ },
31
+ {
32
+ "path": "checkpoint-21100/experiment_cfg/config.yaml",
33
+ "size": 5959,
34
+ "sha256": "44bd448809f236ba9971c402a793b5c5343f0773a99aec33ffa2805363bd9893"
35
+ },
36
+ {
37
+ "path": "checkpoint-21100/experiment_cfg/dataset_statistics.json",
38
+ "size": 6511,
39
+ "sha256": "80b7e441e87693041fb1167bde36e80b16ca87c640d4e726c37b9d35ff283eb2"
40
+ },
41
+ {
42
+ "path": "checkpoint-21100/experiment_cfg/final_model_config.json",
43
+ "size": 1543,
44
+ "sha256": "3cfb92dc78faef4e74ea51d2313adad02028cf21f1070e54210d0862051142de"
45
+ },
46
+ {
47
+ "path": "checkpoint-21100/experiment_cfg/final_processor_config.json",
48
+ "size": 2548956,
49
+ "sha256": "fc94d947bae756babf51187c79d21dc5bb8a0ab2653417bd3d7e63807fb47fae"
50
+ },
51
+ {
52
+ "path": "checkpoint-21100/global_step21100/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt",
53
+ "size": 3041236165,
54
+ "sha256": "a5b2a2c4cffec3ae3991d6274a15e78b7952b499b0fa489da2ee9a8e0a6008da"
55
+ },
56
+ {
57
+ "path": "checkpoint-21100/global_step21100/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt",
58
+ "size": 3041225157,
59
+ "sha256": "eb8fccd7b7d4f5af384d59b9a1af24521c4ce7cd9745ce6eb8db1a0c70fed841"
60
+ },
61
+ {
62
+ "path": "checkpoint-21100/global_step21100/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt",
63
+ "size": 3041223941,
64
+ "sha256": "3e9839009b02b34dd7677f1ea056cbbb4fdbb9c4a35140bbdcbf81638e0855e0"
65
+ },
66
+ {
67
+ "path": "checkpoint-21100/global_step21100/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt",
68
+ "size": 3041223877,
69
+ "sha256": "0a8c2edf2f1ad38063e5160167d1dff1b228a9696099bcac56637accccaf8afa"
70
+ },
71
+ {
72
+ "path": "checkpoint-21100/global_step21100/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt",
73
+ "size": 3041224069,
74
+ "sha256": "0b8a808e771ee5820faf3b48a5b2a90bbf0c12fe8e68ba1d64871fe00e73238b"
75
+ },
76
+ {
77
+ "path": "checkpoint-21100/global_step21100/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt",
78
+ "size": 3041223109,
79
+ "sha256": "92fb3791e263da034ea8aaf5f5c3a3fc330d3eb060d1d81e2b4a3e119c749f86"
80
+ },
81
+ {
82
+ "path": "checkpoint-21100/global_step21100/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt",
83
+ "size": 3041214405,
84
+ "sha256": "097145ed1d18ed72c14003f6890ec728486894ae82cd00f086a61474a9fdc661"
85
+ },
86
+ {
87
+ "path": "checkpoint-21100/global_step21100/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt",
88
+ "size": 3041221061,
89
+ "sha256": "683215a0eaa901719b76eb940067dcb0fef2faf2782c9844293e50b664738bfe"
90
+ },
91
+ {
92
+ "path": "checkpoint-21100/global_step21100/mp_rank_00_model_states.pt",
93
+ "size": 8521564777,
94
+ "sha256": "aab77284037440d98a7c1a79e6057fda9e34432e033e0fa2580cc66b8f69a6bf"
95
+ },
96
+ {
97
+ "path": "checkpoint-21100/latest",
98
+ "size": 16,
99
+ "sha256": "bce45f95cfc571e5c3bc834b246305b7e5f2fc46ee84dcd111949baa5f97d598"
100
+ },
101
+ {
102
+ "path": "checkpoint-21100/model-00001-of-00002.safetensors",
103
+ "size": 4990519232,
104
+ "sha256": "71789b3b904446ebf11712fa0f979ecd22495a7aea31e9183c6f0d410f3b3020"
105
+ },
106
+ {
107
+ "path": "checkpoint-21100/model-00002-of-00002.safetensors",
108
+ "size": 1919980184,
109
+ "sha256": "e39ce9aff702f34e28b9b312b3f23e8578dbe5f11374bbec27c48756adbf3d28"
110
+ },
111
+ {
112
+ "path": "checkpoint-21100/model.safetensors.index.json",
113
+ "size": 104985,
114
+ "sha256": "407804ea5a62f4f8823f48811ae0edbb82fac101e9cf4d7273e6e2f692bb4d59"
115
+ },
116
+ {
117
+ "path": "checkpoint-21100/processor_config.json",
118
+ "size": 27465,
119
+ "sha256": "ac8f8e03dcdeb5ea4f8adbe3291aecf7feaea4f623b63d83456334592c1b5214"
120
+ },
121
+ {
122
+ "path": "checkpoint-21100/rng_state_0.pth",
123
+ "size": 16325,
124
+ "sha256": "34f7c22cde7f1679513776d3a56693b76b6362a25d15f9a7f7edbf1ad0da8c7f"
125
+ },
126
+ {
127
+ "path": "checkpoint-21100/rng_state_1.pth",
128
+ "size": 16325,
129
+ "sha256": "a2151ff4c2d3507f84bce96d29c29fe5a210030b53eb514ed9723e3505798940"
130
+ },
131
+ {
132
+ "path": "checkpoint-21100/rng_state_2.pth",
133
+ "size": 16325,
134
+ "sha256": "26856c64253080822c4206354e8996bf26cbc791a04f2f7e9ade8d6c293f2163"
135
+ },
136
+ {
137
+ "path": "checkpoint-21100/rng_state_3.pth",
138
+ "size": 16325,
139
+ "sha256": "7b37819809978ef21babf985c8157eaf94a9b500c6c1bc971c338e4e387dd5ec"
140
+ },
141
+ {
142
+ "path": "checkpoint-21100/rng_state_4.pth",
143
+ "size": 16325,
144
+ "sha256": "b85d8fb887ef5a0b45586c8b52dc1919ff7dde61e51562a7c661c377e0f1dc61"
145
+ },
146
+ {
147
+ "path": "checkpoint-21100/rng_state_5.pth",
148
+ "size": 16325,
149
+ "sha256": "a75c620a65e592f7b5fd0a2a60421a2bc3432585b4fefa9a40e1a77929efd561"
150
+ },
151
+ {
152
+ "path": "checkpoint-21100/rng_state_6.pth",
153
+ "size": 16325,
154
+ "sha256": "ad4e0327d9b2d593c2760bc2a03927ccf6135d7de353c3953f46957588fa805d"
155
+ },
156
+ {
157
+ "path": "checkpoint-21100/rng_state_7.pth",
158
+ "size": 16325,
159
+ "sha256": "7b2906f2dd09565832d6221075d0a3ae55a23117020a3d73f534443ab9c268c0"
160
+ },
161
+ {
162
+ "path": "checkpoint-21100/scheduler.pt",
163
+ "size": 1401,
164
+ "sha256": "b493a49f0d24b6a59c451e94df8429955744031a4a2d80f354e24dcc8f906076"
165
+ },
166
+ {
167
+ "path": "checkpoint-21100/statistics.json",
168
+ "size": 4299604,
169
+ "sha256": "75197ba4742089064afdd763c00f56aba68a1b3d392a652356496ca09ec7562c"
170
+ },
171
+ {
172
+ "path": "checkpoint-21100/trainer_state.json",
173
+ "size": 139661,
174
+ "sha256": "40f38a3409b09dbf41edd7b058573d851e9393e6d0cd934b4b84f7c9f68e8a3a"
175
+ },
176
+ {
177
+ "path": "checkpoint-21100/training_args.bin",
178
+ "size": 7953,
179
+ "sha256": "fc07b7bffcf221721b9dce804b4fd4ba4d91c1bfab12a2c19565c3d7aff0e6ec"
180
+ },
181
+ {
182
+ "path": "checkpoint-21100/wandb_config.json",
183
+ "size": 59,
184
+ "sha256": "0ac8b31d2c323a7d925344277f66baed9cdb159dc8097c6553e5d7203fd298ae"
185
+ },
186
+ {
187
+ "path": "checkpoint-21100/zero_to_fp32.py",
188
+ "size": 33270,
189
+ "sha256": "e8a2553b434ac043ec63b022db1e768277acf8d4e70f10a2842db6d238977135"
190
+ }
191
+ ],
192
+ "complete_native_checkpoint": true
193
+ }
baseline/gr00t_stack_cups_148/README.md ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # GR00T: stack cups, 148 episodes
2
+
3
+ Training data: 81 original episodes plus 67 added episodes (148 total).
4
+
5
+ Final optimizer step: 21100; action horizon: 40.
6
+ Native checkpoint: [checkpoint-21100](./checkpoint-21100).
7
+
8
+ The complete native checkpoint is preserved, including policy weights, normalization/configuration assets, and training state.
9
+ MANIFEST.json lists the size and SHA-256 of every native checkpoint file.
10
+
11
+ Training completion and CPU parameter read/restore checks passed. This upload does not claim robot inference or downstream evaluation validation.
12
+
13
+ The policy weights are the two SafeTensors shards and their index. The directory also includes native processor/configuration/statistics files and DeepSpeed training state.
baseline/gr00t_stack_cups_148/checkpoint-21100/config.json ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "action_horizon": 40,
3
+ "add_pos_embed": true,
4
+ "apply_sincos_state_encoding": false,
5
+ "architectures": [
6
+ "Gr00tN1d7"
7
+ ],
8
+ "attn_dropout": 0.2,
9
+ "attn_implementation": null,
10
+ "backbone_embedding_dim": 2048,
11
+ "backbone_trainable_params_fp32": true,
12
+ "color_jitter_params": {
13
+ "brightness": 0.3,
14
+ "contrast": 0.4,
15
+ "hue": 0.08,
16
+ "saturation": 0.5
17
+ },
18
+ "crop_fraction": 0.95,
19
+ "diffusion_model_cfg": {
20
+ "attention_head_dim": 48,
21
+ "dropout": 0.2,
22
+ "final_dropout": true,
23
+ "interleave_self_attention": true,
24
+ "norm_type": "ada_norm",
25
+ "num_attention_heads": 32,
26
+ "num_layers": 32,
27
+ "output_dim": 1024,
28
+ "positional_embeddings": null
29
+ },
30
+ "dtype": "bfloat16",
31
+ "exclude_state": false,
32
+ "formalize_language": true,
33
+ "hidden_size": 1024,
34
+ "image_crop_size": [
35
+ 230,
36
+ 230
37
+ ],
38
+ "image_target_size": [
39
+ 256,
40
+ 256
41
+ ],
42
+ "letter_box_transform": false,
43
+ "load_bf16": false,
44
+ "max_action_dim": 132,
45
+ "max_num_embodiments": 32,
46
+ "max_seq_len": 1024,
47
+ "max_state_dim": 132,
48
+ "model_dtype": "bfloat16",
49
+ "model_name": "/code/ironflow-pico-runtime-v1/weights/Cosmos-Reason2-2B",
50
+ "model_type": "Gr00tN1d7",
51
+ "noise_beta_alpha": 1.5,
52
+ "noise_beta_beta": 1.0,
53
+ "noise_s": 0.999,
54
+ "num_inference_timesteps": 4,
55
+ "num_timestep_buckets": 1000,
56
+ "random_history_crop": true,
57
+ "random_rotation_angle": 0,
58
+ "reproject_vision": false,
59
+ "rtc_ramp_rate": 6.0,
60
+ "select_layer": 16,
61
+ "shortest_image_edge": 256,
62
+ "state_dropout_prob": 0.0,
63
+ "state_gaussian_noise_std": 0.0,
64
+ "transformers_version": "4.57.3",
65
+ "tune_diffusion_model": true,
66
+ "tune_linear": true,
67
+ "tune_llm": false,
68
+ "tune_projector": true,
69
+ "tune_top_llm_layers": 0,
70
+ "tune_visual": true,
71
+ "tune_vlln": true,
72
+ "use_albumentations": true,
73
+ "use_alternate_vl_dit": true,
74
+ "use_flash_attention": true,
75
+ "use_future_tokens": false,
76
+ "use_mean_std": false,
77
+ "use_percentiles": true,
78
+ "use_relative_action": false,
79
+ "use_vl_self_attention": true,
80
+ "use_vlln": true,
81
+ "vl_self_attention_cfg": {
82
+ "attention_head_dim": 64,
83
+ "dropout": 0.2,
84
+ "final_dropout": true,
85
+ "num_attention_heads": 32,
86
+ "num_layers": 4,
87
+ "positional_embeddings": null
88
+ }
89
+ }
baseline/gr00t_stack_cups_148/checkpoint-21100/embodiment_id.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "robocasa_panda_omron": 13,
3
+ "oxe_droid": 17,
4
+ "oxe_fractal": 18,
5
+ "oxe_language_table": 19,
6
+ "oxe_bridge": 20,
7
+ "unknown": 22,
8
+ "gr1_unified": 20,
9
+ "agibot": 26,
10
+ "sim_behavior_r1_pro": 23,
11
+ "xdof": 24,
12
+ "xdof_oss_data": 25,
13
+ "unitree_g1_full_body_with_waist_height_nav_cmd": 25,
14
+ "real_r1_pro_sharpa": 27,
15
+ "real_r1_pro_sharpa_add_view": 27,
16
+ "real_r1_pro_sharpa_relative_arm_joint": 26,
17
+ "real_r1_pro_sharpa_delta_eef": 26,
18
+ "real_r1_pro_sharpa_absolute_eef": 26,
19
+ "real_r1_pro_sharpa_meanstd": 26,
20
+ "real_r1_pro_sharpa_relative_eef": 26,
21
+ "real_r1_pro_sharpa_relative_eef_add_view": 26,
22
+ "real_r1_pro_sharpa_relative_eef_relative_hand": 26,
23
+ "real_r1_pro_sharpa_relative_eef_human": 26,
24
+ "real_r1_pro_sharpa_relative_eef_human_add_view": 26,
25
+ "real_r1_pro_sharpa_relative_eef_human_relative_hand": 26,
26
+ "real_r1_pro_sharpa_relative_eef_egodex": 26,
27
+ "real_r1_pro_sharpa_relative_eef_egodex_relative_hand": 26,
28
+ "real_r1_pro_sharpa_relative_eef_egodex_wrist_only": 26,
29
+ "real_r1_pro_sharpa_relative_eef_maxinsights": 26,
30
+ "real_r1_pro_sharpa_relative_eef_maxinsights_relative_hand": 26,
31
+ "real_r1_pro_sharpa_relative_eef_mecka": 26,
32
+ "real_r1_pro_sharpa_relative_eef_mecka_relative_hand": 26,
33
+ "real_g1_relative_eef_absolute_joints": 25,
34
+ "real_g1_relative_eef_absolute_joints_wrist_cam": 25,
35
+ "real_g1_relative_eef_relative_joints": 25,
36
+ "real_r1_pro_sharpa_relative_eef_relative_hand_relative_joint": 26,
37
+ "real_r1_pro_sharpa_relative_joint": 29,
38
+ "oxe_droid_relative_eef_relative_joint": 24,
39
+ "oxe_droid_relative_eef_relative_joint_swapped": 24,
40
+ "oxe_droid_relative_eef_relative_joint_upweight_z": 24,
41
+ "oxe_droid_relative_eef_relative_joint_upweight_z_swapped": 24,
42
+ "oxe_droid_relative_eef_relative_joint_3view": 24,
43
+ "oxe_droid_relative_eef_relative_joint_3view_swapped": 24,
44
+ "oxe_droid_relative_eef": 24,
45
+ "oxe_droid_joint_position_relative": 24,
46
+ "xdof_relative_eef_relative_joint": 27,
47
+ "xdof_relative_eef_relative_joint_subtask": 27,
48
+ "xdof_relative_eef": 27,
49
+ "xdof_relative_joint": 28,
50
+ "simpler_env_google": 0,
51
+ "simpler_env_widowx": 1,
52
+ "libero_sim": 2,
53
+ "droid_sim": 3,
54
+ "robocasa_gr1_tabletop": 10,
55
+ "new_embodiment": 10,
56
+ "unitree_g1_sonic": 11,
57
+ "unitree_g1_full_state_relative_pose_joints": 25
58
+ }
baseline/gr00t_stack_cups_148/checkpoint-21100/experiment_cfg/conf.yaml ADDED
@@ -0,0 +1,252 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ load_config_path: null
2
+ model:
3
+ model_type: Gr00tN1d7
4
+ model_dtype: bfloat16
5
+ model_name: /code/ironflow-pico-runtime-v1/weights/Cosmos-Reason2-2B
6
+ backbone_model_type: qwen
7
+ model_revision: null
8
+ tune_top_llm_layers: 0
9
+ backbone_embedding_dim: 2048
10
+ tune_llm: false
11
+ tune_visual: true
12
+ select_layer: 16
13
+ reproject_vision: false
14
+ use_flash_attention: true
15
+ load_bf16: false
16
+ backbone_trainable_params_fp32: true
17
+ image_crop_size: null
18
+ image_target_size: null
19
+ shortest_image_edge: 256
20
+ crop_fraction: 1.0
21
+ random_rotation_angle: null
22
+ color_jitter_params:
23
+ brightness: 0.0
24
+ contrast: 0.0
25
+ saturation: 0.0
26
+ hue: 0.0
27
+ use_albumentations_transforms: true
28
+ letter_box_transform: false
29
+ extra_augmentation_config: null
30
+ formalize_language: true
31
+ apply_sincos_state_encoding: false
32
+ use_percentiles: true
33
+ use_relative_action: false
34
+ max_state_dim: 132
35
+ max_action_dim: 132
36
+ action_horizon: 40
37
+ hidden_size: 1024
38
+ input_embedding_dim: 1536
39
+ state_history_length: 1
40
+ add_pos_embed: true
41
+ attn_dropout: 0.2
42
+ use_vlln: true
43
+ max_seq_len: 1024
44
+ use_alternate_vl_dit: true
45
+ attend_text_every_n_blocks: 2
46
+ diffusion_model_cfg:
47
+ attention_head_dim: 48
48
+ dropout: 0.2
49
+ final_dropout: true
50
+ interleave_self_attention: true
51
+ norm_type: ada_norm
52
+ num_attention_heads: 32
53
+ num_layers: 32
54
+ output_dim: 1024
55
+ positional_embeddings: null
56
+ num_inference_timesteps: 4
57
+ noise_beta_alpha: 1.5
58
+ noise_beta_beta: 1.0
59
+ noise_s: 0.999
60
+ num_timestep_buckets: 1000
61
+ tune_projector: true
62
+ tune_diffusion_model: true
63
+ tune_vlln: true
64
+ state_dropout_prob: 0.0
65
+ exclude_state: false
66
+ use_mean_std: false
67
+ max_num_embodiments: 32
68
+ data:
69
+ datasets:
70
+ - dataset_paths:
71
+ - /workspace/group_share/rc-wbc/shouqx@xiaopeng.com/stack_cups_148_four_models_20261106/gr00t/assets
72
+ embodiment_tag: new_embodiment
73
+ mix_ratio: 1.0
74
+ dataset_type: physical_embodiment
75
+ val_dataset_path: null
76
+ modality_configs:
77
+ new_embodiment:
78
+ video:
79
+ delta_indices:
80
+ - 0
81
+ modality_keys:
82
+ - head_left
83
+ - wrist_left
84
+ - wrist_right
85
+ sin_cos_embedding_keys: null
86
+ mean_std_embedding_keys: null
87
+ action_configs: null
88
+ state:
89
+ delta_indices:
90
+ - 0
91
+ modality_keys:
92
+ - left_arm
93
+ - left_gripper
94
+ - right_arm
95
+ - right_gripper
96
+ sin_cos_embedding_keys: null
97
+ mean_std_embedding_keys: null
98
+ action_configs: null
99
+ action:
100
+ delta_indices:
101
+ - 0
102
+ - 1
103
+ - 2
104
+ - 3
105
+ - 4
106
+ - 5
107
+ - 6
108
+ - 7
109
+ - 8
110
+ - 9
111
+ - 10
112
+ - 11
113
+ - 12
114
+ - 13
115
+ - 14
116
+ - 15
117
+ - 16
118
+ - 17
119
+ - 18
120
+ - 19
121
+ - 20
122
+ - 21
123
+ - 22
124
+ - 23
125
+ - 24
126
+ - 25
127
+ - 26
128
+ - 27
129
+ - 28
130
+ - 29
131
+ - 30
132
+ - 31
133
+ - 32
134
+ - 33
135
+ - 34
136
+ - 35
137
+ - 36
138
+ - 37
139
+ - 38
140
+ - 39
141
+ modality_keys:
142
+ - left_arm
143
+ - left_gripper
144
+ - right_arm
145
+ - right_gripper
146
+ sin_cos_embedding_keys: null
147
+ mean_std_embedding_keys: null
148
+ action_configs:
149
+ - rep: ABSOLUTE
150
+ type: NON_EEF
151
+ format: DEFAULT
152
+ state_key: left_arm
153
+ - rep: ABSOLUTE
154
+ type: NON_EEF
155
+ format: DEFAULT
156
+ state_key: left_gripper
157
+ - rep: ABSOLUTE
158
+ type: NON_EEF
159
+ format: DEFAULT
160
+ state_key: right_arm
161
+ - rep: ABSOLUTE
162
+ type: NON_EEF
163
+ format: DEFAULT
164
+ state_key: right_gripper
165
+ language:
166
+ delta_indices:
167
+ - 0
168
+ modality_keys:
169
+ - annotation.human.task_description
170
+ sin_cos_embedding_keys: null
171
+ mean_std_embedding_keys: null
172
+ action_configs: null
173
+ download_cache: false
174
+ shard_size: 1024
175
+ episode_sampling_rate: 0.1
176
+ num_shards_per_epoch: 100000
177
+ ds_weights_alpha: null
178
+ override_pretraining_statistics: true
179
+ mode: single_turn
180
+ random_chop: 0.0
181
+ mock_dataset_mode: false
182
+ shuffle: true
183
+ seed: 0
184
+ multiprocessing_context: fork
185
+ allow_padding: false
186
+ subsample_ratio: 1.0
187
+ image_crop_size:
188
+ - 244
189
+ - 244
190
+ image_target_size:
191
+ - 224
192
+ - 224
193
+ training:
194
+ output_dir: /workspace/group_share/rc-wbc/shouqx@xiaopeng.com/stack_cups_148_four_models_20261106/gr00t/checkpoints
195
+ experiment_name: null
196
+ max_steps: 21100
197
+ global_batch_size: 256
198
+ per_gpu_batch_size: null
199
+ gradient_accumulation_steps: 1
200
+ learning_rate: 5.0e-05
201
+ lr_scheduler_type: cosine
202
+ weight_decay: 1.0e-10
203
+ warmup_ratio: 0.0
204
+ warmup_steps: 500
205
+ max_grad_norm: 1.0
206
+ optim: adamw_torch_fused
207
+ start_from_checkpoint: /code/ironflow-pico-runtime-v1/weights/GR00T-N1.7-3B
208
+ skip_weight_loading: false
209
+ tf32: true
210
+ fp16: false
211
+ bf16: true
212
+ eval_bf16: true
213
+ logging_steps: 20
214
+ save_steps: 2000
215
+ save_total_limit: 2
216
+ save_vl_model: false
217
+ save_only_model: false
218
+ resume_from_checkpoint: false
219
+ upload_checkpoints: false
220
+ upload_every: 1000
221
+ upload_last_n_checkpoints: 5
222
+ max_concurrent_uploads: 2
223
+ eval_strategy: 'no'
224
+ eval_steps: 500
225
+ eval_set_split_ratio: 0.1
226
+ eval_batch_size: 2
227
+ save_best_eval_metric_name: ''
228
+ save_best_eval_metric_greater_is_better: true
229
+ deepspeed_stage: 2
230
+ gradient_checkpointing: false
231
+ transformers_trust_remote_code: true
232
+ transformers_local_files_only: true
233
+ transformers_cache_dir: /code/stack-cups-148-gr00t-bifrost-2026110617482401-shouqx/hf
234
+ transformers_access_token: null
235
+ use_ddp: false
236
+ ddp_bucket_cap_mb: 100
237
+ num_gpus: 8
238
+ dataloader_num_workers: 4
239
+ remove_unused_columns: false
240
+ use_wandb: true
241
+ wandb_project: finetune-gr00t-n1d7
242
+ enable_profiling: false
243
+ max_retries: 3
244
+ assert_loss_less_than: null
245
+ add_rl_callback: false
246
+ enable_open_loop_eval: false
247
+ open_loop_eval_traj_ids:
248
+ - 0
249
+ open_loop_eval_steps_per_traj: 100
250
+ open_loop_eval_plot_indices: null
251
+ max_steps: 21100
252
+ save_steps: 2000
baseline/gr00t_stack_cups_148/checkpoint-21100/experiment_cfg/config.yaml ADDED
@@ -0,0 +1,250 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ load_config_path: null
2
+ model:
3
+ model_type: Gr00tN1d7
4
+ model_dtype: bfloat16
5
+ model_name: /code/ironflow-pico-runtime-v1/weights/Cosmos-Reason2-2B
6
+ backbone_model_type: qwen
7
+ model_revision: null
8
+ tune_top_llm_layers: 0
9
+ backbone_embedding_dim: 2048
10
+ tune_llm: false
11
+ tune_visual: true
12
+ select_layer: 16
13
+ reproject_vision: false
14
+ use_flash_attention: true
15
+ load_bf16: false
16
+ backbone_trainable_params_fp32: true
17
+ image_crop_size: null
18
+ image_target_size: null
19
+ shortest_image_edge: 256
20
+ crop_fraction: 1.0
21
+ random_rotation_angle: null
22
+ color_jitter_params:
23
+ brightness: 0.0
24
+ contrast: 0.0
25
+ saturation: 0.0
26
+ hue: 0.0
27
+ use_albumentations_transforms: true
28
+ letter_box_transform: false
29
+ extra_augmentation_config: null
30
+ formalize_language: true
31
+ apply_sincos_state_encoding: false
32
+ use_percentiles: true
33
+ use_relative_action: false
34
+ max_state_dim: 132
35
+ max_action_dim: 132
36
+ action_horizon: 40
37
+ hidden_size: 1024
38
+ input_embedding_dim: 1536
39
+ state_history_length: 1
40
+ add_pos_embed: true
41
+ attn_dropout: 0.2
42
+ use_vlln: true
43
+ max_seq_len: 1024
44
+ use_alternate_vl_dit: true
45
+ attend_text_every_n_blocks: 2
46
+ diffusion_model_cfg:
47
+ attention_head_dim: 48
48
+ dropout: 0.2
49
+ final_dropout: true
50
+ interleave_self_attention: true
51
+ norm_type: ada_norm
52
+ num_attention_heads: 32
53
+ num_layers: 32
54
+ output_dim: 1024
55
+ positional_embeddings: null
56
+ num_inference_timesteps: 4
57
+ noise_beta_alpha: 1.5
58
+ noise_beta_beta: 1.0
59
+ noise_s: 0.999
60
+ num_timestep_buckets: 1000
61
+ tune_projector: true
62
+ tune_diffusion_model: true
63
+ tune_vlln: true
64
+ state_dropout_prob: 0.0
65
+ exclude_state: false
66
+ use_mean_std: false
67
+ max_num_embodiments: 32
68
+ data:
69
+ datasets:
70
+ - dataset_paths:
71
+ - /workspace/group_share/rc-wbc/shouqx@xiaopeng.com/stack_cups_148_four_models_20261106/gr00t/assets
72
+ embodiment_tag: new_embodiment
73
+ mix_ratio: 1.0
74
+ dataset_type: physical_embodiment
75
+ val_dataset_path: null
76
+ modality_configs:
77
+ new_embodiment:
78
+ video:
79
+ delta_indices:
80
+ - 0
81
+ modality_keys:
82
+ - head_left
83
+ - wrist_left
84
+ - wrist_right
85
+ sin_cos_embedding_keys: null
86
+ mean_std_embedding_keys: null
87
+ action_configs: null
88
+ state:
89
+ delta_indices:
90
+ - 0
91
+ modality_keys:
92
+ - left_arm
93
+ - left_gripper
94
+ - right_arm
95
+ - right_gripper
96
+ sin_cos_embedding_keys: null
97
+ mean_std_embedding_keys: null
98
+ action_configs: null
99
+ action:
100
+ delta_indices:
101
+ - 0
102
+ - 1
103
+ - 2
104
+ - 3
105
+ - 4
106
+ - 5
107
+ - 6
108
+ - 7
109
+ - 8
110
+ - 9
111
+ - 10
112
+ - 11
113
+ - 12
114
+ - 13
115
+ - 14
116
+ - 15
117
+ - 16
118
+ - 17
119
+ - 18
120
+ - 19
121
+ - 20
122
+ - 21
123
+ - 22
124
+ - 23
125
+ - 24
126
+ - 25
127
+ - 26
128
+ - 27
129
+ - 28
130
+ - 29
131
+ - 30
132
+ - 31
133
+ - 32
134
+ - 33
135
+ - 34
136
+ - 35
137
+ - 36
138
+ - 37
139
+ - 38
140
+ - 39
141
+ modality_keys:
142
+ - left_arm
143
+ - left_gripper
144
+ - right_arm
145
+ - right_gripper
146
+ sin_cos_embedding_keys: null
147
+ mean_std_embedding_keys: null
148
+ action_configs:
149
+ - rep: absolute
150
+ type: non_eef
151
+ format: default
152
+ state_key: left_arm
153
+ - rep: absolute
154
+ type: non_eef
155
+ format: default
156
+ state_key: left_gripper
157
+ - rep: absolute
158
+ type: non_eef
159
+ format: default
160
+ state_key: right_arm
161
+ - rep: absolute
162
+ type: non_eef
163
+ format: default
164
+ state_key: right_gripper
165
+ language:
166
+ delta_indices:
167
+ - 0
168
+ modality_keys:
169
+ - annotation.human.task_description
170
+ sin_cos_embedding_keys: null
171
+ mean_std_embedding_keys: null
172
+ action_configs: null
173
+ download_cache: false
174
+ shard_size: 1024
175
+ episode_sampling_rate: 0.1
176
+ num_shards_per_epoch: 100000
177
+ ds_weights_alpha: null
178
+ override_pretraining_statistics: true
179
+ mode: single_turn
180
+ random_chop: 0.0
181
+ mock_dataset_mode: false
182
+ shuffle: true
183
+ seed: 0
184
+ multiprocessing_context: fork
185
+ allow_padding: false
186
+ subsample_ratio: 1.0
187
+ image_crop_size:
188
+ - 244
189
+ - 244
190
+ image_target_size:
191
+ - 224
192
+ - 224
193
+ training:
194
+ output_dir: /workspace/group_share/rc-wbc/shouqx@xiaopeng.com/stack_cups_148_four_models_20261106/gr00t/checkpoints
195
+ experiment_name: null
196
+ max_steps: 21100
197
+ global_batch_size: 256
198
+ per_gpu_batch_size: null
199
+ gradient_accumulation_steps: 1
200
+ learning_rate: 5.0e-05
201
+ lr_scheduler_type: cosine
202
+ weight_decay: 1.0e-10
203
+ warmup_ratio: 0.0
204
+ warmup_steps: 500
205
+ max_grad_norm: 1.0
206
+ optim: adamw_torch_fused
207
+ start_from_checkpoint: /code/ironflow-pico-runtime-v1/weights/GR00T-N1.7-3B
208
+ skip_weight_loading: false
209
+ tf32: true
210
+ fp16: false
211
+ bf16: true
212
+ eval_bf16: true
213
+ logging_steps: 20
214
+ save_steps: 2000
215
+ save_total_limit: 2
216
+ save_vl_model: false
217
+ save_only_model: false
218
+ resume_from_checkpoint: false
219
+ upload_checkpoints: false
220
+ upload_every: 1000
221
+ upload_last_n_checkpoints: 5
222
+ max_concurrent_uploads: 2
223
+ eval_strategy: 'no'
224
+ eval_steps: 500
225
+ eval_set_split_ratio: 0.1
226
+ eval_batch_size: 2
227
+ save_best_eval_metric_name: ''
228
+ save_best_eval_metric_greater_is_better: true
229
+ deepspeed_stage: 2
230
+ gradient_checkpointing: false
231
+ transformers_trust_remote_code: true
232
+ transformers_local_files_only: true
233
+ transformers_cache_dir: /code/stack-cups-148-gr00t-bifrost-2026110617482401-shouqx/hf
234
+ transformers_access_token: null
235
+ use_ddp: false
236
+ ddp_bucket_cap_mb: 100
237
+ num_gpus: 8
238
+ dataloader_num_workers: 4
239
+ remove_unused_columns: false
240
+ use_wandb: true
241
+ wandb_project: finetune-gr00t-n1d7
242
+ enable_profiling: false
243
+ max_retries: 3
244
+ assert_loss_less_than: null
245
+ add_rl_callback: false
246
+ enable_open_loop_eval: false
247
+ open_loop_eval_traj_ids:
248
+ - 0
249
+ open_loop_eval_steps_per_traj: 100
250
+ open_loop_eval_plot_indices: null
baseline/gr00t_stack_cups_148/checkpoint-21100/experiment_cfg/dataset_statistics.json ADDED
@@ -0,0 +1,288 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "new_embodiment": {
3
+ "state": {
4
+ "left_arm": {
5
+ "mean": [
6
+ -0.1625075198268781,
7
+ 1.0842620647445143,
8
+ -0.36731704502858703,
9
+ -0.41283875757757366,
10
+ 0.03702329523675885,
11
+ -0.030393585719784302
12
+ ],
13
+ "std": [
14
+ 0.2700353873323502,
15
+ 1.1467528567093037,
16
+ 0.5530823686632277,
17
+ 0.5927237779599276,
18
+ 0.12713039475675825,
19
+ 0.06632369407822619
20
+ ],
21
+ "q01": [
22
+ -0.8736169934272766,
23
+ 0.2148936241865158,
24
+ -1.9360382914543153,
25
+ -1.5257446765899658,
26
+ -0.4946638309955597,
27
+ -0.2444680780172348
28
+ ],
29
+ "q99": [
30
+ 0.16063830256462097,
31
+ 3.034255266189575,
32
+ -0.0019148936262354255,
33
+ 0.0642382982373234,
34
+ 0.16808511316776276,
35
+ 0.11829787492752075
36
+ ],
37
+ "min": [
38
+ -1.091489315032959,
39
+ 0.2148936241865158,
40
+ -2.018510580062866,
41
+ -1.564042568206787,
42
+ -1.1272341012954712,
43
+ -0.4861702024936676
44
+ ],
45
+ "max": [
46
+ 0.5544680953025818,
47
+ 3.138085126876831,
48
+ 0.0,
49
+ 1.542553186416626,
50
+ 1.240638256072998,
51
+ 0.5393617153167725
52
+ ]
53
+ },
54
+ "left_gripper": {
55
+ "mean": [
56
+ 88.70043343606748
57
+ ],
58
+ "std": [
59
+ 24.374810634433103
60
+ ],
61
+ "q01": [
62
+ 12.464552879333496
63
+ ],
64
+ "q99": [
65
+ 102.18600463867188
66
+ ],
67
+ "min": [
68
+ 3.08048677444458
69
+ ],
70
+ "max": [
71
+ 102.97866821289062
72
+ ]
73
+ },
74
+ "right_arm": {
75
+ "mean": [
76
+ 0.1001759839971589,
77
+ 1.1130726768189518,
78
+ -0.319068614756271,
79
+ -0.4901851291457013,
80
+ 0.04893345683788824,
81
+ 0.04471603786466137
82
+ ],
83
+ "std": [
84
+ 0.13874521422404704,
85
+ 1.068493942298225,
86
+ 0.43438903122681344,
87
+ 0.6259246013658152,
88
+ 0.08044048709309755,
89
+ 0.031995482570700895
90
+ ],
91
+ "q01": [
92
+ -0.06808510422706604,
93
+ 0.21553191542625427,
94
+ -1.313191533088684,
95
+ -1.5253190994262695,
96
+ -0.11170212924480438,
97
+ -0.05063829943537712
98
+ ],
99
+ "q99": [
100
+ 0.4502127766609192,
101
+ 2.8959574699401855,
102
+ -0.0008510638144798577,
103
+ 0.034255318343639374,
104
+ 0.2665957510471344,
105
+ 0.14464681506156848
106
+ ],
107
+ "min": [
108
+ -0.2080851048231125,
109
+ 0.21553191542625427,
110
+ -1.7904255390167236,
111
+ -1.5502127408981323,
112
+ -0.2957446873188019,
113
+ -0.1517021209001541
114
+ ],
115
+ "max": [
116
+ 0.6497872471809387,
117
+ 3.0423405170440674,
118
+ 0.0,
119
+ 0.43744680285453796,
120
+ 0.3906382918357849,
121
+ 0.23680850863456726
122
+ ]
123
+ },
124
+ "right_gripper": {
125
+ "mean": [
126
+ 85.58191268910615
127
+ ],
128
+ "std": [
129
+ 26.794849672554083
130
+ ],
131
+ "q01": [
132
+ 12.032872200012207
133
+ ],
134
+ "q99": [
135
+ 102.34491729736328
136
+ ],
137
+ "min": [
138
+ 11.066163063049316
139
+ ],
140
+ "max": [
141
+ 103.48764038085938
142
+ ]
143
+ }
144
+ },
145
+ "action": {
146
+ "left_arm": {
147
+ "mean": [
148
+ -0.16473093956759127,
149
+ 0.8681155832258038,
150
+ -0.35903689579197906,
151
+ -0.422772990293642,
152
+ 0.013333487666197312,
153
+ -0.053439910524043906
154
+ ],
155
+ "std": [
156
+ 0.26413852784851743,
157
+ 1.176010024006326,
158
+ 0.5652555232725361,
159
+ 0.6052913922882582,
160
+ 0.13028488266679109,
161
+ 0.06823171888650201
162
+ ],
163
+ "q01": [
164
+ -0.8682331442832947,
165
+ 0.0,
166
+ -1.9711652994155884,
167
+ -1.5707999467849731,
168
+ -0.5215534567832947,
169
+ -0.2730485796928406
170
+ ],
171
+ "q99": [
172
+ 0.14879614114761353,
173
+ 2.8746800422668457,
174
+ 0.0,
175
+ 0.05062136426568031,
176
+ 0.14879614114761353,
177
+ 0.10431069135665894
178
+ ],
179
+ "min": [
180
+ -1.2593982219696045,
181
+ -0.0,
182
+ -2.0,
183
+ -1.5707999467849731,
184
+ -1.5033011436462402,
185
+ -0.6504078507423401
186
+ ],
187
+ "max": [
188
+ 0.575242817401886,
189
+ 3.011204242706299,
190
+ 0.0,
191
+ 1.5707999467849731,
192
+ 1.2716701030731201,
193
+ 0.5384272336959839
194
+ ]
195
+ },
196
+ "left_gripper": {
197
+ "mean": [
198
+ 86.547770940583
199
+ ],
200
+ "std": [
201
+ 33.50286769923162
202
+ ],
203
+ "q01": [
204
+ 0.0
205
+ ],
206
+ "q99": [
207
+ 101.44685363769531
208
+ ],
209
+ "min": [
210
+ 0.0
211
+ ],
212
+ "max": [
213
+ 102.18600463867188
214
+ ]
215
+ },
216
+ "right_arm": {
217
+ "mean": [
218
+ 0.08334164575863477,
219
+ 0.9183426463724823,
220
+ -0.33061070522100666,
221
+ -0.5209033568473217,
222
+ 0.024604943696559547,
223
+ 0.01963253503542509
224
+ ],
225
+ "std": [
226
+ 0.13674243725761884,
227
+ 1.099167413108321,
228
+ 0.4524649041976627,
229
+ 0.6366756189128563,
230
+ 0.08196322531802316,
231
+ 0.032833390548008885
232
+ ],
233
+ "q01": [
234
+ -0.09970875084400177,
235
+ 0.0,
236
+ -1.368310809135437,
237
+ -1.5707999467849731,
238
+ -0.13805827498435974,
239
+ -0.07669904083013535
240
+ ],
241
+ "q99": [
242
+ 0.4264466464519501,
243
+ 2.728951930999756,
244
+ 0.0,
245
+ 0.019941750913858414,
246
+ 0.251572847366333,
247
+ 0.12425244599580765
248
+ ],
249
+ "min": [
250
+ -0.24697090685367584,
251
+ -0.0,
252
+ -2.0,
253
+ -1.5707999467849731,
254
+ -0.4586602449417114,
255
+ -0.25003886222839355
256
+ ],
257
+ "max": [
258
+ 0.6473398804664612,
259
+ 2.8746800422668457,
260
+ 0.0,
261
+ 0.9111846089363098,
262
+ 0.38656315207481384,
263
+ 0.2147573083639145
264
+ ]
265
+ },
266
+ "right_gripper": {
267
+ "mean": [
268
+ 79.80457299498252
269
+ ],
270
+ "std": [
271
+ 39.14279967697801
272
+ ],
273
+ "q01": [
274
+ 0.0
275
+ ],
276
+ "q99": [
277
+ 102.08477783203125
278
+ ],
279
+ "min": [
280
+ 0.0
281
+ ],
282
+ "max": [
283
+ 102.34491729736328
284
+ ]
285
+ }
286
+ }
287
+ }
288
+ }
baseline/gr00t_stack_cups_148/checkpoint-21100/experiment_cfg/final_model_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_type": "Gr00tN1d7",
3
+ "model_dtype": "bfloat16",
4
+ "model_name": "/code/ironflow-pico-runtime-v1/weights/Cosmos-Reason2-2B",
5
+ "backbone_model_type": "qwen",
6
+ "model_revision": null,
7
+ "tune_top_llm_layers": 0,
8
+ "backbone_embedding_dim": 2048,
9
+ "tune_llm": false,
10
+ "tune_visual": true,
11
+ "select_layer": 16,
12
+ "reproject_vision": false,
13
+ "use_flash_attention": true,
14
+ "load_bf16": false,
15
+ "backbone_trainable_params_fp32": true,
16
+ "letter_box_transform": false,
17
+ "extra_augmentation_config": null,
18
+ "apply_sincos_state_encoding": false,
19
+ "use_percentiles": true,
20
+ "use_relative_action": false,
21
+ "max_state_dim": 132,
22
+ "max_action_dim": 132,
23
+ "action_horizon": 40,
24
+ "hidden_size": 1024,
25
+ "input_embedding_dim": 1536,
26
+ "state_history_length": 1,
27
+ "add_pos_embed": true,
28
+ "attn_dropout": 0.2,
29
+ "use_vlln": true,
30
+ "max_seq_len": 1024,
31
+ "use_alternate_vl_dit": true,
32
+ "attend_text_every_n_blocks": 2,
33
+ "diffusion_model_cfg": {
34
+ "attention_head_dim": 48,
35
+ "dropout": 0.2,
36
+ "final_dropout": true,
37
+ "interleave_self_attention": true,
38
+ "norm_type": "ada_norm",
39
+ "num_attention_heads": 32,
40
+ "num_layers": 32,
41
+ "output_dim": 1024,
42
+ "positional_embeddings": null
43
+ },
44
+ "num_inference_timesteps": 4,
45
+ "noise_beta_alpha": 1.5,
46
+ "noise_beta_beta": 1.0,
47
+ "noise_s": 0.999,
48
+ "num_timestep_buckets": 1000,
49
+ "tune_projector": true,
50
+ "tune_diffusion_model": true,
51
+ "tune_vlln": true,
52
+ "state_dropout_prob": 0.0,
53
+ "exclude_state": false,
54
+ "use_mean_std": false,
55
+ "max_num_embodiments": 32
56
+ }
baseline/gr00t_stack_cups_148/checkpoint-21100/experiment_cfg/final_processor_config.json ADDED
The diff for this file is too large to render. See raw diff
 
baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5b2a2c4cffec3ae3991d6274a15e78b7952b499b0fa489da2ee9a8e0a6008da
3
+ size 3041236165
baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb8fccd7b7d4f5af384d59b9a1af24521c4ce7cd9745ce6eb8db1a0c70fed841
3
+ size 3041225157
baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e9839009b02b34dd7677f1ea056cbbb4fdbb9c4a35140bbdcbf81638e0855e0
3
+ size 3041223941
baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a8c2edf2f1ad38063e5160167d1dff1b228a9696099bcac56637accccaf8afa
3
+ size 3041223877
baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b8a808e771ee5820faf3b48a5b2a90bbf0c12fe8e68ba1d64871fe00e73238b
3
+ size 3041224069
baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92fb3791e263da034ea8aaf5f5c3a3fc330d3eb060d1d81e2b4a3e119c749f86
3
+ size 3041223109
baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:097145ed1d18ed72c14003f6890ec728486894ae82cd00f086a61474a9fdc661
3
+ size 3041214405
baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:683215a0eaa901719b76eb940067dcb0fef2faf2782c9844293e50b664738bfe
3
+ size 3041221061
baseline/gr00t_stack_cups_148/checkpoint-21100/global_step21100/mp_rank_00_model_states.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aab77284037440d98a7c1a79e6057fda9e34432e033e0fa2580cc66b8f69a6bf
3
+ size 8521564777
baseline/gr00t_stack_cups_148/checkpoint-21100/latest ADDED
@@ -0,0 +1 @@
 
 
1
+ global_step21100
baseline/gr00t_stack_cups_148/checkpoint-21100/model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71789b3b904446ebf11712fa0f979ecd22495a7aea31e9183c6f0d410f3b3020
3
+ size 4990519232
baseline/gr00t_stack_cups_148/checkpoint-21100/model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e39ce9aff702f34e28b9b312b3f23e8578dbe5f11374bbec27c48756adbf3d28
3
+ size 1919980184
baseline/gr00t_stack_cups_148/checkpoint-21100/model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
baseline/gr00t_stack_cups_148/checkpoint-21100/processor_config.json ADDED
@@ -0,0 +1,1164 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "processor_class": "Gr00tN1d7Processor",
3
+ "processor_kwargs": {
4
+ "modality_configs": {
5
+ "real_g1_relative_eef_relative_joints": {
6
+ "video": {
7
+ "delta_indices": [
8
+ -20,
9
+ 0
10
+ ],
11
+ "modality_keys": [
12
+ "ego_view"
13
+ ],
14
+ "sin_cos_embedding_keys": null,
15
+ "mean_std_embedding_keys": null,
16
+ "action_configs": null
17
+ },
18
+ "state": {
19
+ "delta_indices": [
20
+ 0
21
+ ],
22
+ "modality_keys": [
23
+ "left_wrist_eef_9d",
24
+ "right_wrist_eef_9d",
25
+ "left_hand",
26
+ "right_hand",
27
+ "left_arm",
28
+ "right_arm",
29
+ "waist"
30
+ ],
31
+ "sin_cos_embedding_keys": null,
32
+ "mean_std_embedding_keys": null,
33
+ "action_configs": null
34
+ },
35
+ "action": {
36
+ "delta_indices": [
37
+ 0,
38
+ 1,
39
+ 2,
40
+ 3,
41
+ 4,
42
+ 5,
43
+ 6,
44
+ 7,
45
+ 8,
46
+ 9,
47
+ 10,
48
+ 11,
49
+ 12,
50
+ 13,
51
+ 14,
52
+ 15,
53
+ 16,
54
+ 17,
55
+ 18,
56
+ 19,
57
+ 20,
58
+ 21,
59
+ 22,
60
+ 23,
61
+ 24,
62
+ 25,
63
+ 26,
64
+ 27,
65
+ 28,
66
+ 29,
67
+ 30,
68
+ 31,
69
+ 32,
70
+ 33,
71
+ 34,
72
+ 35,
73
+ 36,
74
+ 37,
75
+ 38,
76
+ 39
77
+ ],
78
+ "modality_keys": [
79
+ "left_wrist_eef_9d",
80
+ "right_wrist_eef_9d",
81
+ "left_hand",
82
+ "right_hand",
83
+ "left_arm",
84
+ "right_arm",
85
+ "waist",
86
+ "base_height_command",
87
+ "navigate_command"
88
+ ],
89
+ "sin_cos_embedding_keys": null,
90
+ "mean_std_embedding_keys": null,
91
+ "action_configs": [
92
+ {
93
+ "rep": "RELATIVE",
94
+ "type": "EEF",
95
+ "format": "XYZ_ROT6D",
96
+ "state_key": "left_wrist_eef_9d"
97
+ },
98
+ {
99
+ "rep": "RELATIVE",
100
+ "type": "EEF",
101
+ "format": "XYZ_ROT6D",
102
+ "state_key": "right_wrist_eef_9d"
103
+ },
104
+ {
105
+ "rep": "ABSOLUTE",
106
+ "type": "NON_EEF",
107
+ "format": "DEFAULT",
108
+ "state_key": "left_hand"
109
+ },
110
+ {
111
+ "rep": "ABSOLUTE",
112
+ "type": "NON_EEF",
113
+ "format": "DEFAULT",
114
+ "state_key": "right_hand"
115
+ },
116
+ {
117
+ "rep": "RELATIVE",
118
+ "type": "NON_EEF",
119
+ "format": "DEFAULT",
120
+ "state_key": "left_arm"
121
+ },
122
+ {
123
+ "rep": "RELATIVE",
124
+ "type": "NON_EEF",
125
+ "format": "DEFAULT",
126
+ "state_key": "right_arm"
127
+ },
128
+ {
129
+ "rep": "ABSOLUTE",
130
+ "type": "NON_EEF",
131
+ "format": "DEFAULT",
132
+ "state_key": "waist"
133
+ },
134
+ {
135
+ "rep": "ABSOLUTE",
136
+ "type": "NON_EEF",
137
+ "format": "DEFAULT",
138
+ "state_key": "base_height_command"
139
+ },
140
+ {
141
+ "rep": "ABSOLUTE",
142
+ "type": "NON_EEF",
143
+ "format": "DEFAULT",
144
+ "state_key": "navigate_command"
145
+ }
146
+ ]
147
+ },
148
+ "language": {
149
+ "delta_indices": [
150
+ 0
151
+ ],
152
+ "modality_keys": [
153
+ "annotation.human.task_description"
154
+ ],
155
+ "sin_cos_embedding_keys": null,
156
+ "mean_std_embedding_keys": null,
157
+ "action_configs": null
158
+ }
159
+ },
160
+ "real_r1_pro_sharpa_relative_eef_mecka": {
161
+ "video": {
162
+ "delta_indices": [
163
+ -30,
164
+ 0
165
+ ],
166
+ "modality_keys": [
167
+ "ego_view_cropratio_res320x240_freq30"
168
+ ],
169
+ "sin_cos_embedding_keys": null,
170
+ "mean_std_embedding_keys": null,
171
+ "action_configs": null
172
+ },
173
+ "state": {
174
+ "delta_indices": [
175
+ 0
176
+ ],
177
+ "modality_keys": [
178
+ "left_wrist_eef",
179
+ "right_wrist_eef",
180
+ "left_hand_joints",
181
+ "right_hand_joints"
182
+ ],
183
+ "sin_cos_embedding_keys": null,
184
+ "mean_std_embedding_keys": null,
185
+ "action_configs": null
186
+ },
187
+ "action": {
188
+ "delta_indices": [
189
+ 0,
190
+ 1,
191
+ 2,
192
+ 3,
193
+ 4,
194
+ 5,
195
+ 6,
196
+ 7,
197
+ 8,
198
+ 9,
199
+ 10,
200
+ 11,
201
+ 12,
202
+ 13,
203
+ 14,
204
+ 15,
205
+ 16,
206
+ 17,
207
+ 18,
208
+ 19,
209
+ 20,
210
+ 21,
211
+ 22,
212
+ 23,
213
+ 24,
214
+ 25,
215
+ 26,
216
+ 27,
217
+ 28,
218
+ 29,
219
+ 30,
220
+ 31,
221
+ 32,
222
+ 33,
223
+ 34,
224
+ 35,
225
+ 36,
226
+ 37,
227
+ 38,
228
+ 39
229
+ ],
230
+ "modality_keys": [
231
+ "left_wrist_eef",
232
+ "right_wrist_eef",
233
+ "left_hand_joints",
234
+ "right_hand_joints"
235
+ ],
236
+ "sin_cos_embedding_keys": null,
237
+ "mean_std_embedding_keys": null,
238
+ "action_configs": [
239
+ {
240
+ "rep": "RELATIVE",
241
+ "type": "EEF",
242
+ "format": "XYZ_ROT6D",
243
+ "state_key": "left_wrist_eef"
244
+ },
245
+ {
246
+ "rep": "RELATIVE",
247
+ "type": "EEF",
248
+ "format": "XYZ_ROT6D",
249
+ "state_key": "right_wrist_eef"
250
+ },
251
+ {
252
+ "rep": "ABSOLUTE",
253
+ "type": "NON_EEF",
254
+ "format": "DEFAULT",
255
+ "state_key": "left_hand_joints"
256
+ },
257
+ {
258
+ "rep": "ABSOLUTE",
259
+ "type": "NON_EEF",
260
+ "format": "DEFAULT",
261
+ "state_key": "right_hand_joints"
262
+ }
263
+ ]
264
+ },
265
+ "language": {
266
+ "delta_indices": [
267
+ 0
268
+ ],
269
+ "modality_keys": [
270
+ "annotation.human.coarse_action"
271
+ ],
272
+ "sin_cos_embedding_keys": null,
273
+ "mean_std_embedding_keys": null,
274
+ "action_configs": null
275
+ }
276
+ },
277
+ "real_r1_pro_sharpa_relative_eef_human": {
278
+ "video": {
279
+ "delta_indices": [
280
+ -20,
281
+ 0
282
+ ],
283
+ "modality_keys": [
284
+ "ego_view_res320x240_freq20",
285
+ "left_wrist_view_res320x240_freq20",
286
+ "right_wrist_view_res320x240_freq20"
287
+ ],
288
+ "sin_cos_embedding_keys": null,
289
+ "mean_std_embedding_keys": null,
290
+ "action_configs": null
291
+ },
292
+ "state": {
293
+ "delta_indices": [
294
+ 0
295
+ ],
296
+ "modality_keys": [
297
+ "left_wrist_eef",
298
+ "right_wrist_eef",
299
+ "left_hand_joints",
300
+ "right_hand_joints"
301
+ ],
302
+ "sin_cos_embedding_keys": null,
303
+ "mean_std_embedding_keys": null,
304
+ "action_configs": null
305
+ },
306
+ "action": {
307
+ "delta_indices": [
308
+ 0,
309
+ 1,
310
+ 2,
311
+ 3,
312
+ 4,
313
+ 5,
314
+ 6,
315
+ 7,
316
+ 8,
317
+ 9,
318
+ 10,
319
+ 11,
320
+ 12,
321
+ 13,
322
+ 14,
323
+ 15,
324
+ 16,
325
+ 17,
326
+ 18,
327
+ 19,
328
+ 20,
329
+ 21,
330
+ 22,
331
+ 23,
332
+ 24,
333
+ 25,
334
+ 26,
335
+ 27,
336
+ 28,
337
+ 29,
338
+ 30,
339
+ 31,
340
+ 32,
341
+ 33,
342
+ 34,
343
+ 35,
344
+ 36,
345
+ 37,
346
+ 38,
347
+ 39
348
+ ],
349
+ "modality_keys": [
350
+ "left_wrist_eef",
351
+ "right_wrist_eef",
352
+ "left_hand_joints",
353
+ "right_hand_joints"
354
+ ],
355
+ "sin_cos_embedding_keys": null,
356
+ "mean_std_embedding_keys": null,
357
+ "action_configs": [
358
+ {
359
+ "rep": "RELATIVE",
360
+ "type": "EEF",
361
+ "format": "XYZ_ROT6D",
362
+ "state_key": "left_wrist_eef"
363
+ },
364
+ {
365
+ "rep": "RELATIVE",
366
+ "type": "EEF",
367
+ "format": "XYZ_ROT6D",
368
+ "state_key": "right_wrist_eef"
369
+ },
370
+ {
371
+ "rep": "ABSOLUTE",
372
+ "type": "NON_EEF",
373
+ "format": "DEFAULT",
374
+ "state_key": "left_hand_joints"
375
+ },
376
+ {
377
+ "rep": "ABSOLUTE",
378
+ "type": "NON_EEF",
379
+ "format": "DEFAULT",
380
+ "state_key": "right_hand_joints"
381
+ }
382
+ ]
383
+ },
384
+ "language": {
385
+ "delta_indices": [
386
+ 0
387
+ ],
388
+ "modality_keys": [
389
+ "annotation.human.coarse_action"
390
+ ],
391
+ "sin_cos_embedding_keys": null,
392
+ "mean_std_embedding_keys": null,
393
+ "action_configs": null
394
+ }
395
+ },
396
+ "real_r1_pro_sharpa_relative_eef": {
397
+ "video": {
398
+ "delta_indices": [
399
+ -20,
400
+ 0
401
+ ],
402
+ "modality_keys": [
403
+ "ego_view_res320x240_freq20",
404
+ "left_wrist_view_res320x240_freq20",
405
+ "right_wrist_view_res320x240_freq20"
406
+ ],
407
+ "sin_cos_embedding_keys": null,
408
+ "mean_std_embedding_keys": null,
409
+ "action_configs": null
410
+ },
411
+ "state": {
412
+ "delta_indices": [
413
+ 0
414
+ ],
415
+ "modality_keys": [
416
+ "left_wrist_eef",
417
+ "right_wrist_eef",
418
+ "left_hand_joints",
419
+ "right_hand_joints"
420
+ ],
421
+ "sin_cos_embedding_keys": null,
422
+ "mean_std_embedding_keys": null,
423
+ "action_configs": null
424
+ },
425
+ "action": {
426
+ "delta_indices": [
427
+ 0,
428
+ 1,
429
+ 2,
430
+ 3,
431
+ 4,
432
+ 5,
433
+ 6,
434
+ 7,
435
+ 8,
436
+ 9,
437
+ 10,
438
+ 11,
439
+ 12,
440
+ 13,
441
+ 14,
442
+ 15,
443
+ 16,
444
+ 17,
445
+ 18,
446
+ 19,
447
+ 20,
448
+ 21,
449
+ 22,
450
+ 23,
451
+ 24,
452
+ 25,
453
+ 26,
454
+ 27,
455
+ 28,
456
+ 29,
457
+ 30,
458
+ 31,
459
+ 32,
460
+ 33,
461
+ 34,
462
+ 35,
463
+ 36,
464
+ 37,
465
+ 38,
466
+ 39
467
+ ],
468
+ "modality_keys": [
469
+ "left_wrist_eef",
470
+ "right_wrist_eef",
471
+ "left_hand_joints",
472
+ "right_hand_joints"
473
+ ],
474
+ "sin_cos_embedding_keys": null,
475
+ "mean_std_embedding_keys": null,
476
+ "action_configs": [
477
+ {
478
+ "rep": "RELATIVE",
479
+ "type": "EEF",
480
+ "format": "XYZ_ROT6D",
481
+ "state_key": "left_wrist_eef"
482
+ },
483
+ {
484
+ "rep": "RELATIVE",
485
+ "type": "EEF",
486
+ "format": "XYZ_ROT6D",
487
+ "state_key": "right_wrist_eef"
488
+ },
489
+ {
490
+ "rep": "ABSOLUTE",
491
+ "type": "NON_EEF",
492
+ "format": "DEFAULT",
493
+ "state_key": "left_hand_joints"
494
+ },
495
+ {
496
+ "rep": "ABSOLUTE",
497
+ "type": "NON_EEF",
498
+ "format": "DEFAULT",
499
+ "state_key": "right_hand_joints"
500
+ }
501
+ ]
502
+ },
503
+ "language": {
504
+ "delta_indices": [
505
+ 0
506
+ ],
507
+ "modality_keys": [
508
+ "annotation.human.coarse_action"
509
+ ],
510
+ "sin_cos_embedding_keys": null,
511
+ "mean_std_embedding_keys": null,
512
+ "action_configs": null
513
+ }
514
+ },
515
+ "xdof_relative_eef_relative_joint": {
516
+ "video": {
517
+ "delta_indices": [
518
+ -30,
519
+ 0
520
+ ],
521
+ "modality_keys": [
522
+ "top_camera-images-rgb_320_240",
523
+ "left_camera-images-rgb_320_240",
524
+ "right_camera-images-rgb_320_240"
525
+ ],
526
+ "sin_cos_embedding_keys": null,
527
+ "mean_std_embedding_keys": null,
528
+ "action_configs": null
529
+ },
530
+ "state": {
531
+ "delta_indices": [
532
+ 0
533
+ ],
534
+ "modality_keys": [
535
+ "left_wrist_eef",
536
+ "right_wrist_eef",
537
+ "left_gripper_pos",
538
+ "right_gripper_pos",
539
+ "left_joint_pos",
540
+ "right_joint_pos"
541
+ ],
542
+ "sin_cos_embedding_keys": null,
543
+ "mean_std_embedding_keys": null,
544
+ "action_configs": null
545
+ },
546
+ "action": {
547
+ "delta_indices": [
548
+ 0,
549
+ 1,
550
+ 2,
551
+ 3,
552
+ 4,
553
+ 5,
554
+ 6,
555
+ 7,
556
+ 8,
557
+ 9,
558
+ 10,
559
+ 11,
560
+ 12,
561
+ 13,
562
+ 14,
563
+ 15,
564
+ 16,
565
+ 17,
566
+ 18,
567
+ 19,
568
+ 20,
569
+ 21,
570
+ 22,
571
+ 23,
572
+ 24,
573
+ 25,
574
+ 26,
575
+ 27,
576
+ 28,
577
+ 29,
578
+ 30,
579
+ 31,
580
+ 32,
581
+ 33,
582
+ 34,
583
+ 35,
584
+ 36,
585
+ 37,
586
+ 38,
587
+ 39
588
+ ],
589
+ "modality_keys": [
590
+ "left_wrist_eef",
591
+ "right_wrist_eef",
592
+ "left_gripper_pos",
593
+ "right_gripper_pos",
594
+ "left_joint_pos",
595
+ "right_joint_pos"
596
+ ],
597
+ "sin_cos_embedding_keys": null,
598
+ "mean_std_embedding_keys": null,
599
+ "action_configs": [
600
+ {
601
+ "rep": "RELATIVE",
602
+ "type": "EEF",
603
+ "format": "XYZ_ROT6D",
604
+ "state_key": "left_wrist_eef"
605
+ },
606
+ {
607
+ "rep": "RELATIVE",
608
+ "type": "EEF",
609
+ "format": "XYZ_ROT6D",
610
+ "state_key": "right_wrist_eef"
611
+ },
612
+ {
613
+ "rep": "ABSOLUTE",
614
+ "type": "NON_EEF",
615
+ "format": "DEFAULT",
616
+ "state_key": "left_gripper_pos"
617
+ },
618
+ {
619
+ "rep": "ABSOLUTE",
620
+ "type": "NON_EEF",
621
+ "format": "DEFAULT",
622
+ "state_key": "right_gripper_pos"
623
+ },
624
+ {
625
+ "rep": "RELATIVE",
626
+ "type": "NON_EEF",
627
+ "format": "DEFAULT",
628
+ "state_key": "left_joint_pos"
629
+ },
630
+ {
631
+ "rep": "RELATIVE",
632
+ "type": "NON_EEF",
633
+ "format": "DEFAULT",
634
+ "state_key": "right_joint_pos"
635
+ }
636
+ ]
637
+ },
638
+ "language": {
639
+ "delta_indices": [
640
+ 0
641
+ ],
642
+ "modality_keys": [
643
+ "annotation.task"
644
+ ],
645
+ "sin_cos_embedding_keys": null,
646
+ "mean_std_embedding_keys": null,
647
+ "action_configs": null
648
+ }
649
+ },
650
+ "real_r1_pro_sharpa_relative_eef_maxinsights": {
651
+ "video": {
652
+ "delta_indices": [
653
+ -30,
654
+ 0
655
+ ],
656
+ "modality_keys": [
657
+ "ego_view_cropratio_res320x240_freq30"
658
+ ],
659
+ "sin_cos_embedding_keys": null,
660
+ "mean_std_embedding_keys": null,
661
+ "action_configs": null
662
+ },
663
+ "state": {
664
+ "delta_indices": [
665
+ 0
666
+ ],
667
+ "modality_keys": [
668
+ "left_wrist_eef",
669
+ "right_wrist_eef",
670
+ "left_hand_joints",
671
+ "right_hand_joints"
672
+ ],
673
+ "sin_cos_embedding_keys": null,
674
+ "mean_std_embedding_keys": null,
675
+ "action_configs": null
676
+ },
677
+ "action": {
678
+ "delta_indices": [
679
+ 0,
680
+ 1,
681
+ 2,
682
+ 3,
683
+ 4,
684
+ 5,
685
+ 6,
686
+ 7,
687
+ 8,
688
+ 9,
689
+ 10,
690
+ 11,
691
+ 12,
692
+ 13,
693
+ 14,
694
+ 15,
695
+ 16,
696
+ 17,
697
+ 18,
698
+ 19,
699
+ 20,
700
+ 21,
701
+ 22,
702
+ 23,
703
+ 24,
704
+ 25,
705
+ 26,
706
+ 27,
707
+ 28,
708
+ 29,
709
+ 30,
710
+ 31,
711
+ 32,
712
+ 33,
713
+ 34,
714
+ 35,
715
+ 36,
716
+ 37,
717
+ 38,
718
+ 39
719
+ ],
720
+ "modality_keys": [
721
+ "left_wrist_eef",
722
+ "right_wrist_eef",
723
+ "left_hand_joints",
724
+ "right_hand_joints"
725
+ ],
726
+ "sin_cos_embedding_keys": null,
727
+ "mean_std_embedding_keys": null,
728
+ "action_configs": [
729
+ {
730
+ "rep": "RELATIVE",
731
+ "type": "EEF",
732
+ "format": "XYZ_ROT6D",
733
+ "state_key": "left_wrist_eef"
734
+ },
735
+ {
736
+ "rep": "RELATIVE",
737
+ "type": "EEF",
738
+ "format": "XYZ_ROT6D",
739
+ "state_key": "right_wrist_eef"
740
+ },
741
+ {
742
+ "rep": "ABSOLUTE",
743
+ "type": "NON_EEF",
744
+ "format": "DEFAULT",
745
+ "state_key": "left_hand_joints"
746
+ },
747
+ {
748
+ "rep": "ABSOLUTE",
749
+ "type": "NON_EEF",
750
+ "format": "DEFAULT",
751
+ "state_key": "right_hand_joints"
752
+ }
753
+ ]
754
+ },
755
+ "language": {
756
+ "delta_indices": [
757
+ 0
758
+ ],
759
+ "modality_keys": [
760
+ "annotation.human.coarse_action"
761
+ ],
762
+ "sin_cos_embedding_keys": null,
763
+ "mean_std_embedding_keys": null,
764
+ "action_configs": null
765
+ }
766
+ },
767
+ "xdof_relative_eef_relative_joint_subtask": {
768
+ "video": {
769
+ "delta_indices": [
770
+ -30,
771
+ 0
772
+ ],
773
+ "modality_keys": [
774
+ "top_camera-images-rgb_320_240",
775
+ "left_camera-images-rgb_320_240",
776
+ "right_camera-images-rgb_320_240"
777
+ ],
778
+ "sin_cos_embedding_keys": null,
779
+ "mean_std_embedding_keys": null,
780
+ "action_configs": null
781
+ },
782
+ "state": {
783
+ "delta_indices": [
784
+ 0
785
+ ],
786
+ "modality_keys": [
787
+ "left_wrist_eef",
788
+ "right_wrist_eef",
789
+ "left_gripper_pos",
790
+ "right_gripper_pos",
791
+ "left_joint_pos",
792
+ "right_joint_pos"
793
+ ],
794
+ "sin_cos_embedding_keys": null,
795
+ "mean_std_embedding_keys": null,
796
+ "action_configs": null
797
+ },
798
+ "action": {
799
+ "delta_indices": [
800
+ 0,
801
+ 1,
802
+ 2,
803
+ 3,
804
+ 4,
805
+ 5,
806
+ 6,
807
+ 7,
808
+ 8,
809
+ 9,
810
+ 10,
811
+ 11,
812
+ 12,
813
+ 13,
814
+ 14,
815
+ 15,
816
+ 16,
817
+ 17,
818
+ 18,
819
+ 19,
820
+ 20,
821
+ 21,
822
+ 22,
823
+ 23,
824
+ 24,
825
+ 25,
826
+ 26,
827
+ 27,
828
+ 28,
829
+ 29,
830
+ 30,
831
+ 31,
832
+ 32,
833
+ 33,
834
+ 34,
835
+ 35,
836
+ 36,
837
+ 37,
838
+ 38,
839
+ 39
840
+ ],
841
+ "modality_keys": [
842
+ "left_wrist_eef",
843
+ "right_wrist_eef",
844
+ "left_gripper_pos",
845
+ "right_gripper_pos",
846
+ "left_joint_pos",
847
+ "right_joint_pos"
848
+ ],
849
+ "sin_cos_embedding_keys": null,
850
+ "mean_std_embedding_keys": null,
851
+ "action_configs": [
852
+ {
853
+ "rep": "RELATIVE",
854
+ "type": "EEF",
855
+ "format": "XYZ_ROT6D",
856
+ "state_key": "left_wrist_eef"
857
+ },
858
+ {
859
+ "rep": "RELATIVE",
860
+ "type": "EEF",
861
+ "format": "XYZ_ROT6D",
862
+ "state_key": "right_wrist_eef"
863
+ },
864
+ {
865
+ "rep": "ABSOLUTE",
866
+ "type": "NON_EEF",
867
+ "format": "DEFAULT",
868
+ "state_key": "left_gripper_pos"
869
+ },
870
+ {
871
+ "rep": "ABSOLUTE",
872
+ "type": "NON_EEF",
873
+ "format": "DEFAULT",
874
+ "state_key": "right_gripper_pos"
875
+ },
876
+ {
877
+ "rep": "RELATIVE",
878
+ "type": "NON_EEF",
879
+ "format": "DEFAULT",
880
+ "state_key": "left_joint_pos"
881
+ },
882
+ {
883
+ "rep": "RELATIVE",
884
+ "type": "NON_EEF",
885
+ "format": "DEFAULT",
886
+ "state_key": "right_joint_pos"
887
+ }
888
+ ]
889
+ },
890
+ "language": {
891
+ "delta_indices": [
892
+ 0
893
+ ],
894
+ "modality_keys": [
895
+ "annotation.sub_task"
896
+ ],
897
+ "sin_cos_embedding_keys": null,
898
+ "mean_std_embedding_keys": null,
899
+ "action_configs": null
900
+ }
901
+ },
902
+ "oxe_droid_relative_eef_relative_joint": {
903
+ "video": {
904
+ "delta_indices": [
905
+ -15,
906
+ 0
907
+ ],
908
+ "modality_keys": [
909
+ "exterior_image_1_left",
910
+ "wrist_image_left"
911
+ ],
912
+ "sin_cos_embedding_keys": null,
913
+ "mean_std_embedding_keys": null,
914
+ "action_configs": null
915
+ },
916
+ "state": {
917
+ "delta_indices": [
918
+ 0
919
+ ],
920
+ "modality_keys": [
921
+ "eef_9d",
922
+ "gripper_position",
923
+ "joint_position"
924
+ ],
925
+ "sin_cos_embedding_keys": null,
926
+ "mean_std_embedding_keys": null,
927
+ "action_configs": null
928
+ },
929
+ "action": {
930
+ "delta_indices": [
931
+ 0,
932
+ 1,
933
+ 2,
934
+ 3,
935
+ 4,
936
+ 5,
937
+ 6,
938
+ 7,
939
+ 8,
940
+ 9,
941
+ 10,
942
+ 11,
943
+ 12,
944
+ 13,
945
+ 14,
946
+ 15,
947
+ 16,
948
+ 17,
949
+ 18,
950
+ 19,
951
+ 20,
952
+ 21,
953
+ 22,
954
+ 23,
955
+ 24,
956
+ 25,
957
+ 26,
958
+ 27,
959
+ 28,
960
+ 29,
961
+ 30,
962
+ 31,
963
+ 32,
964
+ 33,
965
+ 34,
966
+ 35,
967
+ 36,
968
+ 37,
969
+ 38,
970
+ 39
971
+ ],
972
+ "modality_keys": [
973
+ "eef_9d",
974
+ "gripper_position",
975
+ "joint_position"
976
+ ],
977
+ "sin_cos_embedding_keys": null,
978
+ "mean_std_embedding_keys": null,
979
+ "action_configs": [
980
+ {
981
+ "rep": "RELATIVE",
982
+ "type": "EEF",
983
+ "format": "XYZ_ROT6D",
984
+ "state_key": "eef_9d"
985
+ },
986
+ {
987
+ "rep": "ABSOLUTE",
988
+ "type": "NON_EEF",
989
+ "format": "DEFAULT",
990
+ "state_key": "gripper_position"
991
+ },
992
+ {
993
+ "rep": "RELATIVE",
994
+ "type": "NON_EEF",
995
+ "format": "DEFAULT",
996
+ "state_key": "joint_position"
997
+ }
998
+ ]
999
+ },
1000
+ "language": {
1001
+ "delta_indices": [
1002
+ 0
1003
+ ],
1004
+ "modality_keys": [
1005
+ "annotation.language.language_instruction"
1006
+ ],
1007
+ "sin_cos_embedding_keys": null,
1008
+ "mean_std_embedding_keys": null,
1009
+ "action_configs": null
1010
+ }
1011
+ },
1012
+ "new_embodiment": {
1013
+ "video": {
1014
+ "delta_indices": [
1015
+ 0
1016
+ ],
1017
+ "modality_keys": [
1018
+ "head_left",
1019
+ "wrist_left",
1020
+ "wrist_right"
1021
+ ],
1022
+ "sin_cos_embedding_keys": null,
1023
+ "mean_std_embedding_keys": null,
1024
+ "action_configs": null
1025
+ },
1026
+ "state": {
1027
+ "delta_indices": [
1028
+ 0
1029
+ ],
1030
+ "modality_keys": [
1031
+ "left_arm",
1032
+ "left_gripper",
1033
+ "right_arm",
1034
+ "right_gripper"
1035
+ ],
1036
+ "sin_cos_embedding_keys": null,
1037
+ "mean_std_embedding_keys": null,
1038
+ "action_configs": null
1039
+ },
1040
+ "action": {
1041
+ "delta_indices": [
1042
+ 0,
1043
+ 1,
1044
+ 2,
1045
+ 3,
1046
+ 4,
1047
+ 5,
1048
+ 6,
1049
+ 7,
1050
+ 8,
1051
+ 9,
1052
+ 10,
1053
+ 11,
1054
+ 12,
1055
+ 13,
1056
+ 14,
1057
+ 15,
1058
+ 16,
1059
+ 17,
1060
+ 18,
1061
+ 19,
1062
+ 20,
1063
+ 21,
1064
+ 22,
1065
+ 23,
1066
+ 24,
1067
+ 25,
1068
+ 26,
1069
+ 27,
1070
+ 28,
1071
+ 29,
1072
+ 30,
1073
+ 31,
1074
+ 32,
1075
+ 33,
1076
+ 34,
1077
+ 35,
1078
+ 36,
1079
+ 37,
1080
+ 38,
1081
+ 39
1082
+ ],
1083
+ "modality_keys": [
1084
+ "left_arm",
1085
+ "left_gripper",
1086
+ "right_arm",
1087
+ "right_gripper"
1088
+ ],
1089
+ "sin_cos_embedding_keys": null,
1090
+ "mean_std_embedding_keys": null,
1091
+ "action_configs": [
1092
+ {
1093
+ "rep": "ABSOLUTE",
1094
+ "type": "NON_EEF",
1095
+ "format": "DEFAULT",
1096
+ "state_key": "left_arm"
1097
+ },
1098
+ {
1099
+ "rep": "ABSOLUTE",
1100
+ "type": "NON_EEF",
1101
+ "format": "DEFAULT",
1102
+ "state_key": "left_gripper"
1103
+ },
1104
+ {
1105
+ "rep": "ABSOLUTE",
1106
+ "type": "NON_EEF",
1107
+ "format": "DEFAULT",
1108
+ "state_key": "right_arm"
1109
+ },
1110
+ {
1111
+ "rep": "ABSOLUTE",
1112
+ "type": "NON_EEF",
1113
+ "format": "DEFAULT",
1114
+ "state_key": "right_gripper"
1115
+ }
1116
+ ]
1117
+ },
1118
+ "language": {
1119
+ "delta_indices": [
1120
+ 0
1121
+ ],
1122
+ "modality_keys": [
1123
+ "annotation.human.task_description"
1124
+ ],
1125
+ "sin_cos_embedding_keys": null,
1126
+ "mean_std_embedding_keys": null,
1127
+ "action_configs": null
1128
+ }
1129
+ }
1130
+ },
1131
+ "image_crop_size": [
1132
+ 230,
1133
+ 230
1134
+ ],
1135
+ "image_target_size": [
1136
+ 256,
1137
+ 256
1138
+ ],
1139
+ "use_albumentations": true,
1140
+ "random_rotation_angle": 0,
1141
+ "color_jitter_params": {
1142
+ "brightness": 0.0,
1143
+ "contrast": 0.0,
1144
+ "saturation": 0.0,
1145
+ "hue": 0.0
1146
+ },
1147
+ "shortest_image_edge": 256,
1148
+ "crop_fraction": 0.95,
1149
+ "letter_box_transform": false,
1150
+ "model_name": "/code/ironflow-pico-runtime-v1/weights/Cosmos-Reason2-2B",
1151
+ "model_type": "qwen",
1152
+ "formalize_language": true,
1153
+ "max_state_dim": 132,
1154
+ "max_action_dim": 132,
1155
+ "max_action_horizon": 40,
1156
+ "use_percentiles": true,
1157
+ "use_mean_std": false,
1158
+ "clip_outliers": true,
1159
+ "apply_sincos_state_encoding": false,
1160
+ "use_relative_action": false,
1161
+ "exclude_state": false,
1162
+ "state_dropout_prob": 0.0
1163
+ }
1164
+ }
baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34f7c22cde7f1679513776d3a56693b76b6362a25d15f9a7f7edbf1ad0da8c7f
3
+ size 16325
baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2151ff4c2d3507f84bce96d29c29fe5a210030b53eb514ed9723e3505798940
3
+ size 16325
baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26856c64253080822c4206354e8996bf26cbc791a04f2f7e9ade8d6c293f2163
3
+ size 16325
baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b37819809978ef21babf985c8157eaf94a9b500c6c1bc971c338e4e387dd5ec
3
+ size 16325
baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b85d8fb887ef5a0b45586c8b52dc1919ff7dde61e51562a7c661c377e0f1dc61
3
+ size 16325
baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a75c620a65e592f7b5fd0a2a60421a2bc3432585b4fefa9a40e1a77929efd561
3
+ size 16325
baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad4e0327d9b2d593c2760bc2a03927ccf6135d7de353c3953f46957588fa805d
3
+ size 16325
baseline/gr00t_stack_cups_148/checkpoint-21100/rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b2906f2dd09565832d6221075d0a3ae55a23117020a3d73f534443ab9c268c0
3
+ size 16325
baseline/gr00t_stack_cups_148/checkpoint-21100/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b493a49f0d24b6a59c451e94df8429955744031a4a2d80f354e24dcc8f906076
3
+ size 1401
baseline/gr00t_stack_cups_148/checkpoint-21100/statistics.json ADDED
The diff for this file is too large to render. See raw diff
 
baseline/gr00t_stack_cups_148/checkpoint-21100/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
baseline/gr00t_stack_cups_148/checkpoint-21100/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc07b7bffcf221721b9dce804b4fd4ba4d91c1bfab12a2c19565c3d7aff0e6ec
3
+ size 7953
baseline/gr00t_stack_cups_148/checkpoint-21100/wandb_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"project": "finetune-gr00t-n1d7", "run_id": "checkpoints"}
baseline/gr00t_stack_cups_148/checkpoint-21100/zero_to_fp32.py ADDED
@@ -0,0 +1,760 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env python
2
+
3
+ # Copyright (c) Microsoft Corporation.
4
+ # SPDX-License-Identifier: Apache-2.0
5
+
6
+ # DeepSpeed Team
7
+
8
+ # This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
9
+ # copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
10
+ # the future. Once extracted, the weights don't require DeepSpeed and can be used in any
11
+ # application.
12
+ #
13
+ # example:
14
+ # python zero_to_fp32.py . output_dir/
15
+ # or
16
+ # python zero_to_fp32.py . output_dir/ --safe_serialization
17
+
18
+ import argparse
19
+ import torch
20
+ import glob
21
+ import math
22
+ import os
23
+ import re
24
+ import gc
25
+ import json
26
+ import numpy as np
27
+ from tqdm import tqdm
28
+ from collections import OrderedDict
29
+ from dataclasses import dataclass
30
+
31
+ # while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
32
+ # DeepSpeed data structures it has to be available in the current python environment.
33
+ from deepspeed.utils import logger
34
+ from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
35
+ FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
36
+ FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
37
+
38
+
39
+ @dataclass
40
+ class zero_model_state:
41
+ buffers: dict()
42
+ param_shapes: dict()
43
+ shared_params: list
44
+ ds_version: int
45
+ frozen_param_shapes: dict()
46
+ frozen_param_fragments: dict()
47
+
48
+
49
+ debug = 0
50
+
51
+ # load to cpu
52
+ device = torch.device('cpu')
53
+
54
+
55
+ def atoi(text):
56
+ return int(text) if text.isdigit() else text
57
+
58
+
59
+ def natural_keys(text):
60
+ '''
61
+ alist.sort(key=natural_keys) sorts in human order
62
+ http://nedbatchelder.com/blog/200712/human_sorting.html
63
+ (See Toothy's implementation in the comments)
64
+ '''
65
+ return [atoi(c) for c in re.split(r'(\d+)', text)]
66
+
67
+
68
+ def get_model_state_file(checkpoint_dir, zero_stage):
69
+ if not os.path.isdir(checkpoint_dir):
70
+ raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
71
+
72
+ # there should be only one file
73
+ if zero_stage <= 2:
74
+ file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
75
+ elif zero_stage == 3:
76
+ file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
77
+
78
+ if not os.path.exists(file):
79
+ raise FileNotFoundError(f"can't find model states file at '{file}'")
80
+
81
+ return file
82
+
83
+
84
+ def get_checkpoint_files(checkpoint_dir, glob_pattern):
85
+ # XXX: need to test that this simple glob rule works for multi-node setup too
86
+ ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
87
+
88
+ if len(ckpt_files) == 0:
89
+ raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
90
+
91
+ return ckpt_files
92
+
93
+
94
+ def get_optim_files(checkpoint_dir):
95
+ return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
96
+
97
+
98
+ def get_model_state_files(checkpoint_dir):
99
+ return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
100
+
101
+
102
+ def parse_model_states(files):
103
+ zero_model_states = []
104
+ for file in files:
105
+ state_dict = torch.load(file, map_location=device, weights_only=False)
106
+
107
+ if BUFFER_NAMES not in state_dict:
108
+ raise ValueError(f"{file} is not a model state checkpoint")
109
+ buffer_names = state_dict[BUFFER_NAMES]
110
+ if debug:
111
+ print("Found buffers:", buffer_names)
112
+
113
+ # recover just the buffers while restoring them to fp32 if they were saved in fp16
114
+ buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
115
+ param_shapes = state_dict[PARAM_SHAPES]
116
+
117
+ # collect parameters that are included in param_shapes
118
+ param_names = []
119
+ for s in param_shapes:
120
+ for name in s.keys():
121
+ param_names.append(name)
122
+
123
+ # update with frozen parameters
124
+ frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
125
+ if frozen_param_shapes is not None:
126
+ if debug:
127
+ print(f"Found frozen_param_shapes: {frozen_param_shapes}")
128
+ param_names += list(frozen_param_shapes.keys())
129
+
130
+ # handle shared params
131
+ shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
132
+
133
+ ds_version = state_dict.get(DS_VERSION, None)
134
+
135
+ frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
136
+
137
+ z_model_state = zero_model_state(buffers=buffers,
138
+ param_shapes=param_shapes,
139
+ shared_params=shared_params,
140
+ ds_version=ds_version,
141
+ frozen_param_shapes=frozen_param_shapes,
142
+ frozen_param_fragments=frozen_param_fragments)
143
+ zero_model_states.append(z_model_state)
144
+
145
+ return zero_model_states
146
+
147
+
148
+ def parse_optim_states(files, ds_checkpoint_dir):
149
+ total_files = len(files)
150
+ state_dicts = []
151
+ for f in tqdm(files, desc='Loading checkpoint shards'):
152
+ state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
153
+ # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
154
+ # and also handle the case where it was already removed by another helper script
155
+ state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
156
+ state_dicts.append(state_dict)
157
+
158
+ if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]:
159
+ raise ValueError(f"{files[0]} is not a zero checkpoint")
160
+ zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
161
+ world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
162
+
163
+ # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
164
+ # parameters can be different from data parallelism for non-expert parameters. So we can just
165
+ # use the max of the partition_count to get the dp world_size.
166
+
167
+ if type(world_size) is list:
168
+ world_size = max(world_size)
169
+
170
+ if world_size != total_files:
171
+ raise ValueError(
172
+ f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
173
+ "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
174
+ )
175
+
176
+ # the groups are named differently in each stage
177
+ if zero_stage <= 2:
178
+ fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
179
+ elif zero_stage == 3:
180
+ fp32_groups_key = FP32_FLAT_GROUPS
181
+ else:
182
+ raise ValueError(f"unknown zero stage {zero_stage}")
183
+
184
+ fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
185
+ return zero_stage, world_size, fp32_flat_groups
186
+
187
+
188
+ def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
189
+ """
190
+ Returns fp32 state_dict reconstructed from ds checkpoint
191
+
192
+ Args:
193
+ - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
194
+
195
+ """
196
+ print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
197
+
198
+ optim_files = get_optim_files(ds_checkpoint_dir)
199
+ zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
200
+ print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
201
+
202
+ model_files = get_model_state_files(ds_checkpoint_dir)
203
+
204
+ zero_model_states = parse_model_states(model_files)
205
+ print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
206
+
207
+ if zero_stage <= 2:
208
+ return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
209
+ exclude_frozen_parameters)
210
+ elif zero_stage == 3:
211
+ return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
212
+ exclude_frozen_parameters)
213
+
214
+
215
+ def _zero2_merge_frozen_params(state_dict, zero_model_states):
216
+ if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
217
+ return
218
+
219
+ frozen_param_shapes = zero_model_states[0].frozen_param_shapes
220
+ frozen_param_fragments = zero_model_states[0].frozen_param_fragments
221
+
222
+ if debug:
223
+ num_elem = sum(s.numel() for s in frozen_param_shapes.values())
224
+ print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
225
+
226
+ wanted_params = len(frozen_param_shapes)
227
+ wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
228
+ avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
229
+ print(f'Frozen params: Have {avail_numel} numels to process.')
230
+ print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
231
+
232
+ total_params = 0
233
+ total_numel = 0
234
+ for name, shape in frozen_param_shapes.items():
235
+ total_params += 1
236
+ unpartitioned_numel = shape.numel()
237
+ total_numel += unpartitioned_numel
238
+
239
+ state_dict[name] = frozen_param_fragments[name]
240
+
241
+ if debug:
242
+ print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
243
+
244
+ print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
245
+
246
+
247
+ def _has_callable(obj, fn):
248
+ attr = getattr(obj, fn, None)
249
+ return callable(attr)
250
+
251
+
252
+ def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
253
+ param_shapes = zero_model_states[0].param_shapes
254
+
255
+ # Reconstruction protocol:
256
+ #
257
+ # XXX: document this
258
+
259
+ if debug:
260
+ for i in range(world_size):
261
+ for j in range(len(fp32_flat_groups[0])):
262
+ print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
263
+
264
+ # XXX: memory usage doubles here (zero2)
265
+ num_param_groups = len(fp32_flat_groups[0])
266
+ merged_single_partition_of_fp32_groups = []
267
+ for i in range(num_param_groups):
268
+ merged_partitions = [sd[i] for sd in fp32_flat_groups]
269
+ full_single_fp32_vector = torch.cat(merged_partitions, 0)
270
+ merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
271
+ avail_numel = sum(
272
+ [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
273
+
274
+ if debug:
275
+ wanted_params = sum([len(shapes) for shapes in param_shapes])
276
+ wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
277
+ # not asserting if there is a mismatch due to possible padding
278
+ print(f"Have {avail_numel} numels to process.")
279
+ print(f"Need {wanted_numel} numels in {wanted_params} params.")
280
+
281
+ # params
282
+ # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
283
+ # out-of-core computing solution
284
+ total_numel = 0
285
+ total_params = 0
286
+ for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
287
+ offset = 0
288
+ avail_numel = full_single_fp32_vector.numel()
289
+ for name, shape in shapes.items():
290
+
291
+ unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
292
+ total_numel += unpartitioned_numel
293
+ total_params += 1
294
+
295
+ if debug:
296
+ print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
297
+ state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
298
+ offset += unpartitioned_numel
299
+
300
+ # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
301
+ # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
302
+ # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
303
+ # live optimizer object, so we are checking that the numbers are within the right range
304
+ align_to = 2 * world_size
305
+
306
+ def zero2_align(x):
307
+ return align_to * math.ceil(x / align_to)
308
+
309
+ if debug:
310
+ print(f"original offset={offset}, avail_numel={avail_numel}")
311
+
312
+ offset = zero2_align(offset)
313
+ avail_numel = zero2_align(avail_numel)
314
+
315
+ if debug:
316
+ print(f"aligned offset={offset}, avail_numel={avail_numel}")
317
+
318
+ # Sanity check
319
+ if offset != avail_numel:
320
+ raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
321
+
322
+ print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
323
+
324
+
325
+ def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
326
+ exclude_frozen_parameters):
327
+ state_dict = OrderedDict()
328
+
329
+ # buffers
330
+ buffers = zero_model_states[0].buffers
331
+ state_dict.update(buffers)
332
+ if debug:
333
+ print(f"added {len(buffers)} buffers")
334
+
335
+ if not exclude_frozen_parameters:
336
+ _zero2_merge_frozen_params(state_dict, zero_model_states)
337
+
338
+ _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
339
+
340
+ # recover shared parameters
341
+ for pair in zero_model_states[0].shared_params:
342
+ if pair[1] in state_dict:
343
+ state_dict[pair[0]] = state_dict[pair[1]]
344
+
345
+ return state_dict
346
+
347
+
348
+ def zero3_partitioned_param_info(unpartitioned_numel, world_size):
349
+ remainder = unpartitioned_numel % world_size
350
+ padding_numel = (world_size - remainder) if remainder else 0
351
+ partitioned_numel = math.ceil(unpartitioned_numel / world_size)
352
+ return partitioned_numel, padding_numel
353
+
354
+
355
+ def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
356
+ if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
357
+ return
358
+
359
+ if debug:
360
+ for i in range(world_size):
361
+ num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
362
+ print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
363
+
364
+ frozen_param_shapes = zero_model_states[0].frozen_param_shapes
365
+ wanted_params = len(frozen_param_shapes)
366
+ wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
367
+ avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
368
+ print(f'Frozen params: Have {avail_numel} numels to process.')
369
+ print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
370
+
371
+ total_params = 0
372
+ total_numel = 0
373
+ for name, shape in zero_model_states[0].frozen_param_shapes.items():
374
+ total_params += 1
375
+ unpartitioned_numel = shape.numel()
376
+ total_numel += unpartitioned_numel
377
+
378
+ param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
379
+ state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
380
+
381
+ partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
382
+
383
+ if debug:
384
+ print(
385
+ f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
386
+ )
387
+
388
+ print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
389
+
390
+
391
+ class GatheredTensor:
392
+ """
393
+ A pseudo tensor that collects partitioned weights.
394
+ It is more memory efficient when there are multiple groups.
395
+ """
396
+
397
+ def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
398
+ self.flat_groups = flat_groups
399
+ self.flat_groups_offset = flat_groups_offset
400
+ self.offset = offset
401
+ self.partitioned_numel = partitioned_numel
402
+ self.shape = shape
403
+ self.dtype = self.flat_groups[0][0].dtype
404
+
405
+ def contiguous(self):
406
+ """
407
+ Merge partitioned weights from flat_groups into a single tensor.
408
+ """
409
+ end_idx = self.offset + self.partitioned_numel
410
+ world_size = len(self.flat_groups)
411
+ pad_flat_param_chunks = []
412
+
413
+ for rank_i in range(world_size):
414
+ # for each rank, we need to collect weights from related group/groups
415
+ flat_groups_at_rank_i = self.flat_groups[rank_i]
416
+ start_group_id = None
417
+ end_group_id = None
418
+ for group_id in range(len(self.flat_groups_offset)):
419
+ if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
420
+ start_group_id = group_id
421
+ if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
422
+ end_group_id = group_id
423
+ break
424
+ # collect weights from related group/groups
425
+ for group_id in range(start_group_id, end_group_id + 1):
426
+ flat_tensor = flat_groups_at_rank_i[group_id]
427
+ start_offset = self.offset - self.flat_groups_offset[group_id]
428
+ end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
429
+ pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
430
+
431
+ # collect weights from all ranks
432
+ pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
433
+ param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
434
+ return param
435
+
436
+
437
+ def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
438
+ param_shapes = zero_model_states[0].param_shapes
439
+ avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
440
+
441
+ # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
442
+ # param, re-consolidating each param, while dealing with padding if any
443
+
444
+ # merge list of dicts, preserving order
445
+ param_shapes = {k: v for d in param_shapes for k, v in d.items()}
446
+
447
+ if debug:
448
+ for i in range(world_size):
449
+ print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
450
+
451
+ wanted_params = len(param_shapes)
452
+ wanted_numel = sum(shape.numel() for shape in param_shapes.values())
453
+ # not asserting if there is a mismatch due to possible padding
454
+ avail_numel = fp32_flat_groups[0].numel() * world_size
455
+ print(f"Trainable params: Have {avail_numel} numels to process.")
456
+ print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
457
+
458
+ # params
459
+ # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
460
+ # out-of-core computing solution
461
+ offset = 0
462
+ total_numel = 0
463
+ total_params = 0
464
+ flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
465
+ for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
466
+ unpartitioned_numel = shape.numel()
467
+ total_numel += unpartitioned_numel
468
+ total_params += 1
469
+ partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
470
+
471
+ if debug:
472
+ print(
473
+ f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
474
+ )
475
+
476
+ # memory efficient tensor
477
+ tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
478
+ state_dict[name] = tensor
479
+ offset += partitioned_numel
480
+
481
+ offset *= world_size
482
+
483
+ # Sanity check
484
+ if offset != avail_numel:
485
+ raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
486
+
487
+ print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
488
+
489
+
490
+ def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
491
+ exclude_frozen_parameters):
492
+ state_dict = OrderedDict()
493
+
494
+ # buffers
495
+ buffers = zero_model_states[0].buffers
496
+ state_dict.update(buffers)
497
+ if debug:
498
+ print(f"added {len(buffers)} buffers")
499
+
500
+ if not exclude_frozen_parameters:
501
+ _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
502
+
503
+ _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
504
+
505
+ # recover shared parameters
506
+ for pair in zero_model_states[0].shared_params:
507
+ if pair[1] in state_dict:
508
+ state_dict[pair[0]] = state_dict[pair[1]]
509
+
510
+ return state_dict
511
+
512
+
513
+ def to_torch_tensor(state_dict, return_empty_tensor=False):
514
+ """
515
+ Convert state_dict of GatheredTensor to torch tensor
516
+ """
517
+ torch_state_dict = {}
518
+ converted_tensors = {}
519
+ for name, tensor in state_dict.items():
520
+ tensor_id = id(tensor)
521
+ if tensor_id in converted_tensors: # shared tensors
522
+ shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
523
+ torch_state_dict[name] = shared_tensor
524
+ else:
525
+ converted_tensors[tensor_id] = name
526
+ if return_empty_tensor:
527
+ torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
528
+ else:
529
+ torch_state_dict[name] = tensor.contiguous()
530
+ return torch_state_dict
531
+
532
+
533
+ def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
534
+ tag=None,
535
+ exclude_frozen_parameters=False,
536
+ lazy_mode=False):
537
+ """
538
+ Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
539
+ ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
540
+ via a model hub.
541
+
542
+ Args:
543
+ - ``checkpoint_dir``: path to the desired checkpoint folder
544
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
545
+ - ``exclude_frozen_parameters``: exclude frozen parameters
546
+ - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
547
+ Convert the pesduo tensor to torch tensor by ``.contiguous()``
548
+
549
+ Returns:
550
+ - pytorch ``state_dict``
551
+
552
+ A typical usage might be ::
553
+
554
+ from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
555
+ # do the training and checkpoint saving
556
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
557
+ model = model.cpu() # move to cpu
558
+ model.load_state_dict(state_dict)
559
+ # submit to model hub or save the model to share with others
560
+
561
+ In this example the ``model`` will no longer be usable in the deepspeed context of the same
562
+ application. i.e. you will need to re-initialize the deepspeed engine, since
563
+ ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
564
+
565
+ If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
566
+
567
+ Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
568
+ You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
569
+ the checkpoint. Or you can load state_dict in lazy mode ::
570
+
571
+ from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
572
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
573
+ for name, lazy_tensor in state_dict.item():
574
+ tensor = lazy_tensor.contiguous() # to cpu
575
+ print(name, tensor)
576
+ # del tensor to release memory if it no longer in use
577
+ """
578
+ if tag is None:
579
+ latest_path = os.path.join(checkpoint_dir, 'latest')
580
+ if os.path.isfile(latest_path):
581
+ with open(latest_path, 'r') as fd:
582
+ tag = fd.read().strip()
583
+ else:
584
+ raise ValueError(f"Unable to find 'latest' file at {latest_path}")
585
+
586
+ ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
587
+
588
+ if not os.path.isdir(ds_checkpoint_dir):
589
+ raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
590
+
591
+ state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
592
+ if lazy_mode:
593
+ return state_dict
594
+ else:
595
+ return to_torch_tensor(state_dict)
596
+
597
+
598
+ def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
599
+ output_dir,
600
+ max_shard_size="5GB",
601
+ safe_serialization=False,
602
+ tag=None,
603
+ exclude_frozen_parameters=False):
604
+ """
605
+ Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
606
+ loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
607
+
608
+ Args:
609
+ - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
610
+ - ``output_dir``: directory to the pytorch fp32 state_dict output files
611
+ - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
612
+ - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
613
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
614
+ - ``exclude_frozen_parameters``: exclude frozen parameters
615
+ """
616
+
617
+ # Dependency pre-check
618
+ if safe_serialization:
619
+ try:
620
+ from safetensors.torch import save_file
621
+ except ImportError:
622
+ print('If you want to use `safe_serialization`, please `pip install safetensors`')
623
+ raise
624
+ if max_shard_size is not None:
625
+ try:
626
+ from huggingface_hub import split_torch_state_dict_into_shards
627
+ except ImportError:
628
+ print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
629
+ raise
630
+
631
+ # Convert zero checkpoint to state_dict
632
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
633
+ tag,
634
+ exclude_frozen_parameters,
635
+ lazy_mode=True)
636
+
637
+ # Shard the model if it is too big.
638
+ weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
639
+ if max_shard_size is not None:
640
+ filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
641
+ # an memory-efficient approach for sharding
642
+ empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
643
+ state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
644
+ filename_pattern=filename_pattern,
645
+ max_shard_size=max_shard_size)
646
+ else:
647
+ from collections import namedtuple
648
+ StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
649
+ state_dict_split = StateDictSplit(is_sharded=False,
650
+ filename_to_tensors={weights_name: list(state_dict.keys())})
651
+
652
+ # Save the model by shard
653
+ os.makedirs(output_dir, exist_ok=True)
654
+ filename_to_tensors = state_dict_split.filename_to_tensors.items()
655
+ for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
656
+ shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
657
+ shard_state_dict = to_torch_tensor(shard_state_dict)
658
+ output_path = os.path.join(output_dir, shard_file)
659
+ if safe_serialization:
660
+ save_file(shard_state_dict, output_path, metadata={"format": "pt"})
661
+ else:
662
+ torch.save(shard_state_dict, output_path)
663
+ # release the memory of current shard
664
+ for tensor_name in list(shard_state_dict.keys()):
665
+ del state_dict[tensor_name]
666
+ del shard_state_dict[tensor_name]
667
+ del shard_state_dict
668
+ gc.collect()
669
+
670
+ # Save index if sharded
671
+ if state_dict_split.is_sharded:
672
+ index = {
673
+ "metadata": state_dict_split.metadata,
674
+ "weight_map": state_dict_split.tensor_to_filename,
675
+ }
676
+ save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
677
+ save_index_file = os.path.join(output_dir, save_index_file)
678
+ with open(save_index_file, "w", encoding="utf-8") as f:
679
+ content = json.dumps(index, indent=2, sort_keys=True) + "\n"
680
+ f.write(content)
681
+
682
+
683
+ def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
684
+ """
685
+ 1. Put the provided model to cpu
686
+ 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
687
+ 3. Load it into the provided model
688
+
689
+ Args:
690
+ - ``model``: the model object to update
691
+ - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
692
+ - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
693
+
694
+ Returns:
695
+ - ``model`: modified model
696
+
697
+ Make sure you have plenty of CPU memory available before you call this function. If you don't
698
+ have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
699
+ conveniently placed for you in the checkpoint folder.
700
+
701
+ A typical usage might be ::
702
+
703
+ from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
704
+ model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
705
+ # submit to model hub or save the model to share with others
706
+
707
+ Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
708
+ of the same application. i.e. you will need to re-initialize the deepspeed engine, since
709
+ ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
710
+
711
+ """
712
+ logger.info("Extracting fp32 weights")
713
+ state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
714
+
715
+ logger.info("Overwriting model with fp32 weights")
716
+ model = model.cpu()
717
+ model.load_state_dict(state_dict, strict=False)
718
+
719
+ return model
720
+
721
+
722
+ if __name__ == "__main__":
723
+ parser = argparse.ArgumentParser()
724
+ parser.add_argument("checkpoint_dir",
725
+ type=str,
726
+ help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
727
+ parser.add_argument("output_dir",
728
+ type=str,
729
+ help="directory to the pytorch fp32 state_dict output files"
730
+ "(e.g. path/checkpoint-12-output/)")
731
+ parser.add_argument(
732
+ "--max_shard_size",
733
+ type=str,
734
+ default="5GB",
735
+ help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
736
+ "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
737
+ "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
738
+ "without CPU OOM issues.")
739
+ parser.add_argument(
740
+ "--safe_serialization",
741
+ default=False,
742
+ action='store_true',
743
+ help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
744
+ parser.add_argument("-t",
745
+ "--tag",
746
+ type=str,
747
+ default=None,
748
+ help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
749
+ parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
750
+ parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
751
+ args = parser.parse_args()
752
+
753
+ debug = args.debug
754
+
755
+ convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
756
+ args.output_dir,
757
+ max_shard_size=args.max_shard_size,
758
+ safe_serialization=args.safe_serialization,
759
+ tag=args.tag,
760
+ exclude_frozen_parameters=args.exclude_frozen_parameters)
baseline/pi05_stack_cups_148/10000/_CHECKPOINT_METADATA ADDED
@@ -0,0 +1 @@
 
 
1
+ {"item_handlers": {"assets": "openpi.training.checkpoints.CallbackHandler", "params": "orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler", "train_state": "orbax.checkpoint._src.handlers.pytree_checkpoint_handler.PyTreeCheckpointHandler"}, "metrics": {}, "performance_metrics": {}, "init_timestamp_nsecs": 1793975600237098653, "commit_timestamp_nsecs": 1793975624028994132, "custom_metadata": {}}
baseline/pi05_stack_cups_148/10000/assets/galaxea_r1_lite_stack_cups/norm_stats.json ADDED
@@ -0,0 +1,136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "norm_stats": {
3
+ "state": {
4
+ "mean": [
5
+ -0.1625075198268781,
6
+ 1.0842620647445143,
7
+ -0.36731704502858703,
8
+ -0.41283875757757366,
9
+ 0.03702329523675885,
10
+ -0.030393585719784302,
11
+ 88.70043343606748,
12
+ 0.1001759839971589,
13
+ 1.1130726768189518,
14
+ -0.319068614756271,
15
+ -0.4901851291457013,
16
+ 0.04893345683788824,
17
+ 0.04471603786466137,
18
+ 85.58191268910615
19
+ ],
20
+ "std": [
21
+ 0.2700353873323502,
22
+ 1.1467528567093037,
23
+ 0.5530823686632277,
24
+ 0.5927237779599276,
25
+ 0.12713039475675825,
26
+ 0.06632369407822619,
27
+ 24.374810634433103,
28
+ 0.13874521422404704,
29
+ 1.068493942298225,
30
+ 0.43438903122681344,
31
+ 0.6259246013658152,
32
+ 0.08044048709309755,
33
+ 0.031995482570700895,
34
+ 26.794849672554083
35
+ ],
36
+ "q01": [
37
+ -0.8736169934272766,
38
+ 0.2148936241865158,
39
+ -1.9360382914543153,
40
+ -1.5257446765899658,
41
+ -0.4946638309955597,
42
+ -0.2444680780172348,
43
+ 12.464552879333496,
44
+ -0.06808510422706604,
45
+ 0.21553191542625427,
46
+ -1.313191533088684,
47
+ -1.5253190994262695,
48
+ -0.11170212924480438,
49
+ -0.05063829943537712,
50
+ 12.032872200012207
51
+ ],
52
+ "q99": [
53
+ 0.16063830256462097,
54
+ 3.034255266189575,
55
+ -0.0019148936262354255,
56
+ 0.0642382982373234,
57
+ 0.16808511316776276,
58
+ 0.11829787492752075,
59
+ 102.18600463867188,
60
+ 0.4502127766609192,
61
+ 2.8959574699401855,
62
+ -0.0008510638144798577,
63
+ 0.034255318343639374,
64
+ 0.2665957510471344,
65
+ 0.14464681506156848,
66
+ 102.34491729736328
67
+ ]
68
+ },
69
+ "actions": {
70
+ "mean": [
71
+ -0.16473093956759127,
72
+ 0.8681155832258038,
73
+ -0.35903689579197906,
74
+ -0.422772990293642,
75
+ 0.013333487666197312,
76
+ -0.053439910524043906,
77
+ 86.547770940583,
78
+ 0.08334164575863477,
79
+ 0.9183426463724823,
80
+ -0.33061070522100666,
81
+ -0.5209033568473217,
82
+ 0.024604943696559547,
83
+ 0.01963253503542509,
84
+ 79.80457299498252
85
+ ],
86
+ "std": [
87
+ 0.26413852784851743,
88
+ 1.176010024006326,
89
+ 0.5652555232725361,
90
+ 0.6052913922882582,
91
+ 0.13028488266679109,
92
+ 0.06823171888650201,
93
+ 33.50286769923162,
94
+ 0.13674243725761884,
95
+ 1.099167413108321,
96
+ 0.4524649041976627,
97
+ 0.6366756189128563,
98
+ 0.08196322531802316,
99
+ 0.032833390548008885,
100
+ 39.14279967697801
101
+ ],
102
+ "q01": [
103
+ -0.8682331442832947,
104
+ 0.0,
105
+ -1.9711652994155884,
106
+ -1.5707999467849731,
107
+ -0.5215534567832947,
108
+ -0.2730485796928406,
109
+ 0.0,
110
+ -0.09970875084400177,
111
+ 0.0,
112
+ -1.368310809135437,
113
+ -1.5707999467849731,
114
+ -0.13805827498435974,
115
+ -0.07669904083013535,
116
+ 0.0
117
+ ],
118
+ "q99": [
119
+ 0.14879614114761353,
120
+ 2.8746800422668457,
121
+ 0.0,
122
+ 0.05062136426568031,
123
+ 0.14879614114761353,
124
+ 0.10431069135665894,
125
+ 101.44685363769531,
126
+ 0.4264466464519501,
127
+ 2.728951930999756,
128
+ 0.0,
129
+ 0.019941750913858414,
130
+ 0.251572847366333,
131
+ 0.12425244599580765,
132
+ 102.08477783203125
133
+ ]
134
+ }
135
+ }
136
+ }
baseline/pi05_stack_cups_148/10000/params/_METADATA ADDED
@@ -0,0 +1 @@
 
 
1
+ {"tree_metadata": {"('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [144]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoder_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoder_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [144]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 144]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_0', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_0", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 144]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 144]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'LayerNorm_1', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "LayerNorm_1", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 144]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 538]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 144, 4304]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 144]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MlpBlock_0', 'Dense_1', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MlpBlock_0", "key_type": 2}, {"key": "Dense_1", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 538, 1152]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 2, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'key', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "key", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 144, 16, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 144]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'out', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 2, 72, 1152]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 2, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'query', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "query", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 144, 16, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 2, 72]}}, "('params', 'PaliGemma', 'img', 'Transformer', 'encoderblock', 'MultiHeadDotProductAttention_0', 'value', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "Transformer", "key_type": 2}, {"key": "encoderblock", "key_type": 2}, {"key": "MultiHeadDotProductAttention_0", "key_type": 2}, {"key": "value", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [27, 144, 16, 72]}}, "('params', 'PaliGemma', 'img', 'embedding', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [144]}}, "('params', 'PaliGemma', 'img', 'embedding', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "embedding", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [14, 14, 3, 144]}}, "('params', 'PaliGemma', 'img', 'head', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [256]}}, "('params', 'PaliGemma', 'img', 'head', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "head", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [144, 2048]}}, "('params', 'PaliGemma', 'img', 'pos_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "img", "key_type": 2}, {"key": "pos_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [1, 32, 1152]}}, "('params', 'PaliGemma', 'llm', 'embedder', 'input_embedding', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "embedder", "key_type": 2}, {"key": "input_embedding", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [32144, 2048]}}, "('params', 'PaliGemma', 'llm', 'final_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [256]}}, "('params', 'PaliGemma', 'llm', 'final_norm_1', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [384]}}, "('params', 'PaliGemma', 'llm', 'final_norm_1', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "final_norm_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [128, 3072]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'lora_a', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "lora_a", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 256, 16]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'lora_b', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "lora_b", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 16, 2048]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 256, 2048]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum_1', 'lora_a', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum_1", "key_type": 2}, {"key": "lora_a", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 256, 32]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum_1', 'lora_b', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum_1", "key_type": 2}, {"key": "lora_b", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 32, 1024]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'attn_vec_einsum_1', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "attn_vec_einsum_1", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 256, 1024]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'lora_a', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "lora_a", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 256, 16]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'lora_b', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "lora_b", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 2, 256]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 256, 256]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum_1', 'lora_a', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum_1", "key_type": 2}, {"key": "lora_a", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 128, 32]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum_1', 'lora_b', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum_1", "key_type": 2}, {"key": "lora_b", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 4, 256]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'kv_einsum_1', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "kv_einsum_1", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 1, 128, 256]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'lora_a', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "lora_a", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 2048, 16]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'lora_b', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "lora_b", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 16, 256]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 2048, 256]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum_1', 'lora_a', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum_1", "key_type": 2}, {"key": "lora_a", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 1024, 32]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum_1', 'lora_b', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum_1", "key_type": 2}, {"key": "lora_b", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 32, 256]}}, "('params', 'PaliGemma', 'llm', 'layers', 'attn', 'q_einsum_1', 'w', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "attn", "key_type": 2}, {"key": "q_einsum_1", "key_type": 2}, {"key": "w", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 1, 1024, 256]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 256, 16384]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum_lora_a', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum_lora_a", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 256, 16]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'gating_einsum_lora_b', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "gating_einsum_lora_b", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2, 16384]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2048, 2048]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear_lora_a', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear_lora_a", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2048, 16]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp', 'linear_lora_b', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp", "key_type": 2}, {"key": "linear_lora_b", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 2048]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp_1', 'gating_einsum', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp_1", "key_type": 2}, {"key": "gating_einsum", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 128, 4096]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp_1', 'gating_einsum_lora_a', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp_1", "key_type": 2}, {"key": "gating_einsum_lora_a", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 128, 32]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp_1', 'gating_einsum_lora_b', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp_1", "key_type": 2}, {"key": "gating_einsum_lora_b", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 2, 4, 4096]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp_1', 'linear', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp_1", "key_type": 2}, {"key": "linear", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512, 1024]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp_1', 'linear_lora_a', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp_1", "key_type": 2}, {"key": "linear_lora_a", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 512, 32]}}, "('params', 'PaliGemma', 'llm', 'layers', 'mlp_1', 'linear_lora_b', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "mlp_1", "key_type": 2}, {"key": "linear_lora_b", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 4, 1024]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 256]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm_1', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 384]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_attention_norm_1', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_attention_norm_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 128, 3072]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm', 'scale', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm", "key_type": 2}, {"key": "scale", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 256]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm_1', 'Dense_0', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 384]}}, "('params', 'PaliGemma', 'llm', 'layers', 'pre_ffw_norm_1', 'Dense_0', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "PaliGemma", "key_type": 2}, {"key": "llm", "key_type": 2}, {"key": "layers", "key_type": 2}, {"key": "pre_ffw_norm_1", "key_type": 2}, {"key": "Dense_0", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [18, 128, 3072]}}, "('params', 'action_in_proj', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "action_in_proj", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [128]}}, "('params', 'action_in_proj', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "action_in_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [4, 1024]}}, "('params', 'action_out_proj', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "action_out_proj", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [4]}}, "('params', 'action_out_proj', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "action_out_proj", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [128, 32]}}, "('params', 'time_mlp_in', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "time_mlp_in", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [128]}}, "('params', 'time_mlp_in', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "time_mlp_in", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [128, 1024]}}, "('params', 'time_mlp_out', 'bias', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "time_mlp_out", "key_type": 2}, {"key": "bias", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [128]}}, "('params', 'time_mlp_out', 'kernel', 'value')": {"key_metadata": [{"key": "params", "key_type": 2}, {"key": "time_mlp_out", "key_type": 2}, {"key": "kernel", "key_type": 2}, {"key": "value", "key_type": 2}], "value_metadata": {"value_type": "jax.Array", "skip_deserialize": false, "write_shape": [128, 1024]}}}, "use_zarr3": false, "store_array_data_equal_to_fill_value": true, "custom_metadata": null}
baseline/pi05_stack_cups_148/10000/params/_sharding ADDED
@@ -0,0 +1 @@
 
 
1
+ {"cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2Rlcl9ub3JtLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC52YWx1ZS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5rZXkua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5vdXQua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk11bHRpSGVhZERvdFByb2R1Y3RBdHRlbnRpb25fMC5xdWVyeS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLk1scEJsb2NrXzAuRGVuc2VfMS5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8wLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuVHJhbnNmb3JtZXIuZW5jb2RlcmJsb2NrLkxheWVyTm9ybV8xLnNjYWxlLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuZW1iZWRkaW5nLmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcuaGVhZC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5pbWcucG9zX2VtYmVkZGluZy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZW1iZWRkZXIuaW5wdXRfZW1iZWRkaW5nLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZmluYWxfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZmluYWxfbm9ybV8xLkRlbnNlXzAuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0uZmluYWxfbm9ybV8xLkRlbnNlXzAua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5nYXRpbmdfZWluc3VtLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5nYXRpbmdfZWluc3VtX2xvcmFfYS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5nYXRpbmdfZWluc3VtX2xvcmFfYi52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5saW5lYXIudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5saW5lYXJfbG9yYV9hLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scC5saW5lYXJfbG9yYV9iLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scF8xLmdhdGluZ19laW5zdW0udmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scF8xLmdhdGluZ19laW5zdW1fbG9yYV9hLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scF8xLmdhdGluZ19laW5zdW1fbG9yYV9iLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scF8xLmxpbmVhci52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scF8xLmxpbmVhcl9sb3JhX2EudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLm1scF8xLmxpbmVhcl9sb3JhX2IudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtLmxvcmFfYS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtLmxvcmFfYi52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtXzEubG9yYV9hLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtXzEubG9yYV9iLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4uYXR0bl92ZWNfZWluc3VtXzEudy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtLmxvcmFfYS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtLmxvcmFfYi52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtLncudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtXzEubG9yYV9hLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtXzEubG9yYV9iLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ua3ZfZWluc3VtXzEudy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW0ubG9yYV9hLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW0ubG9yYV9iLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW0udy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW1fMS53LnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW1fMS5sb3JhX2EudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLmF0dG4ucV9laW5zdW1fMS5sb3JhX2IudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9hdHRlbnRpb25fbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9hdHRlbnRpb25fbm9ybV8xLkRlbnNlXzAuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9hdHRlbnRpb25fbm9ybV8xLkRlbnNlXzAua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9mZndfbm9ybS5zY2FsZS52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9mZndfbm9ybV8xLkRlbnNlXzAuYmlhcy52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLlBhbGlHZW1tYS5sbG0ubGF5ZXJzLnByZV9mZndfbm9ybV8xLkRlbnNlXzAua2VybmVsLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLmFjdGlvbl9pbl9wcm9qLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLmFjdGlvbl9pbl9wcm9qLmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLmFjdGlvbl9vdXRfcHJvai5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLmFjdGlvbl9vdXRfcHJvai5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLnRpbWVfbWxwX291dC5iaWFzLnZhbHVl":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLnRpbWVfbWxwX291dC5rZXJuZWwudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLnRpbWVfbWxwX2luLmJpYXMudmFsdWU=":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}","cGFyYW1zLnRpbWVfbWxwX2luLmtlcm5lbC52YWx1ZQ==":"{\"sharding_type\": \"NamedSharding\", \"shape\": [8, 1], \"axis_names\": [\"batch\", \"fsdp\"], \"partition_spec\": [], \"device_mesh\": {\"mesh\": [[{\"id\": 0}], [{\"id\": 1}], [{\"id\": 2}], [{\"id\": 3}], [{\"id\": 4}], [{\"id\": 5}], [{\"id\": 6}], [{\"id\": 7}]]}}"}
baseline/pi05_stack_cups_148/10000/params/array_metadatas/process_0 ADDED
@@ -0,0 +1 @@
 
 
1
+ {"array_metadatas": [{"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.bias.value", "write_shape": [144], "chunk_shape": [144], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoder_norm.scale.value", "write_shape": [144], "chunk_shape": [144], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.bias.value", "write_shape": [27, 144], "chunk_shape": [27, 144], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_0.scale.value", "write_shape": [27, 144], "chunk_shape": [27, 144], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.bias.value", "write_shape": [27, 144], "chunk_shape": [27, 144], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.LayerNorm_1.scale.value", "write_shape": [27, 144], "chunk_shape": [27, 144], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.bias.value", "write_shape": [27, 538], "chunk_shape": [27, 538], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_0.kernel.value", "write_shape": [27, 144, 4304], "chunk_shape": [27, 144, 4304], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.bias.value", "write_shape": [27, 144], "chunk_shape": [27, 144], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MlpBlock_0.Dense_1.kernel.value", "write_shape": [27, 538, 1152], "chunk_shape": [27, 538, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.bias.value", "write_shape": [27, 2, 72], "chunk_shape": [27, 2, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.key.kernel.value", "write_shape": [27, 144, 16, 72], "chunk_shape": [27, 144, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.bias.value", "write_shape": [27, 144], "chunk_shape": [27, 144], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.out.kernel.value", "write_shape": [27, 2, 72, 1152], "chunk_shape": [27, 2, 72, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.bias.value", "write_shape": [27, 2, 72], "chunk_shape": [27, 2, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.query.kernel.value", "write_shape": [27, 144, 16, 72], "chunk_shape": [27, 144, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.bias.value", "write_shape": [27, 2, 72], "chunk_shape": [27, 2, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.Transformer.encoderblock.MultiHeadDotProductAttention_0.value.kernel.value", "write_shape": [27, 144, 16, 72], "chunk_shape": [27, 144, 16, 72], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.bias.value", "write_shape": [144], "chunk_shape": [144], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.embedding.kernel.value", "write_shape": [14, 14, 3, 144], "chunk_shape": [14, 14, 3, 144], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.bias.value", "write_shape": [256], "chunk_shape": [256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.head.kernel.value", "write_shape": [144, 2048], "chunk_shape": [144, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.img.pos_embedding.value", "write_shape": [1, 32, 1152], "chunk_shape": [1, 32, 1152], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.embedder.input_embedding.value", "write_shape": [32144, 2048], "chunk_shape": [32144, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.final_norm.scale.value", "write_shape": [256], "chunk_shape": [256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.final_norm_1.Dense_0.bias.value", "write_shape": [384], "chunk_shape": [384], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.final_norm_1.Dense_0.kernel.value", "write_shape": [128, 3072], "chunk_shape": [128, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum.lora_a.value", "write_shape": [18, 1, 256, 16], "chunk_shape": [18, 1, 256, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum.lora_b.value", "write_shape": [18, 1, 16, 2048], "chunk_shape": [18, 1, 16, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum.w.value", "write_shape": [18, 1, 256, 2048], "chunk_shape": [18, 1, 256, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum_1.lora_a.value", "write_shape": [18, 1, 256, 32], "chunk_shape": [18, 1, 256, 32], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum_1.lora_b.value", "write_shape": [18, 1, 32, 1024], "chunk_shape": [18, 1, 32, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.attn_vec_einsum_1.w.value", "write_shape": [18, 1, 256, 1024], "chunk_shape": [18, 1, 256, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum.lora_a.value", "write_shape": [18, 2, 1, 256, 16], "chunk_shape": [18, 2, 1, 256, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum.lora_b.value", "write_shape": [18, 2, 1, 2, 256], "chunk_shape": [18, 2, 1, 2, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum.w.value", "write_shape": [18, 2, 1, 256, 256], "chunk_shape": [18, 2, 1, 256, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum_1.lora_a.value", "write_shape": [18, 2, 1, 128, 32], "chunk_shape": [18, 2, 1, 128, 32], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum_1.lora_b.value", "write_shape": [18, 2, 1, 4, 256], "chunk_shape": [18, 2, 1, 4, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.kv_einsum_1.w.value", "write_shape": [18, 2, 1, 128, 256], "chunk_shape": [18, 2, 1, 128, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum.lora_a.value", "write_shape": [18, 1, 2048, 16], "chunk_shape": [18, 1, 2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum.lora_b.value", "write_shape": [18, 1, 16, 256], "chunk_shape": [18, 1, 16, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum.w.value", "write_shape": [18, 1, 2048, 256], "chunk_shape": [18, 1, 2048, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum_1.lora_a.value", "write_shape": [18, 1, 1024, 32], "chunk_shape": [18, 1, 1024, 32], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum_1.lora_b.value", "write_shape": [18, 1, 32, 256], "chunk_shape": [18, 1, 32, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.attn.q_einsum_1.w.value", "write_shape": [18, 1, 1024, 256], "chunk_shape": [18, 1, 1024, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.gating_einsum.value", "write_shape": [18, 2, 256, 16384], "chunk_shape": [18, 2, 256, 16384], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.gating_einsum_lora_a.value", "write_shape": [18, 2, 256, 16], "chunk_shape": [18, 2, 256, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.gating_einsum_lora_b.value", "write_shape": [18, 2, 2, 16384], "chunk_shape": [18, 2, 2, 16384], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.linear.value", "write_shape": [18, 2048, 2048], "chunk_shape": [18, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.linear_lora_a.value", "write_shape": [18, 2048, 16], "chunk_shape": [18, 2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp.linear_lora_b.value", "write_shape": [18, 2, 2048], "chunk_shape": [18, 2, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp_1.gating_einsum.value", "write_shape": [18, 2, 128, 4096], "chunk_shape": [18, 2, 128, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp_1.gating_einsum_lora_a.value", "write_shape": [18, 2, 128, 32], "chunk_shape": [18, 2, 128, 32], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp_1.gating_einsum_lora_b.value", "write_shape": [18, 2, 4, 4096], "chunk_shape": [18, 2, 4, 4096], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp_1.linear.value", "write_shape": [18, 512, 1024], "chunk_shape": [18, 512, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp_1.linear_lora_a.value", "write_shape": [18, 512, 32], "chunk_shape": [18, 512, 32], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.mlp_1.linear_lora_b.value", "write_shape": [18, 4, 1024], "chunk_shape": [18, 4, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_attention_norm.scale.value", "write_shape": [18, 256], "chunk_shape": [18, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_attention_norm_1.Dense_0.bias.value", "write_shape": [18, 384], "chunk_shape": [18, 384], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_attention_norm_1.Dense_0.kernel.value", "write_shape": [18, 128, 3072], "chunk_shape": [18, 128, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_ffw_norm.scale.value", "write_shape": [18, 256], "chunk_shape": [18, 256], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_ffw_norm_1.Dense_0.bias.value", "write_shape": [18, 384], "chunk_shape": [18, 384], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.PaliGemma.llm.layers.pre_ffw_norm_1.Dense_0.kernel.value", "write_shape": [18, 128, 3072], "chunk_shape": [18, 128, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.action_in_proj.bias.value", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.action_in_proj.kernel.value", "write_shape": [4, 1024], "chunk_shape": [4, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.action_out_proj.bias.value", "write_shape": [4], "chunk_shape": [4], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.action_out_proj.kernel.value", "write_shape": [128, 32], "chunk_shape": [128, 32], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.time_mlp_in.bias.value", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.time_mlp_in.kernel.value", "write_shape": [128, 1024], "chunk_shape": [128, 1024], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.time_mlp_out.bias.value", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "params.time_mlp_out.kernel.value", "write_shape": [128, 1024], "chunk_shape": [128, 1024], "ext_metadata": null}}]}
baseline/pi05_stack_cups_148/10000/params/d/4c1273032c55c08244f3a733657fcac0 ADDED
Binary file (41.4 kB). View file
 
baseline/pi05_stack_cups_148/10000/params/manifest.ocdbt ADDED
Binary file (120 Bytes). View file
 
baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/319e3d0585c3de362e83981c20d3fe89 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92a32f0e502eab92ba5be7220dccc99aaa8545d54886f88a184964f1b4cb7d70
3
+ size 472802556
baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/3c032c830f6cc7175afe2811c1a18827 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62c8a58d326ec3db170ed385fb1abb9f81a8ed9f38ec53f6857ec567392be33c
3
+ size 1470771998
baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/58f68e6e347aedcae64dc9ca8caa082b ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9eb13457fc97f4f9b3087e020891f34cd3f087900b5404079800faaec0fb527d
3
+ size 2160225551
baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/5b8109f6166eafdac21828028a913405 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:40941adec6467d076c00b5addcf833f811ef6e58377b9d6c1bd08296155bf947
3
+ size 226695
baseline/pi05_stack_cups_148/10000/params/ocdbt.process_0/d/6bd188f310ebefce038484a33ca74dc0 ADDED
Binary file (246 Bytes). View file