nimitt0 commited on
Commit
2ddce9a
·
verified ·
1 Parent(s): 8b07ef7

Add files using upload-large-folder tool

Browse files
checkpoint-1000/config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "action_dim": 32,
3
+ "action_head_cfg": {
4
+ "action_dim": 32,
5
+ "action_horizon": 16,
6
+ "add_pos_embed": true,
7
+ "backbone_embedding_dim": 2048,
8
+ "diffusion_model_cfg": {
9
+ "attention_head_dim": 48,
10
+ "cross_attention_dim": 2048,
11
+ "dropout": 0.2,
12
+ "final_dropout": true,
13
+ "interleave_self_attention": true,
14
+ "norm_type": "ada_norm",
15
+ "num_attention_heads": 32,
16
+ "num_layers": 16,
17
+ "output_dim": 1024,
18
+ "positional_embeddings": null
19
+ },
20
+ "hidden_size": 1024,
21
+ "input_embedding_dim": 1536,
22
+ "max_action_dim": 32,
23
+ "max_state_dim": 64,
24
+ "model_dtype": "float32",
25
+ "noise_beta_alpha": 1.5,
26
+ "noise_beta_beta": 1.0,
27
+ "noise_s": 0.999,
28
+ "num_inference_timesteps": 4,
29
+ "num_target_vision_tokens": 32,
30
+ "num_timestep_buckets": 1000,
31
+ "tune_diffusion_model": true,
32
+ "tune_projector": true,
33
+ "use_vlln": true,
34
+ "vl_self_attention_cfg": {
35
+ "attention_head_dim": 64,
36
+ "dropout": 0.2,
37
+ "final_dropout": true,
38
+ "num_attention_heads": 32,
39
+ "num_layers": 4,
40
+ "positional_embeddings": null
41
+ }
42
+ },
43
+ "action_horizon": 16,
44
+ "architectures": [
45
+ "GR00T_N1_5"
46
+ ],
47
+ "attn_implementation": null,
48
+ "backbone_cfg": {
49
+ "eagle_path": "NVEagle/eagle_er-qwen3_1_7B-Siglip2_400M_stage1_5_128gpu_er_v7_1mlp_nops",
50
+ "load_bf16": false,
51
+ "project_to_dim": null,
52
+ "reproject_vision": false,
53
+ "select_layer": 12,
54
+ "tune_llm": false,
55
+ "tune_visual": true,
56
+ "use_flash_attention": true
57
+ },
58
+ "compute_dtype": "bfloat16",
59
+ "hidden_size": 2048,
60
+ "model_dtype": "float32",
61
+ "model_type": "gr00t_n1_5",
62
+ "torch_dtype": "bfloat16",
63
+ "transformers_version": "4.51.3"
64
+ }
checkpoint-1000/experiment_cfg/metadata.json ADDED
@@ -0,0 +1,195 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "new_embodiment": {
3
+ "statistics": {
4
+ "state": {
5
+ "single_arm": {
6
+ "max": [
7
+ 2.0000243186950684,
8
+ 2.9169764518737793,
9
+ -0.038691196590662,
10
+ 1.2000755071640015,
11
+ 1.5494728088378906
12
+ ],
13
+ "min": [
14
+ -1.6060552597045898,
15
+ -0.0013021699851378798,
16
+ -2.8603312969207764,
17
+ -0.24804846942424774,
18
+ -2.4647815227508545
19
+ ],
20
+ "mean": [
21
+ 0.5791597962379456,
22
+ 1.1458725929260254,
23
+ -1.20786452293396,
24
+ 0.7232474684715271,
25
+ -0.6134248375892639
26
+ ],
27
+ "std": [
28
+ 0.9811665415763855,
29
+ 0.7337602376937866,
30
+ 0.6142615079879761,
31
+ 0.23586146533489227,
32
+ 0.966107964515686
33
+ ],
34
+ "q01": [
35
+ -1.4408677732944488,
36
+ -1.425088475116354e-05,
37
+ -2.781974565982819,
38
+ 0.27418837130069734,
39
+ -1.9999565029144286
40
+ ],
41
+ "q99": [
42
+ 1.9999998807907104,
43
+ 2.6177230525016784,
44
+ -0.3294296935200691,
45
+ 1.2000004053115845,
46
+ 1.4374910080432892
47
+ ]
48
+ },
49
+ "gripper": {
50
+ "max": [
51
+ 0.7455036640167236
52
+ ],
53
+ "min": [
54
+ -0.09471277892589569
55
+ ],
56
+ "mean": [
57
+ 0.000568521732930094
58
+ ],
59
+ "std": [
60
+ 0.14351940155029297
61
+ ],
62
+ "q01": [
63
+ -0.05334045834839344
64
+ ],
65
+ "q99": [
66
+ 0.5978925746679306
67
+ ]
68
+ }
69
+ },
70
+ "action": {
71
+ "single_arm": {
72
+ "max": [
73
+ 2.0000085830688477,
74
+ 2.9173049926757812,
75
+ -0.0662246122956276,
76
+ 1.2000755071640015,
77
+ 1.549481987953186
78
+ ],
79
+ "min": [
80
+ -1.5781737565994263,
81
+ -0.0013021699851378798,
82
+ -2.8603312969207764,
83
+ -0.24804846942424774,
84
+ -2.4347522258758545
85
+ ],
86
+ "mean": [
87
+ 0.5763245820999146,
88
+ 1.1609699726104736,
89
+ -1.2109930515289307,
90
+ 0.7210843563079834,
91
+ -0.6085058450698853
92
+ ],
93
+ "std": [
94
+ 0.9860737323760986,
95
+ 0.7361332774162292,
96
+ 0.6108555197715759,
97
+ 0.23312774300575256,
98
+ 0.9712501168251038
99
+ ],
100
+ "q01": [
101
+ -1.4462056386470794,
102
+ -1.4080941818974679e-05,
103
+ -2.7542264652252197,
104
+ 0.27880221664905547,
105
+ -1.9997829270362855
106
+ ],
107
+ "q99": [
108
+ 1.9999998807907104,
109
+ 2.6269060873985284,
110
+ -0.33764528483152395,
111
+ 1.200000286102295,
112
+ 1.4436372423171995
113
+ ]
114
+ },
115
+ "gripper": {
116
+ "max": [
117
+ 0.7455036640167236
118
+ ],
119
+ "min": [
120
+ -0.07934218645095825
121
+ ],
122
+ "mean": [
123
+ -0.0036394144408404827
124
+ ],
125
+ "std": [
126
+ 0.13472281396389008
127
+ ],
128
+ "q01": [
129
+ -0.05393668111413717
130
+ ],
131
+ "q99": [
132
+ 0.56789559841156
133
+ ]
134
+ }
135
+ }
136
+ },
137
+ "modalities": {
138
+ "video": {
139
+ "front": {
140
+ "resolution": [
141
+ 640,
142
+ 480
143
+ ],
144
+ "channels": 3,
145
+ "fps": 30.0
146
+ },
147
+ "wrist": {
148
+ "resolution": [
149
+ 640,
150
+ 480
151
+ ],
152
+ "channels": 3,
153
+ "fps": 30.0
154
+ }
155
+ },
156
+ "state": {
157
+ "single_arm": {
158
+ "absolute": true,
159
+ "rotation_type": null,
160
+ "shape": [
161
+ 5
162
+ ],
163
+ "continuous": true
164
+ },
165
+ "gripper": {
166
+ "absolute": true,
167
+ "rotation_type": null,
168
+ "shape": [
169
+ 1
170
+ ],
171
+ "continuous": true
172
+ }
173
+ },
174
+ "action": {
175
+ "single_arm": {
176
+ "absolute": true,
177
+ "rotation_type": null,
178
+ "shape": [
179
+ 5
180
+ ],
181
+ "continuous": true
182
+ },
183
+ "gripper": {
184
+ "absolute": true,
185
+ "rotation_type": null,
186
+ "shape": [
187
+ 1
188
+ ],
189
+ "continuous": true
190
+ }
191
+ }
192
+ },
193
+ "embodiment_tag": "new_embodiment"
194
+ }
195
+ }
checkpoint-1000/model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ee4d510847221ac72a25ea6eec1cfa59563281cdf1930f6e7cfae1cd46a0b82
3
+ size 4999367032
checkpoint-1000/model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1efc712847eff4f518876aa393fcf81da1a5cd7fdda2dc7f113a0ec6faea2f4d
3
+ size 2586705312
checkpoint-1000/model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b47a8f2ecd4442b624403aed730638ade0620e71a34a499e07f7671948a0ae4
3
+ size 8550720062
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db318eef8b61b1e618d544199e422f2b7c381c3b1db41c0fb2b5c07049d39864
3
+ size 14244
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3834182320079d17ccef3fad4a4c43c5ea55e060a5542d782f76379baf9a4c3a
3
+ size 1064
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,734 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.41981528127623846,
6
+ "eval_steps": 500,
7
+ "global_step": 1000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0041981528127623844,
14
+ "grad_norm": 5.267488956451416,
15
+ "learning_rate": 9e-06,
16
+ "loss": 1.0193,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.008396305625524769,
21
+ "grad_norm": 2.3370044231414795,
22
+ "learning_rate": 1.9e-05,
23
+ "loss": 0.3362,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.012594458438287154,
28
+ "grad_norm": 2.1865203380584717,
29
+ "learning_rate": 2.9e-05,
30
+ "loss": 0.2021,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.016792611251049538,
35
+ "grad_norm": 1.625726342201233,
36
+ "learning_rate": 3.9000000000000006e-05,
37
+ "loss": 0.1466,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.020990764063811923,
42
+ "grad_norm": 1.1913143396377563,
43
+ "learning_rate": 4.9e-05,
44
+ "loss": 0.1184,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.02518891687657431,
49
+ "grad_norm": 1.6674305200576782,
50
+ "learning_rate": 5.9e-05,
51
+ "loss": 0.1028,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.02938706968933669,
56
+ "grad_norm": 0.8989318013191223,
57
+ "learning_rate": 6.9e-05,
58
+ "loss": 0.0836,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.033585222502099076,
63
+ "grad_norm": 0.8966989517211914,
64
+ "learning_rate": 7.900000000000001e-05,
65
+ "loss": 0.0851,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.037783375314861464,
70
+ "grad_norm": 0.7644183039665222,
71
+ "learning_rate": 8.900000000000001e-05,
72
+ "loss": 0.0717,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.041981528127623846,
77
+ "grad_norm": 1.272506833076477,
78
+ "learning_rate": 9.900000000000001e-05,
79
+ "loss": 0.0774,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.04617968094038623,
84
+ "grad_norm": 1.4885015487670898,
85
+ "learning_rate": 9.999446382823013e-05,
86
+ "loss": 0.0712,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.05037783375314862,
91
+ "grad_norm": 1.201401710510254,
92
+ "learning_rate": 9.997532801828658e-05,
93
+ "loss": 0.072,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.054575986565911,
98
+ "grad_norm": 1.367532730102539,
99
+ "learning_rate": 9.99425294526634e-05,
100
+ "loss": 0.0682,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.05877413937867338,
105
+ "grad_norm": 0.9127843976020813,
106
+ "learning_rate": 9.989607709816091e-05,
107
+ "loss": 0.0659,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.06297229219143577,
112
+ "grad_norm": 0.7713141441345215,
113
+ "learning_rate": 9.983598365438902e-05,
114
+ "loss": 0.0528,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.06717044500419815,
119
+ "grad_norm": 0.9186705946922302,
120
+ "learning_rate": 9.976226555029522e-05,
121
+ "loss": 0.0495,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.07136859781696053,
126
+ "grad_norm": 0.5967724919319153,
127
+ "learning_rate": 9.967494293967312e-05,
128
+ "loss": 0.0522,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.07556675062972293,
133
+ "grad_norm": 1.0857634544372559,
134
+ "learning_rate": 9.95740396956525e-05,
135
+ "loss": 0.0578,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.07976490344248531,
140
+ "grad_norm": 0.5992318987846375,
141
+ "learning_rate": 9.945958340417283e-05,
142
+ "loss": 0.0478,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.08396305625524769,
147
+ "grad_norm": 0.5689753293991089,
148
+ "learning_rate": 9.93316053564413e-05,
149
+ "loss": 0.0478,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.08816120906801007,
154
+ "grad_norm": 0.9272230863571167,
155
+ "learning_rate": 9.919014054037836e-05,
156
+ "loss": 0.0527,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.09235936188077246,
161
+ "grad_norm": 1.0262973308563232,
162
+ "learning_rate": 9.903522763105218e-05,
163
+ "loss": 0.051,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.09655751469353484,
168
+ "grad_norm": 0.6636225581169128,
169
+ "learning_rate": 9.886690898010535e-05,
170
+ "loss": 0.0429,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.10075566750629723,
175
+ "grad_norm": 0.6513682007789612,
176
+ "learning_rate": 9.868523060417646e-05,
177
+ "loss": 0.0443,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.10495382031905962,
182
+ "grad_norm": 0.7679399251937866,
183
+ "learning_rate": 9.849024217231935e-05,
184
+ "loss": 0.0406,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.109151973131822,
189
+ "grad_norm": 0.5520310401916504,
190
+ "learning_rate": 9.82819969924244e-05,
191
+ "loss": 0.0345,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.11335012594458438,
196
+ "grad_norm": 0.7136860489845276,
197
+ "learning_rate": 9.806055199664446e-05,
198
+ "loss": 0.0487,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.11754827875734676,
203
+ "grad_norm": 0.5643390417098999,
204
+ "learning_rate": 9.782596772583026e-05,
205
+ "loss": 0.0419,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.12174643157010916,
210
+ "grad_norm": 0.6041691899299622,
211
+ "learning_rate": 9.757830831297914e-05,
212
+ "loss": 0.0479,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.12594458438287154,
217
+ "grad_norm": 0.5313841700553894,
218
+ "learning_rate": 9.731764146570173e-05,
219
+ "loss": 0.037,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.13014273719563393,
224
+ "grad_norm": 0.4854077100753784,
225
+ "learning_rate": 9.704403844771128e-05,
226
+ "loss": 0.0368,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.1343408900083963,
231
+ "grad_norm": 0.44807952642440796,
232
+ "learning_rate": 9.675757405934103e-05,
233
+ "loss": 0.0363,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.1385390428211587,
238
+ "grad_norm": 0.51119464635849,
239
+ "learning_rate": 9.645832661709444e-05,
240
+ "loss": 0.0308,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.14273719563392107,
245
+ "grad_norm": 0.5140641927719116,
246
+ "learning_rate": 9.614637793223425e-05,
247
+ "loss": 0.0316,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.14693534844668346,
252
+ "grad_norm": 0.5101410150527954,
253
+ "learning_rate": 9.582181328841611e-05,
254
+ "loss": 0.0307,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.15113350125944586,
259
+ "grad_norm": 0.6274232864379883,
260
+ "learning_rate": 9.548472141837286e-05,
261
+ "loss": 0.0324,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.15533165407220823,
266
+ "grad_norm": 0.6747106313705444,
267
+ "learning_rate": 9.513519447965595e-05,
268
+ "loss": 0.0353,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.15952980688497062,
273
+ "grad_norm": 0.38412219285964966,
274
+ "learning_rate": 9.477332802944044e-05,
275
+ "loss": 0.0311,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.163727959697733,
280
+ "grad_norm": 0.3586186468601227,
281
+ "learning_rate": 9.439922099840054e-05,
282
+ "loss": 0.032,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.16792611251049538,
287
+ "grad_norm": 0.38126131892204285,
288
+ "learning_rate": 9.401297566366318e-05,
289
+ "loss": 0.0266,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.17212426532325778,
294
+ "grad_norm": 0.3684011995792389,
295
+ "learning_rate": 9.36146976208462e-05,
296
+ "loss": 0.0375,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.17632241813602015,
301
+ "grad_norm": 0.5006753206253052,
302
+ "learning_rate": 9.320449575518972e-05,
303
+ "loss": 0.028,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.18052057094878254,
308
+ "grad_norm": 0.5954798460006714,
309
+ "learning_rate": 9.278248221178798e-05,
310
+ "loss": 0.0329,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.1847187237615449,
315
+ "grad_norm": 0.745754063129425,
316
+ "learning_rate": 9.234877236492997e-05,
317
+ "loss": 0.0339,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.1889168765743073,
322
+ "grad_norm": 0.6667306423187256,
323
+ "learning_rate": 9.190348478655724e-05,
324
+ "loss": 0.0316,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.19311502938706968,
329
+ "grad_norm": 0.7456135153770447,
330
+ "learning_rate": 9.144674121384757e-05,
331
+ "loss": 0.0394,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.19731318219983207,
336
+ "grad_norm": 0.4713476002216339,
337
+ "learning_rate": 9.097866651593317e-05,
338
+ "loss": 0.031,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.20151133501259447,
343
+ "grad_norm": 0.9486730098724365,
344
+ "learning_rate": 9.049938865976275e-05,
345
+ "loss": 0.0335,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.20570948782535683,
350
+ "grad_norm": 0.8080450296401978,
351
+ "learning_rate": 9.000903867511666e-05,
352
+ "loss": 0.0326,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.20990764063811923,
357
+ "grad_norm": 0.6074867844581604,
358
+ "learning_rate": 8.950775061878453e-05,
359
+ "loss": 0.0326,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.2141057934508816,
364
+ "grad_norm": 0.4722858965396881,
365
+ "learning_rate": 8.899566153791566e-05,
366
+ "loss": 0.0362,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.218303946263644,
371
+ "grad_norm": 0.3163154721260071,
372
+ "learning_rate": 8.84729114325516e-05,
373
+ "loss": 0.0312,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.2225020990764064,
378
+ "grad_norm": 0.45854002237319946,
379
+ "learning_rate": 8.79396432173515e-05,
380
+ "loss": 0.0319,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.22670025188916876,
385
+ "grad_norm": 0.5331682562828064,
386
+ "learning_rate": 8.739600268252078e-05,
387
+ "loss": 0.0302,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.23089840470193115,
392
+ "grad_norm": 0.44824129343032837,
393
+ "learning_rate": 8.684213845395339e-05,
394
+ "loss": 0.0256,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.23509655751469352,
399
+ "grad_norm": 0.38797256350517273,
400
+ "learning_rate": 8.627820195259918e-05,
401
+ "loss": 0.024,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.23929471032745592,
406
+ "grad_norm": 0.44604647159576416,
407
+ "learning_rate": 8.570434735306671e-05,
408
+ "loss": 0.0265,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.2434928631402183,
413
+ "grad_norm": 0.46292269229888916,
414
+ "learning_rate": 8.512073154147362e-05,
415
+ "loss": 0.0238,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.24769101595298068,
420
+ "grad_norm": 0.6517325043678284,
421
+ "learning_rate": 8.452751407255541e-05,
422
+ "loss": 0.0283,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.2518891687657431,
427
+ "grad_norm": 0.3638853430747986,
428
+ "learning_rate": 8.392485712604483e-05,
429
+ "loss": 0.0269,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.25608732157850544,
434
+ "grad_norm": 0.3911242187023163,
435
+ "learning_rate": 8.331292546233362e-05,
436
+ "loss": 0.0298,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.26028547439126787,
441
+ "grad_norm": 0.4468225836753845,
442
+ "learning_rate": 8.269188637742846e-05,
443
+ "loss": 0.0235,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.26448362720403024,
448
+ "grad_norm": 0.4399245083332062,
449
+ "learning_rate": 8.206190965721419e-05,
450
+ "loss": 0.0217,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.2686817800167926,
455
+ "grad_norm": 0.34588077664375305,
456
+ "learning_rate": 8.14231675310358e-05,
457
+ "loss": 0.0212,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.27287993282955497,
462
+ "grad_norm": 0.3927046060562134,
463
+ "learning_rate": 8.077583462461283e-05,
464
+ "loss": 0.0228,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.2770780856423174,
469
+ "grad_norm": 0.42579185962677,
470
+ "learning_rate": 8.012008791229826e-05,
471
+ "loss": 0.0239,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.28127623845507976,
476
+ "grad_norm": 0.37947288155555725,
477
+ "learning_rate": 7.945610666869568e-05,
478
+ "loss": 0.0302,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.28547439126784213,
483
+ "grad_norm": 0.46608966588974,
484
+ "learning_rate": 7.878407241964729e-05,
485
+ "loss": 0.0269,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.28967254408060455,
490
+ "grad_norm": 0.4380934536457062,
491
+ "learning_rate": 7.810416889260653e-05,
492
+ "loss": 0.0259,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.2938706968933669,
497
+ "grad_norm": 0.4817158281803131,
498
+ "learning_rate": 7.741658196640892e-05,
499
+ "loss": 0.0246,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.2980688497061293,
504
+ "grad_norm": 0.3391236960887909,
505
+ "learning_rate": 7.672149962045457e-05,
506
+ "loss": 0.0285,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.3022670025188917,
511
+ "grad_norm": 0.5921253561973572,
512
+ "learning_rate": 7.60191118833165e-05,
513
+ "loss": 0.0192,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.3064651553316541,
518
+ "grad_norm": 0.3677942156791687,
519
+ "learning_rate": 7.530961078078873e-05,
520
+ "loss": 0.0259,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.31066330814441645,
525
+ "grad_norm": 0.4081045985221863,
526
+ "learning_rate": 7.45931902833884e-05,
527
+ "loss": 0.025,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.3148614609571788,
532
+ "grad_norm": 0.4675191342830658,
533
+ "learning_rate": 7.387004625332608e-05,
534
+ "loss": 0.0223,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.31905961376994124,
539
+ "grad_norm": 0.40352970361709595,
540
+ "learning_rate": 7.3140376390959e-05,
541
+ "loss": 0.0177,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.3232577665827036,
546
+ "grad_norm": 0.4271659851074219,
547
+ "learning_rate": 7.240438018074189e-05,
548
+ "loss": 0.0187,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.327455919395466,
553
+ "grad_norm": 0.4077198803424835,
554
+ "learning_rate": 7.166225883668969e-05,
555
+ "loss": 0.0198,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.3316540722082284,
560
+ "grad_norm": 0.24861399829387665,
561
+ "learning_rate": 7.091421524736784e-05,
562
+ "loss": 0.0182,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.33585222502099077,
567
+ "grad_norm": 0.3340529501438141,
568
+ "learning_rate": 7.016045392042452e-05,
569
+ "loss": 0.0195,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.34005037783375314,
574
+ "grad_norm": 0.30124542117118835,
575
+ "learning_rate": 6.940118092668022e-05,
576
+ "loss": 0.0202,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.34424853064651556,
581
+ "grad_norm": 0.28010880947113037,
582
+ "learning_rate": 6.863660384379017e-05,
583
+ "loss": 0.0192,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.34844668345927793,
588
+ "grad_norm": 0.34497565031051636,
589
+ "learning_rate": 6.786693169949455e-05,
590
+ "loss": 0.0188,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.3526448362720403,
595
+ "grad_norm": 0.38020557165145874,
596
+ "learning_rate": 6.709237491447249e-05,
597
+ "loss": 0.0159,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.35684298908480266,
602
+ "grad_norm": 0.677865207195282,
603
+ "learning_rate": 6.631314524481513e-05,
604
+ "loss": 0.0214,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.3610411418975651,
609
+ "grad_norm": 0.741203784942627,
610
+ "learning_rate": 6.552945572413358e-05,
611
+ "loss": 0.0203,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.36523929471032746,
616
+ "grad_norm": 0.35501229763031006,
617
+ "learning_rate": 6.474152060531768e-05,
618
+ "loss": 0.0193,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.3694374475230898,
623
+ "grad_norm": 0.3309725522994995,
624
+ "learning_rate": 6.394955530196147e-05,
625
+ "loss": 0.0152,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.37363560033585225,
630
+ "grad_norm": 0.36004316806793213,
631
+ "learning_rate": 6.315377632947115e-05,
632
+ "loss": 0.0151,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.3778337531486146,
637
+ "grad_norm": 0.35332006216049194,
638
+ "learning_rate": 6.235440124587198e-05,
639
+ "loss": 0.0174,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.382031905961377,
644
+ "grad_norm": 0.3157385587692261,
645
+ "learning_rate": 6.155164859233012e-05,
646
+ "loss": 0.0125,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.38623005877413935,
651
+ "grad_norm": 0.2917327284812927,
652
+ "learning_rate": 6.074573783340562e-05,
653
+ "loss": 0.0185,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.3904282115869018,
658
+ "grad_norm": 0.20623797178268433,
659
+ "learning_rate": 5.9936889297052986e-05,
660
+ "loss": 0.0141,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.39462636439966414,
665
+ "grad_norm": 0.30338916182518005,
666
+ "learning_rate": 5.912532411438576e-05,
667
+ "loss": 0.0124,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.3988245172124265,
672
+ "grad_norm": 0.46562105417251587,
673
+ "learning_rate": 5.831126415922148e-05,
674
+ "loss": 0.0187,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.40302267002518893,
679
+ "grad_norm": 0.303715318441391,
680
+ "learning_rate": 5.74949319874235e-05,
681
+ "loss": 0.013,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.4072208228379513,
686
+ "grad_norm": 0.33936807513237,
687
+ "learning_rate": 5.667655077605659e-05,
688
+ "loss": 0.0147,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.41141897565071367,
693
+ "grad_norm": 0.26057168841362,
694
+ "learning_rate": 5.585634426237246e-05,
695
+ "loss": 0.0147,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.4156171284634761,
700
+ "grad_norm": 0.30878400802612305,
701
+ "learning_rate": 5.5034536682642224e-05,
702
+ "loss": 0.0208,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.41981528127623846,
707
+ "grad_norm": 0.32619133591651917,
708
+ "learning_rate": 5.4211352710852495e-05,
709
+ "loss": 0.0147,
710
+ "step": 1000
711
+ }
712
+ ],
713
+ "logging_steps": 10,
714
+ "max_steps": 2000,
715
+ "num_input_tokens_seen": 0,
716
+ "num_train_epochs": 1,
717
+ "save_steps": 1000,
718
+ "stateful_callbacks": {
719
+ "TrainerControl": {
720
+ "args": {
721
+ "should_epoch_stop": false,
722
+ "should_evaluate": false,
723
+ "should_log": false,
724
+ "should_save": true,
725
+ "should_training_stop": false
726
+ },
727
+ "attributes": {}
728
+ }
729
+ },
730
+ "total_flos": 0.0,
731
+ "train_batch_size": 32,
732
+ "trial_name": null,
733
+ "trial_params": null
734
+ }
checkpoint-2000/config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "action_dim": 32,
3
+ "action_head_cfg": {
4
+ "action_dim": 32,
5
+ "action_horizon": 16,
6
+ "add_pos_embed": true,
7
+ "backbone_embedding_dim": 2048,
8
+ "diffusion_model_cfg": {
9
+ "attention_head_dim": 48,
10
+ "cross_attention_dim": 2048,
11
+ "dropout": 0.2,
12
+ "final_dropout": true,
13
+ "interleave_self_attention": true,
14
+ "norm_type": "ada_norm",
15
+ "num_attention_heads": 32,
16
+ "num_layers": 16,
17
+ "output_dim": 1024,
18
+ "positional_embeddings": null
19
+ },
20
+ "hidden_size": 1024,
21
+ "input_embedding_dim": 1536,
22
+ "max_action_dim": 32,
23
+ "max_state_dim": 64,
24
+ "model_dtype": "float32",
25
+ "noise_beta_alpha": 1.5,
26
+ "noise_beta_beta": 1.0,
27
+ "noise_s": 0.999,
28
+ "num_inference_timesteps": 4,
29
+ "num_target_vision_tokens": 32,
30
+ "num_timestep_buckets": 1000,
31
+ "tune_diffusion_model": true,
32
+ "tune_projector": true,
33
+ "use_vlln": true,
34
+ "vl_self_attention_cfg": {
35
+ "attention_head_dim": 64,
36
+ "dropout": 0.2,
37
+ "final_dropout": true,
38
+ "num_attention_heads": 32,
39
+ "num_layers": 4,
40
+ "positional_embeddings": null
41
+ }
42
+ },
43
+ "action_horizon": 16,
44
+ "architectures": [
45
+ "GR00T_N1_5"
46
+ ],
47
+ "attn_implementation": null,
48
+ "backbone_cfg": {
49
+ "eagle_path": "NVEagle/eagle_er-qwen3_1_7B-Siglip2_400M_stage1_5_128gpu_er_v7_1mlp_nops",
50
+ "load_bf16": false,
51
+ "project_to_dim": null,
52
+ "reproject_vision": false,
53
+ "select_layer": 12,
54
+ "tune_llm": false,
55
+ "tune_visual": true,
56
+ "use_flash_attention": true
57
+ },
58
+ "compute_dtype": "bfloat16",
59
+ "hidden_size": 2048,
60
+ "model_dtype": "float32",
61
+ "model_type": "gr00t_n1_5",
62
+ "torch_dtype": "bfloat16",
63
+ "transformers_version": "4.51.3"
64
+ }
checkpoint-2000/experiment_cfg/metadata.json ADDED
@@ -0,0 +1,195 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "new_embodiment": {
3
+ "statistics": {
4
+ "state": {
5
+ "single_arm": {
6
+ "max": [
7
+ 2.0000243186950684,
8
+ 2.9169764518737793,
9
+ -0.038691196590662,
10
+ 1.2000755071640015,
11
+ 1.5494728088378906
12
+ ],
13
+ "min": [
14
+ -1.6060552597045898,
15
+ -0.0013021699851378798,
16
+ -2.8603312969207764,
17
+ -0.24804846942424774,
18
+ -2.4647815227508545
19
+ ],
20
+ "mean": [
21
+ 0.5791597962379456,
22
+ 1.1458725929260254,
23
+ -1.20786452293396,
24
+ 0.7232474684715271,
25
+ -0.6134248375892639
26
+ ],
27
+ "std": [
28
+ 0.9811665415763855,
29
+ 0.7337602376937866,
30
+ 0.6142615079879761,
31
+ 0.23586146533489227,
32
+ 0.966107964515686
33
+ ],
34
+ "q01": [
35
+ -1.4408677732944488,
36
+ -1.425088475116354e-05,
37
+ -2.781974565982819,
38
+ 0.27418837130069734,
39
+ -1.9999565029144286
40
+ ],
41
+ "q99": [
42
+ 1.9999998807907104,
43
+ 2.6177230525016784,
44
+ -0.3294296935200691,
45
+ 1.2000004053115845,
46
+ 1.4374910080432892
47
+ ]
48
+ },
49
+ "gripper": {
50
+ "max": [
51
+ 0.7455036640167236
52
+ ],
53
+ "min": [
54
+ -0.09471277892589569
55
+ ],
56
+ "mean": [
57
+ 0.000568521732930094
58
+ ],
59
+ "std": [
60
+ 0.14351940155029297
61
+ ],
62
+ "q01": [
63
+ -0.05334045834839344
64
+ ],
65
+ "q99": [
66
+ 0.5978925746679306
67
+ ]
68
+ }
69
+ },
70
+ "action": {
71
+ "single_arm": {
72
+ "max": [
73
+ 2.0000085830688477,
74
+ 2.9173049926757812,
75
+ -0.0662246122956276,
76
+ 1.2000755071640015,
77
+ 1.549481987953186
78
+ ],
79
+ "min": [
80
+ -1.5781737565994263,
81
+ -0.0013021699851378798,
82
+ -2.8603312969207764,
83
+ -0.24804846942424774,
84
+ -2.4347522258758545
85
+ ],
86
+ "mean": [
87
+ 0.5763245820999146,
88
+ 1.1609699726104736,
89
+ -1.2109930515289307,
90
+ 0.7210843563079834,
91
+ -0.6085058450698853
92
+ ],
93
+ "std": [
94
+ 0.9860737323760986,
95
+ 0.7361332774162292,
96
+ 0.6108555197715759,
97
+ 0.23312774300575256,
98
+ 0.9712501168251038
99
+ ],
100
+ "q01": [
101
+ -1.4462056386470794,
102
+ -1.4080941818974679e-05,
103
+ -2.7542264652252197,
104
+ 0.27880221664905547,
105
+ -1.9997829270362855
106
+ ],
107
+ "q99": [
108
+ 1.9999998807907104,
109
+ 2.6269060873985284,
110
+ -0.33764528483152395,
111
+ 1.200000286102295,
112
+ 1.4436372423171995
113
+ ]
114
+ },
115
+ "gripper": {
116
+ "max": [
117
+ 0.7455036640167236
118
+ ],
119
+ "min": [
120
+ -0.07934218645095825
121
+ ],
122
+ "mean": [
123
+ -0.0036394144408404827
124
+ ],
125
+ "std": [
126
+ 0.13472281396389008
127
+ ],
128
+ "q01": [
129
+ -0.05393668111413717
130
+ ],
131
+ "q99": [
132
+ 0.56789559841156
133
+ ]
134
+ }
135
+ }
136
+ },
137
+ "modalities": {
138
+ "video": {
139
+ "front": {
140
+ "resolution": [
141
+ 640,
142
+ 480
143
+ ],
144
+ "channels": 3,
145
+ "fps": 30.0
146
+ },
147
+ "wrist": {
148
+ "resolution": [
149
+ 640,
150
+ 480
151
+ ],
152
+ "channels": 3,
153
+ "fps": 30.0
154
+ }
155
+ },
156
+ "state": {
157
+ "single_arm": {
158
+ "absolute": true,
159
+ "rotation_type": null,
160
+ "shape": [
161
+ 5
162
+ ],
163
+ "continuous": true
164
+ },
165
+ "gripper": {
166
+ "absolute": true,
167
+ "rotation_type": null,
168
+ "shape": [
169
+ 1
170
+ ],
171
+ "continuous": true
172
+ }
173
+ },
174
+ "action": {
175
+ "single_arm": {
176
+ "absolute": true,
177
+ "rotation_type": null,
178
+ "shape": [
179
+ 5
180
+ ],
181
+ "continuous": true
182
+ },
183
+ "gripper": {
184
+ "absolute": true,
185
+ "rotation_type": null,
186
+ "shape": [
187
+ 1
188
+ ],
189
+ "continuous": true
190
+ }
191
+ }
192
+ },
193
+ "embodiment_tag": "new_embodiment"
194
+ }
195
+ }
checkpoint-2000/model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c90564297b150a222847e1a1cb3b76645f5a645d14c88a22cda9b63a692768e8
3
+ size 4999367032
checkpoint-2000/model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d32a802c118a4bfcc906e341d99cc23fe780b40988d7b776b59df927e1eb40b
3
+ size 2586705312
checkpoint-2000/model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bd0d9fd46106b70ec5a3718011c3342322be8b28029d1424127b5774779ef87
3
+ size 8550720062
checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4da91678d4aaa7f85f545f4fa01e6053bf55c809b86f5c232e944791ea0af641
3
+ size 14244
checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6a1e3232c9dd0558d3b582345b4a537ca2d63d9ae2df7d7189cbdc92c4e384f
3
+ size 1064
checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,1434 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.8396305625524769,
6
+ "eval_steps": 500,
7
+ "global_step": 2000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0041981528127623844,
14
+ "grad_norm": 5.267488956451416,
15
+ "learning_rate": 9e-06,
16
+ "loss": 1.0193,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.008396305625524769,
21
+ "grad_norm": 2.3370044231414795,
22
+ "learning_rate": 1.9e-05,
23
+ "loss": 0.3362,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.012594458438287154,
28
+ "grad_norm": 2.1865203380584717,
29
+ "learning_rate": 2.9e-05,
30
+ "loss": 0.2021,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.016792611251049538,
35
+ "grad_norm": 1.625726342201233,
36
+ "learning_rate": 3.9000000000000006e-05,
37
+ "loss": 0.1466,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.020990764063811923,
42
+ "grad_norm": 1.1913143396377563,
43
+ "learning_rate": 4.9e-05,
44
+ "loss": 0.1184,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.02518891687657431,
49
+ "grad_norm": 1.6674305200576782,
50
+ "learning_rate": 5.9e-05,
51
+ "loss": 0.1028,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.02938706968933669,
56
+ "grad_norm": 0.8989318013191223,
57
+ "learning_rate": 6.9e-05,
58
+ "loss": 0.0836,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.033585222502099076,
63
+ "grad_norm": 0.8966989517211914,
64
+ "learning_rate": 7.900000000000001e-05,
65
+ "loss": 0.0851,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.037783375314861464,
70
+ "grad_norm": 0.7644183039665222,
71
+ "learning_rate": 8.900000000000001e-05,
72
+ "loss": 0.0717,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.041981528127623846,
77
+ "grad_norm": 1.272506833076477,
78
+ "learning_rate": 9.900000000000001e-05,
79
+ "loss": 0.0774,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.04617968094038623,
84
+ "grad_norm": 1.4885015487670898,
85
+ "learning_rate": 9.999446382823013e-05,
86
+ "loss": 0.0712,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.05037783375314862,
91
+ "grad_norm": 1.201401710510254,
92
+ "learning_rate": 9.997532801828658e-05,
93
+ "loss": 0.072,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.054575986565911,
98
+ "grad_norm": 1.367532730102539,
99
+ "learning_rate": 9.99425294526634e-05,
100
+ "loss": 0.0682,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.05877413937867338,
105
+ "grad_norm": 0.9127843976020813,
106
+ "learning_rate": 9.989607709816091e-05,
107
+ "loss": 0.0659,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.06297229219143577,
112
+ "grad_norm": 0.7713141441345215,
113
+ "learning_rate": 9.983598365438902e-05,
114
+ "loss": 0.0528,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.06717044500419815,
119
+ "grad_norm": 0.9186705946922302,
120
+ "learning_rate": 9.976226555029522e-05,
121
+ "loss": 0.0495,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.07136859781696053,
126
+ "grad_norm": 0.5967724919319153,
127
+ "learning_rate": 9.967494293967312e-05,
128
+ "loss": 0.0522,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.07556675062972293,
133
+ "grad_norm": 1.0857634544372559,
134
+ "learning_rate": 9.95740396956525e-05,
135
+ "loss": 0.0578,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.07976490344248531,
140
+ "grad_norm": 0.5992318987846375,
141
+ "learning_rate": 9.945958340417283e-05,
142
+ "loss": 0.0478,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.08396305625524769,
147
+ "grad_norm": 0.5689753293991089,
148
+ "learning_rate": 9.93316053564413e-05,
149
+ "loss": 0.0478,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.08816120906801007,
154
+ "grad_norm": 0.9272230863571167,
155
+ "learning_rate": 9.919014054037836e-05,
156
+ "loss": 0.0527,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.09235936188077246,
161
+ "grad_norm": 1.0262973308563232,
162
+ "learning_rate": 9.903522763105218e-05,
163
+ "loss": 0.051,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.09655751469353484,
168
+ "grad_norm": 0.6636225581169128,
169
+ "learning_rate": 9.886690898010535e-05,
170
+ "loss": 0.0429,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.10075566750629723,
175
+ "grad_norm": 0.6513682007789612,
176
+ "learning_rate": 9.868523060417646e-05,
177
+ "loss": 0.0443,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.10495382031905962,
182
+ "grad_norm": 0.7679399251937866,
183
+ "learning_rate": 9.849024217231935e-05,
184
+ "loss": 0.0406,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.109151973131822,
189
+ "grad_norm": 0.5520310401916504,
190
+ "learning_rate": 9.82819969924244e-05,
191
+ "loss": 0.0345,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.11335012594458438,
196
+ "grad_norm": 0.7136860489845276,
197
+ "learning_rate": 9.806055199664446e-05,
198
+ "loss": 0.0487,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.11754827875734676,
203
+ "grad_norm": 0.5643390417098999,
204
+ "learning_rate": 9.782596772583026e-05,
205
+ "loss": 0.0419,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.12174643157010916,
210
+ "grad_norm": 0.6041691899299622,
211
+ "learning_rate": 9.757830831297914e-05,
212
+ "loss": 0.0479,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.12594458438287154,
217
+ "grad_norm": 0.5313841700553894,
218
+ "learning_rate": 9.731764146570173e-05,
219
+ "loss": 0.037,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.13014273719563393,
224
+ "grad_norm": 0.4854077100753784,
225
+ "learning_rate": 9.704403844771128e-05,
226
+ "loss": 0.0368,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.1343408900083963,
231
+ "grad_norm": 0.44807952642440796,
232
+ "learning_rate": 9.675757405934103e-05,
233
+ "loss": 0.0363,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.1385390428211587,
238
+ "grad_norm": 0.51119464635849,
239
+ "learning_rate": 9.645832661709444e-05,
240
+ "loss": 0.0308,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.14273719563392107,
245
+ "grad_norm": 0.5140641927719116,
246
+ "learning_rate": 9.614637793223425e-05,
247
+ "loss": 0.0316,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.14693534844668346,
252
+ "grad_norm": 0.5101410150527954,
253
+ "learning_rate": 9.582181328841611e-05,
254
+ "loss": 0.0307,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.15113350125944586,
259
+ "grad_norm": 0.6274232864379883,
260
+ "learning_rate": 9.548472141837286e-05,
261
+ "loss": 0.0324,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.15533165407220823,
266
+ "grad_norm": 0.6747106313705444,
267
+ "learning_rate": 9.513519447965595e-05,
268
+ "loss": 0.0353,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.15952980688497062,
273
+ "grad_norm": 0.38412219285964966,
274
+ "learning_rate": 9.477332802944044e-05,
275
+ "loss": 0.0311,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.163727959697733,
280
+ "grad_norm": 0.3586186468601227,
281
+ "learning_rate": 9.439922099840054e-05,
282
+ "loss": 0.032,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.16792611251049538,
287
+ "grad_norm": 0.38126131892204285,
288
+ "learning_rate": 9.401297566366318e-05,
289
+ "loss": 0.0266,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.17212426532325778,
294
+ "grad_norm": 0.3684011995792389,
295
+ "learning_rate": 9.36146976208462e-05,
296
+ "loss": 0.0375,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.17632241813602015,
301
+ "grad_norm": 0.5006753206253052,
302
+ "learning_rate": 9.320449575518972e-05,
303
+ "loss": 0.028,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.18052057094878254,
308
+ "grad_norm": 0.5954798460006714,
309
+ "learning_rate": 9.278248221178798e-05,
310
+ "loss": 0.0329,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.1847187237615449,
315
+ "grad_norm": 0.745754063129425,
316
+ "learning_rate": 9.234877236492997e-05,
317
+ "loss": 0.0339,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.1889168765743073,
322
+ "grad_norm": 0.6667306423187256,
323
+ "learning_rate": 9.190348478655724e-05,
324
+ "loss": 0.0316,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.19311502938706968,
329
+ "grad_norm": 0.7456135153770447,
330
+ "learning_rate": 9.144674121384757e-05,
331
+ "loss": 0.0394,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.19731318219983207,
336
+ "grad_norm": 0.4713476002216339,
337
+ "learning_rate": 9.097866651593317e-05,
338
+ "loss": 0.031,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.20151133501259447,
343
+ "grad_norm": 0.9486730098724365,
344
+ "learning_rate": 9.049938865976275e-05,
345
+ "loss": 0.0335,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.20570948782535683,
350
+ "grad_norm": 0.8080450296401978,
351
+ "learning_rate": 9.000903867511666e-05,
352
+ "loss": 0.0326,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.20990764063811923,
357
+ "grad_norm": 0.6074867844581604,
358
+ "learning_rate": 8.950775061878453e-05,
359
+ "loss": 0.0326,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.2141057934508816,
364
+ "grad_norm": 0.4722858965396881,
365
+ "learning_rate": 8.899566153791566e-05,
366
+ "loss": 0.0362,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.218303946263644,
371
+ "grad_norm": 0.3163154721260071,
372
+ "learning_rate": 8.84729114325516e-05,
373
+ "loss": 0.0312,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.2225020990764064,
378
+ "grad_norm": 0.45854002237319946,
379
+ "learning_rate": 8.79396432173515e-05,
380
+ "loss": 0.0319,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.22670025188916876,
385
+ "grad_norm": 0.5331682562828064,
386
+ "learning_rate": 8.739600268252078e-05,
387
+ "loss": 0.0302,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.23089840470193115,
392
+ "grad_norm": 0.44824129343032837,
393
+ "learning_rate": 8.684213845395339e-05,
394
+ "loss": 0.0256,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.23509655751469352,
399
+ "grad_norm": 0.38797256350517273,
400
+ "learning_rate": 8.627820195259918e-05,
401
+ "loss": 0.024,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.23929471032745592,
406
+ "grad_norm": 0.44604647159576416,
407
+ "learning_rate": 8.570434735306671e-05,
408
+ "loss": 0.0265,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.2434928631402183,
413
+ "grad_norm": 0.46292269229888916,
414
+ "learning_rate": 8.512073154147362e-05,
415
+ "loss": 0.0238,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.24769101595298068,
420
+ "grad_norm": 0.6517325043678284,
421
+ "learning_rate": 8.452751407255541e-05,
422
+ "loss": 0.0283,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.2518891687657431,
427
+ "grad_norm": 0.3638853430747986,
428
+ "learning_rate": 8.392485712604483e-05,
429
+ "loss": 0.0269,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.25608732157850544,
434
+ "grad_norm": 0.3911242187023163,
435
+ "learning_rate": 8.331292546233362e-05,
436
+ "loss": 0.0298,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.26028547439126787,
441
+ "grad_norm": 0.4468225836753845,
442
+ "learning_rate": 8.269188637742846e-05,
443
+ "loss": 0.0235,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.26448362720403024,
448
+ "grad_norm": 0.4399245083332062,
449
+ "learning_rate": 8.206190965721419e-05,
450
+ "loss": 0.0217,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.2686817800167926,
455
+ "grad_norm": 0.34588077664375305,
456
+ "learning_rate": 8.14231675310358e-05,
457
+ "loss": 0.0212,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.27287993282955497,
462
+ "grad_norm": 0.3927046060562134,
463
+ "learning_rate": 8.077583462461283e-05,
464
+ "loss": 0.0228,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.2770780856423174,
469
+ "grad_norm": 0.42579185962677,
470
+ "learning_rate": 8.012008791229826e-05,
471
+ "loss": 0.0239,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.28127623845507976,
476
+ "grad_norm": 0.37947288155555725,
477
+ "learning_rate": 7.945610666869568e-05,
478
+ "loss": 0.0302,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.28547439126784213,
483
+ "grad_norm": 0.46608966588974,
484
+ "learning_rate": 7.878407241964729e-05,
485
+ "loss": 0.0269,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.28967254408060455,
490
+ "grad_norm": 0.4380934536457062,
491
+ "learning_rate": 7.810416889260653e-05,
492
+ "loss": 0.0259,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.2938706968933669,
497
+ "grad_norm": 0.4817158281803131,
498
+ "learning_rate": 7.741658196640892e-05,
499
+ "loss": 0.0246,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.2980688497061293,
504
+ "grad_norm": 0.3391236960887909,
505
+ "learning_rate": 7.672149962045457e-05,
506
+ "loss": 0.0285,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.3022670025188917,
511
+ "grad_norm": 0.5921253561973572,
512
+ "learning_rate": 7.60191118833165e-05,
513
+ "loss": 0.0192,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.3064651553316541,
518
+ "grad_norm": 0.3677942156791687,
519
+ "learning_rate": 7.530961078078873e-05,
520
+ "loss": 0.0259,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.31066330814441645,
525
+ "grad_norm": 0.4081045985221863,
526
+ "learning_rate": 7.45931902833884e-05,
527
+ "loss": 0.025,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.3148614609571788,
532
+ "grad_norm": 0.4675191342830658,
533
+ "learning_rate": 7.387004625332608e-05,
534
+ "loss": 0.0223,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.31905961376994124,
539
+ "grad_norm": 0.40352970361709595,
540
+ "learning_rate": 7.3140376390959e-05,
541
+ "loss": 0.0177,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.3232577665827036,
546
+ "grad_norm": 0.4271659851074219,
547
+ "learning_rate": 7.240438018074189e-05,
548
+ "loss": 0.0187,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.327455919395466,
553
+ "grad_norm": 0.4077198803424835,
554
+ "learning_rate": 7.166225883668969e-05,
555
+ "loss": 0.0198,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.3316540722082284,
560
+ "grad_norm": 0.24861399829387665,
561
+ "learning_rate": 7.091421524736784e-05,
562
+ "loss": 0.0182,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.33585222502099077,
567
+ "grad_norm": 0.3340529501438141,
568
+ "learning_rate": 7.016045392042452e-05,
569
+ "loss": 0.0195,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.34005037783375314,
574
+ "grad_norm": 0.30124542117118835,
575
+ "learning_rate": 6.940118092668022e-05,
576
+ "loss": 0.0202,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.34424853064651556,
581
+ "grad_norm": 0.28010880947113037,
582
+ "learning_rate": 6.863660384379017e-05,
583
+ "loss": 0.0192,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.34844668345927793,
588
+ "grad_norm": 0.34497565031051636,
589
+ "learning_rate": 6.786693169949455e-05,
590
+ "loss": 0.0188,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.3526448362720403,
595
+ "grad_norm": 0.38020557165145874,
596
+ "learning_rate": 6.709237491447249e-05,
597
+ "loss": 0.0159,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.35684298908480266,
602
+ "grad_norm": 0.677865207195282,
603
+ "learning_rate": 6.631314524481513e-05,
604
+ "loss": 0.0214,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.3610411418975651,
609
+ "grad_norm": 0.741203784942627,
610
+ "learning_rate": 6.552945572413358e-05,
611
+ "loss": 0.0203,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.36523929471032746,
616
+ "grad_norm": 0.35501229763031006,
617
+ "learning_rate": 6.474152060531768e-05,
618
+ "loss": 0.0193,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.3694374475230898,
623
+ "grad_norm": 0.3309725522994995,
624
+ "learning_rate": 6.394955530196147e-05,
625
+ "loss": 0.0152,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.37363560033585225,
630
+ "grad_norm": 0.36004316806793213,
631
+ "learning_rate": 6.315377632947115e-05,
632
+ "loss": 0.0151,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.3778337531486146,
637
+ "grad_norm": 0.35332006216049194,
638
+ "learning_rate": 6.235440124587198e-05,
639
+ "loss": 0.0174,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.382031905961377,
644
+ "grad_norm": 0.3157385587692261,
645
+ "learning_rate": 6.155164859233012e-05,
646
+ "loss": 0.0125,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.38623005877413935,
651
+ "grad_norm": 0.2917327284812927,
652
+ "learning_rate": 6.074573783340562e-05,
653
+ "loss": 0.0185,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.3904282115869018,
658
+ "grad_norm": 0.20623797178268433,
659
+ "learning_rate": 5.9936889297052986e-05,
660
+ "loss": 0.0141,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.39462636439966414,
665
+ "grad_norm": 0.30338916182518005,
666
+ "learning_rate": 5.912532411438576e-05,
667
+ "loss": 0.0124,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.3988245172124265,
672
+ "grad_norm": 0.46562105417251587,
673
+ "learning_rate": 5.831126415922148e-05,
674
+ "loss": 0.0187,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.40302267002518893,
679
+ "grad_norm": 0.303715318441391,
680
+ "learning_rate": 5.74949319874235e-05,
681
+ "loss": 0.013,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.4072208228379513,
686
+ "grad_norm": 0.33936807513237,
687
+ "learning_rate": 5.667655077605659e-05,
688
+ "loss": 0.0147,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.41141897565071367,
693
+ "grad_norm": 0.26057168841362,
694
+ "learning_rate": 5.585634426237246e-05,
695
+ "loss": 0.0147,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.4156171284634761,
700
+ "grad_norm": 0.30878400802612305,
701
+ "learning_rate": 5.5034536682642224e-05,
702
+ "loss": 0.0208,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.41981528127623846,
707
+ "grad_norm": 0.32619133591651917,
708
+ "learning_rate": 5.4211352710852495e-05,
709
+ "loss": 0.0147,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.42401343408900083,
714
+ "grad_norm": 0.4049901068210602,
715
+ "learning_rate": 5.3387017397281704e-05,
716
+ "loss": 0.0136,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.4282115869017632,
721
+ "grad_norm": 0.263507217168808,
722
+ "learning_rate": 5.2561756106973656e-05,
723
+ "loss": 0.0142,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.4324097397145256,
728
+ "grad_norm": 0.3729311525821686,
729
+ "learning_rate": 5.1735794458124956e-05,
730
+ "loss": 0.0143,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.436607892527288,
735
+ "grad_norm": 0.433525413274765,
736
+ "learning_rate": 5.0909358260403186e-05,
737
+ "loss": 0.0159,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.44080604534005036,
742
+ "grad_norm": 0.20972338318824768,
743
+ "learning_rate": 5.0082673453212914e-05,
744
+ "loss": 0.0125,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.4450041981528128,
749
+ "grad_norm": 0.44978395104408264,
750
+ "learning_rate": 4.925596604392603e-05,
751
+ "loss": 0.0201,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.44920235096557515,
756
+ "grad_norm": 0.33291032910346985,
757
+ "learning_rate": 4.8429462046093585e-05,
758
+ "loss": 0.0145,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.4534005037783375,
763
+ "grad_norm": 0.3551059663295746,
764
+ "learning_rate": 4.7603387417656026e-05,
765
+ "loss": 0.0164,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.45759865659109994,
770
+ "grad_norm": 0.2902749478816986,
771
+ "learning_rate": 4.677796799916845e-05,
772
+ "loss": 0.0088,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.4617968094038623,
777
+ "grad_norm": 0.4253701865673065,
778
+ "learning_rate": 4.5953429452058135e-05,
779
+ "loss": 0.0149,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.4659949622166247,
784
+ "grad_norm": 0.25408294796943665,
785
+ "learning_rate": 4.5129997196930845e-05,
786
+ "loss": 0.014,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.47019311502938704,
791
+ "grad_norm": 0.3801613450050354,
792
+ "learning_rate": 4.430789635194324e-05,
793
+ "loss": 0.0135,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.47439126784214947,
798
+ "grad_norm": 0.3434455990791321,
799
+ "learning_rate": 4.348735167125771e-05,
800
+ "loss": 0.0175,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.47858942065491183,
805
+ "grad_norm": 0.38031941652297974,
806
+ "learning_rate": 4.2668587483596864e-05,
807
+ "loss": 0.0131,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.4827875734676742,
812
+ "grad_norm": 0.2573353350162506,
813
+ "learning_rate": 4.1851827630914305e-05,
814
+ "loss": 0.0137,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.4869857262804366,
819
+ "grad_norm": 0.31360694766044617,
820
+ "learning_rate": 4.103729540719847e-05,
821
+ "loss": 0.0146,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.491183879093199,
826
+ "grad_norm": 0.32134002447128296,
827
+ "learning_rate": 4.0225213497426276e-05,
828
+ "loss": 0.0133,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.49538203190596136,
833
+ "grad_norm": 0.31462499499320984,
834
+ "learning_rate": 3.9415803916683224e-05,
835
+ "loss": 0.0124,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.4995801847187238,
840
+ "grad_norm": 0.3015807569026947,
841
+ "learning_rate": 3.860928794946682e-05,
842
+ "loss": 0.0147,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.5037783375314862,
847
+ "grad_norm": 0.26400843262672424,
848
+ "learning_rate": 3.780588608918947e-05,
849
+ "loss": 0.0125,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.5079764903442485,
854
+ "grad_norm": 0.27983877062797546,
855
+ "learning_rate": 3.700581797789786e-05,
856
+ "loss": 0.0139,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.5121746431570109,
861
+ "grad_norm": 0.34057655930519104,
862
+ "learning_rate": 3.6209302346225006e-05,
863
+ "loss": 0.0125,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.5163727959697733,
868
+ "grad_norm": 0.3145323097705841,
869
+ "learning_rate": 3.541655695359142e-05,
870
+ "loss": 0.0141,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.5205709487825357,
875
+ "grad_norm": 0.2706829905509949,
876
+ "learning_rate": 3.462779852867197e-05,
877
+ "loss": 0.0112,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.5247691015952981,
882
+ "grad_norm": 0.3848113417625427,
883
+ "learning_rate": 3.384324271014429e-05,
884
+ "loss": 0.0147,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.5289672544080605,
889
+ "grad_norm": 0.32747682929039,
890
+ "learning_rate": 3.3063103987735433e-05,
891
+ "loss": 0.0114,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.5331654072208228,
896
+ "grad_norm": 0.2991146445274353,
897
+ "learning_rate": 3.228759564358248e-05,
898
+ "loss": 0.0141,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.5373635600335852,
903
+ "grad_norm": 0.3067378103733063,
904
+ "learning_rate": 3.1516929693923315e-05,
905
+ "loss": 0.0129,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.5415617128463476,
910
+ "grad_norm": 0.15431927144527435,
911
+ "learning_rate": 3.075131683113352e-05,
912
+ "loss": 0.013,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.5457598656591099,
917
+ "grad_norm": 0.1479986160993576,
918
+ "learning_rate": 2.999096636612518e-05,
919
+ "loss": 0.0121,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.5499580184718724,
924
+ "grad_norm": 0.1798468679189682,
925
+ "learning_rate": 2.9236086171123404e-05,
926
+ "loss": 0.0107,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.5541561712846348,
931
+ "grad_norm": 0.21686813235282898,
932
+ "learning_rate": 2.8486882622836026e-05,
933
+ "loss": 0.0124,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.5583543240973972,
938
+ "grad_norm": 0.40006551146507263,
939
+ "learning_rate": 2.774356054603243e-05,
940
+ "loss": 0.0103,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.5625524769101595,
945
+ "grad_norm": 0.39905092120170593,
946
+ "learning_rate": 2.7006323157546386e-05,
947
+ "loss": 0.0146,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.5667506297229219,
952
+ "grad_norm": 0.156532883644104,
953
+ "learning_rate": 2.6275372010718635e-05,
954
+ "loss": 0.0083,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.5709487825356843,
959
+ "grad_norm": 0.418869286775589,
960
+ "learning_rate": 2.555090694029421e-05,
961
+ "loss": 0.0112,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.5751469353484466,
966
+ "grad_norm": 0.3597627878189087,
967
+ "learning_rate": 2.4833126007789653e-05,
968
+ "loss": 0.0071,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.5793450881612091,
973
+ "grad_norm": 0.3350473940372467,
974
+ "learning_rate": 2.4122225447344875e-05,
975
+ "loss": 0.0094,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.5835432409739715,
980
+ "grad_norm": 0.2572272717952728,
981
+ "learning_rate": 2.341839961207482e-05,
982
+ "loss": 0.0086,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.5877413937867338,
987
+ "grad_norm": 0.44453227519989014,
988
+ "learning_rate": 2.2721840920935196e-05,
989
+ "loss": 0.0104,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.5919395465994962,
994
+ "grad_norm": 0.23089422285556793,
995
+ "learning_rate": 2.2032739806117058e-05,
996
+ "loss": 0.0075,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.5961376994122586,
1001
+ "grad_norm": 0.2570338547229767,
1002
+ "learning_rate": 2.1351284660984572e-05,
1003
+ "loss": 0.0128,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.600335852225021,
1008
+ "grad_norm": 0.4198639690876007,
1009
+ "learning_rate": 2.067766178857013e-05,
1010
+ "loss": 0.0133,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 0.6045340050377834,
1015
+ "grad_norm": 0.4144093990325928,
1016
+ "learning_rate": 2.0012055350640986e-05,
1017
+ "loss": 0.0095,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 0.6087321578505458,
1022
+ "grad_norm": 0.25941231846809387,
1023
+ "learning_rate": 1.9354647317351188e-05,
1024
+ "loss": 0.0105,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 0.6129303106633082,
1029
+ "grad_norm": 0.25724703073501587,
1030
+ "learning_rate": 1.8705617417492883e-05,
1031
+ "loss": 0.0096,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 0.6171284634760705,
1036
+ "grad_norm": 0.2537188231945038,
1037
+ "learning_rate": 1.8065143089360172e-05,
1038
+ "loss": 0.0094,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 0.6213266162888329,
1043
+ "grad_norm": 0.25936058163642883,
1044
+ "learning_rate": 1.743339943223926e-05,
1045
+ "loss": 0.0132,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 0.6255247691015953,
1050
+ "grad_norm": 0.27446359395980835,
1051
+ "learning_rate": 1.6810559158538092e-05,
1052
+ "loss": 0.0084,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 0.6297229219143576,
1057
+ "grad_norm": 0.27379536628723145,
1058
+ "learning_rate": 1.6196792546568472e-05,
1059
+ "loss": 0.0126,
1060
+ "step": 1500
1061
+ },
1062
+ {
1063
+ "epoch": 0.6339210747271201,
1064
+ "grad_norm": 0.16973455250263214,
1065
+ "learning_rate": 1.5592267393993716e-05,
1066
+ "loss": 0.0082,
1067
+ "step": 1510
1068
+ },
1069
+ {
1070
+ "epoch": 0.6381192275398825,
1071
+ "grad_norm": 0.19682660698890686,
1072
+ "learning_rate": 1.4997148971954344e-05,
1073
+ "loss": 0.0119,
1074
+ "step": 1520
1075
+ },
1076
+ {
1077
+ "epoch": 0.6423173803526449,
1078
+ "grad_norm": 0.16340869665145874,
1079
+ "learning_rate": 1.4411599979884744e-05,
1080
+ "loss": 0.0101,
1081
+ "step": 1530
1082
+ },
1083
+ {
1084
+ "epoch": 0.6465155331654072,
1085
+ "grad_norm": 0.27933576703071594,
1086
+ "learning_rate": 1.383578050103268e-05,
1087
+ "loss": 0.0117,
1088
+ "step": 1540
1089
+ },
1090
+ {
1091
+ "epoch": 0.6507136859781696,
1092
+ "grad_norm": 0.10343077033758163,
1093
+ "learning_rate": 1.3269847958694148e-05,
1094
+ "loss": 0.0101,
1095
+ "step": 1550
1096
+ },
1097
+ {
1098
+ "epoch": 0.654911838790932,
1099
+ "grad_norm": 0.15868137776851654,
1100
+ "learning_rate": 1.2713957073175425e-05,
1101
+ "loss": 0.0084,
1102
+ "step": 1560
1103
+ },
1104
+ {
1105
+ "epoch": 0.6591099916036943,
1106
+ "grad_norm": 0.22004792094230652,
1107
+ "learning_rate": 1.2168259819494066e-05,
1108
+ "loss": 0.0087,
1109
+ "step": 1570
1110
+ },
1111
+ {
1112
+ "epoch": 0.6633081444164568,
1113
+ "grad_norm": 0.11181829124689102,
1114
+ "learning_rate": 1.1632905385830484e-05,
1115
+ "loss": 0.0098,
1116
+ "step": 1580
1117
+ },
1118
+ {
1119
+ "epoch": 0.6675062972292192,
1120
+ "grad_norm": 0.1479823738336563,
1121
+ "learning_rate": 1.1108040132741354e-05,
1122
+ "loss": 0.0093,
1123
+ "step": 1590
1124
+ },
1125
+ {
1126
+ "epoch": 0.6717044500419815,
1127
+ "grad_norm": 0.2433646023273468,
1128
+ "learning_rate": 1.059380755314613e-05,
1129
+ "loss": 0.0088,
1130
+ "step": 1600
1131
+ },
1132
+ {
1133
+ "epoch": 0.6759026028547439,
1134
+ "grad_norm": 0.212657168507576,
1135
+ "learning_rate": 1.009034823309749e-05,
1136
+ "loss": 0.0061,
1137
+ "step": 1610
1138
+ },
1139
+ {
1140
+ "epoch": 0.6801007556675063,
1141
+ "grad_norm": 0.10653030127286911,
1142
+ "learning_rate": 9.597799813346525e-06,
1143
+ "loss": 0.0068,
1144
+ "step": 1620
1145
+ },
1146
+ {
1147
+ "epoch": 0.6842989084802686,
1148
+ "grad_norm": 0.10436981171369553,
1149
+ "learning_rate": 9.116296951713133e-06,
1150
+ "loss": 0.0064,
1151
+ "step": 1630
1152
+ },
1153
+ {
1154
+ "epoch": 0.6884970612930311,
1155
+ "grad_norm": 0.28674691915512085,
1156
+ "learning_rate": 8.645971286271904e-06,
1157
+ "loss": 0.0083,
1158
+ "step": 1640
1159
+ },
1160
+ {
1161
+ "epoch": 0.6926952141057935,
1162
+ "grad_norm": 0.2074134796857834,
1163
+ "learning_rate": 8.186951399363613e-06,
1164
+ "loss": 0.0103,
1165
+ "step": 1650
1166
+ },
1167
+ {
1168
+ "epoch": 0.6968933669185559,
1169
+ "grad_norm": 0.160618856549263,
1170
+ "learning_rate": 7.739362782442021e-06,
1171
+ "loss": 0.0051,
1172
+ "step": 1660
1173
+ },
1174
+ {
1175
+ "epoch": 0.7010915197313182,
1176
+ "grad_norm": 0.2147001475095749,
1177
+ "learning_rate": 7.30332780176588e-06,
1178
+ "loss": 0.0085,
1179
+ "step": 1670
1180
+ },
1181
+ {
1182
+ "epoch": 0.7052896725440806,
1183
+ "grad_norm": 0.27866339683532715,
1184
+ "learning_rate": 6.878965664945108e-06,
1185
+ "loss": 0.0089,
1186
+ "step": 1680
1187
+ },
1188
+ {
1189
+ "epoch": 0.709487825356843,
1190
+ "grad_norm": 0.22506844997406006,
1191
+ "learning_rate": 6.466392388350695e-06,
1192
+ "loss": 0.009,
1193
+ "step": 1690
1194
+ },
1195
+ {
1196
+ "epoch": 0.7136859781696053,
1197
+ "grad_norm": 0.2676888406276703,
1198
+ "learning_rate": 6.0657207653969315e-06,
1199
+ "loss": 0.0074,
1200
+ "step": 1700
1201
+ },
1202
+ {
1203
+ "epoch": 0.7178841309823678,
1204
+ "grad_norm": 0.19724851846694946,
1205
+ "learning_rate": 5.67706033570487e-06,
1206
+ "loss": 0.0074,
1207
+ "step": 1710
1208
+ },
1209
+ {
1210
+ "epoch": 0.7220822837951302,
1211
+ "grad_norm": 0.18323613703250885,
1212
+ "learning_rate": 5.300517355155215e-06,
1213
+ "loss": 0.0093,
1214
+ "step": 1720
1215
+ },
1216
+ {
1217
+ "epoch": 0.7262804366078925,
1218
+ "grad_norm": 0.2785090208053589,
1219
+ "learning_rate": 4.936194766839103e-06,
1220
+ "loss": 0.0093,
1221
+ "step": 1730
1222
+ },
1223
+ {
1224
+ "epoch": 0.7304785894206549,
1225
+ "grad_norm": 0.16792871057987213,
1226
+ "learning_rate": 4.5841921729144424e-06,
1227
+ "loss": 0.0063,
1228
+ "step": 1740
1229
+ },
1230
+ {
1231
+ "epoch": 0.7346767422334173,
1232
+ "grad_norm": 0.1223788782954216,
1233
+ "learning_rate": 4.244605807375679e-06,
1234
+ "loss": 0.009,
1235
+ "step": 1750
1236
+ },
1237
+ {
1238
+ "epoch": 0.7388748950461796,
1239
+ "grad_norm": 0.13132202625274658,
1240
+ "learning_rate": 3.917528509744412e-06,
1241
+ "loss": 0.0093,
1242
+ "step": 1760
1243
+ },
1244
+ {
1245
+ "epoch": 0.743073047858942,
1246
+ "grad_norm": 0.2555523216724396,
1247
+ "learning_rate": 3.60304969968796e-06,
1248
+ "loss": 0.0078,
1249
+ "step": 1770
1250
+ },
1251
+ {
1252
+ "epoch": 0.7472712006717045,
1253
+ "grad_norm": 0.16517667472362518,
1254
+ "learning_rate": 3.301255352572946e-06,
1255
+ "loss": 0.0074,
1256
+ "step": 1780
1257
+ },
1258
+ {
1259
+ "epoch": 0.7514693534844669,
1260
+ "grad_norm": 0.1717490404844284,
1261
+ "learning_rate": 3.0122279759604745e-06,
1262
+ "loss": 0.0116,
1263
+ "step": 1790
1264
+ },
1265
+ {
1266
+ "epoch": 0.7556675062972292,
1267
+ "grad_norm": 0.1151094138622284,
1268
+ "learning_rate": 2.73604658704939e-06,
1269
+ "loss": 0.0075,
1270
+ "step": 1800
1271
+ },
1272
+ {
1273
+ "epoch": 0.7598656591099916,
1274
+ "grad_norm": 0.19842703640460968,
1275
+ "learning_rate": 2.4727866910737583e-06,
1276
+ "loss": 0.0065,
1277
+ "step": 1810
1278
+ },
1279
+ {
1280
+ "epoch": 0.764063811922754,
1281
+ "grad_norm": 0.2035660296678543,
1282
+ "learning_rate": 2.222520260660521e-06,
1283
+ "loss": 0.0057,
1284
+ "step": 1820
1285
+ },
1286
+ {
1287
+ "epoch": 0.7682619647355163,
1288
+ "grad_norm": 0.16106460988521576,
1289
+ "learning_rate": 1.985315716152847e-06,
1290
+ "loss": 0.0058,
1291
+ "step": 1830
1292
+ },
1293
+ {
1294
+ "epoch": 0.7724601175482787,
1295
+ "grad_norm": 0.20101721584796906,
1296
+ "learning_rate": 1.7612379069047335e-06,
1297
+ "loss": 0.0059,
1298
+ "step": 1840
1299
+ },
1300
+ {
1301
+ "epoch": 0.7766582703610412,
1302
+ "grad_norm": 0.11036290228366852,
1303
+ "learning_rate": 1.550348093551829e-06,
1304
+ "loss": 0.0069,
1305
+ "step": 1850
1306
+ },
1307
+ {
1308
+ "epoch": 0.7808564231738035,
1309
+ "grad_norm": 0.14676934480667114,
1310
+ "learning_rate": 1.3527039312633827e-06,
1311
+ "loss": 0.0082,
1312
+ "step": 1860
1313
+ },
1314
+ {
1315
+ "epoch": 0.7850545759865659,
1316
+ "grad_norm": 0.21743223071098328,
1317
+ "learning_rate": 1.1683594539798893e-06,
1318
+ "loss": 0.0093,
1319
+ "step": 1870
1320
+ },
1321
+ {
1322
+ "epoch": 0.7892527287993283,
1323
+ "grad_norm": 0.09522661566734314,
1324
+ "learning_rate": 9.97365059640787e-07,
1325
+ "loss": 0.0058,
1326
+ "step": 1880
1327
+ },
1328
+ {
1329
+ "epoch": 0.7934508816120907,
1330
+ "grad_norm": 0.12366088479757309,
1331
+ "learning_rate": 8.397674964061075e-07,
1332
+ "loss": 0.0077,
1333
+ "step": 1890
1334
+ },
1335
+ {
1336
+ "epoch": 0.797649034424853,
1337
+ "grad_norm": 0.17567159235477448,
1338
+ "learning_rate": 6.956098498760389e-07,
1339
+ "loss": 0.0105,
1340
+ "step": 1900
1341
+ },
1342
+ {
1343
+ "epoch": 0.8018471872376155,
1344
+ "grad_norm": 0.17301349341869354,
1345
+ "learning_rate": 5.64931531311741e-07,
1346
+ "loss": 0.0072,
1347
+ "step": 1910
1348
+ },
1349
+ {
1350
+ "epoch": 0.8060453400503779,
1351
+ "grad_norm": 0.13477115333080292,
1352
+ "learning_rate": 4.4776826686069305e-07,
1353
+ "loss": 0.008,
1354
+ "step": 1920
1355
+ },
1356
+ {
1357
+ "epoch": 0.8102434928631402,
1358
+ "grad_norm": 0.16665753722190857,
1359
+ "learning_rate": 3.441520877895288e-07,
1360
+ "loss": 0.0078,
1361
+ "step": 1930
1362
+ },
1363
+ {
1364
+ "epoch": 0.8144416456759026,
1365
+ "grad_norm": 0.20697897672653198,
1366
+ "learning_rate": 2.5411132172700194e-07,
1367
+ "loss": 0.0079,
1368
+ "step": 1940
1369
+ },
1370
+ {
1371
+ "epoch": 0.818639798488665,
1372
+ "grad_norm": 0.23099485039710999,
1373
+ "learning_rate": 1.776705849195037e-07,
1374
+ "loss": 0.0104,
1375
+ "step": 1950
1376
+ },
1377
+ {
1378
+ "epoch": 0.8228379513014273,
1379
+ "grad_norm": 0.16362063586711884,
1380
+ "learning_rate": 1.1485077550122402e-07,
1381
+ "loss": 0.0057,
1382
+ "step": 1960
1383
+ },
1384
+ {
1385
+ "epoch": 0.8270361041141897,
1386
+ "grad_norm": 0.12974146008491516,
1387
+ "learning_rate": 6.566906778079917e-08,
1388
+ "loss": 0.0079,
1389
+ "step": 1970
1390
+ },
1391
+ {
1392
+ "epoch": 0.8312342569269522,
1393
+ "grad_norm": 0.1990516036748886,
1394
+ "learning_rate": 3.01389075460512e-08,
1395
+ "loss": 0.0101,
1396
+ "step": 1980
1397
+ },
1398
+ {
1399
+ "epoch": 0.8354324097397146,
1400
+ "grad_norm": 0.15594905614852905,
1401
+ "learning_rate": 8.270008388022721e-09,
1402
+ "loss": 0.0065,
1403
+ "step": 1990
1404
+ },
1405
+ {
1406
+ "epoch": 0.8396305625524769,
1407
+ "grad_norm": 0.07478626072406769,
1408
+ "learning_rate": 6.834904537900144e-11,
1409
+ "loss": 0.0054,
1410
+ "step": 2000
1411
+ }
1412
+ ],
1413
+ "logging_steps": 10,
1414
+ "max_steps": 2000,
1415
+ "num_input_tokens_seen": 0,
1416
+ "num_train_epochs": 1,
1417
+ "save_steps": 1000,
1418
+ "stateful_callbacks": {
1419
+ "TrainerControl": {
1420
+ "args": {
1421
+ "should_epoch_stop": false,
1422
+ "should_evaluate": false,
1423
+ "should_log": false,
1424
+ "should_save": true,
1425
+ "should_training_stop": true
1426
+ },
1427
+ "attributes": {}
1428
+ }
1429
+ },
1430
+ "total_flos": 0.0,
1431
+ "train_batch_size": 32,
1432
+ "trial_name": null,
1433
+ "trial_params": null
1434
+ }
config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "action_dim": 32,
3
+ "action_head_cfg": {
4
+ "action_dim": 32,
5
+ "action_horizon": 16,
6
+ "add_pos_embed": true,
7
+ "backbone_embedding_dim": 2048,
8
+ "diffusion_model_cfg": {
9
+ "attention_head_dim": 48,
10
+ "cross_attention_dim": 2048,
11
+ "dropout": 0.2,
12
+ "final_dropout": true,
13
+ "interleave_self_attention": true,
14
+ "norm_type": "ada_norm",
15
+ "num_attention_heads": 32,
16
+ "num_layers": 16,
17
+ "output_dim": 1024,
18
+ "positional_embeddings": null
19
+ },
20
+ "hidden_size": 1024,
21
+ "input_embedding_dim": 1536,
22
+ "max_action_dim": 32,
23
+ "max_state_dim": 64,
24
+ "model_dtype": "float32",
25
+ "noise_beta_alpha": 1.5,
26
+ "noise_beta_beta": 1.0,
27
+ "noise_s": 0.999,
28
+ "num_inference_timesteps": 4,
29
+ "num_target_vision_tokens": 32,
30
+ "num_timestep_buckets": 1000,
31
+ "tune_diffusion_model": true,
32
+ "tune_projector": true,
33
+ "use_vlln": true,
34
+ "vl_self_attention_cfg": {
35
+ "attention_head_dim": 64,
36
+ "dropout": 0.2,
37
+ "final_dropout": true,
38
+ "num_attention_heads": 32,
39
+ "num_layers": 4,
40
+ "positional_embeddings": null
41
+ }
42
+ },
43
+ "action_horizon": 16,
44
+ "architectures": [
45
+ "GR00T_N1_5"
46
+ ],
47
+ "attn_implementation": null,
48
+ "backbone_cfg": {
49
+ "eagle_path": "NVEagle/eagle_er-qwen3_1_7B-Siglip2_400M_stage1_5_128gpu_er_v7_1mlp_nops",
50
+ "load_bf16": false,
51
+ "project_to_dim": null,
52
+ "reproject_vision": false,
53
+ "select_layer": 12,
54
+ "tune_llm": false,
55
+ "tune_visual": true,
56
+ "use_flash_attention": true
57
+ },
58
+ "compute_dtype": "bfloat16",
59
+ "hidden_size": 2048,
60
+ "model_dtype": "float32",
61
+ "model_type": "gr00t_n1_5",
62
+ "torch_dtype": "bfloat16",
63
+ "transformers_version": "4.51.3"
64
+ }
experiment_cfg/metadata.json ADDED
@@ -0,0 +1,195 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "new_embodiment": {
3
+ "statistics": {
4
+ "state": {
5
+ "single_arm": {
6
+ "max": [
7
+ 2.0000243186950684,
8
+ 2.9169764518737793,
9
+ -0.038691196590662,
10
+ 1.2000755071640015,
11
+ 1.5494728088378906
12
+ ],
13
+ "min": [
14
+ -1.6060552597045898,
15
+ -0.0013021699851378798,
16
+ -2.8603312969207764,
17
+ -0.24804846942424774,
18
+ -2.4647815227508545
19
+ ],
20
+ "mean": [
21
+ 0.5791597962379456,
22
+ 1.1458725929260254,
23
+ -1.20786452293396,
24
+ 0.7232474684715271,
25
+ -0.6134248375892639
26
+ ],
27
+ "std": [
28
+ 0.9811665415763855,
29
+ 0.7337602376937866,
30
+ 0.6142615079879761,
31
+ 0.23586146533489227,
32
+ 0.966107964515686
33
+ ],
34
+ "q01": [
35
+ -1.4408677732944488,
36
+ -1.425088475116354e-05,
37
+ -2.781974565982819,
38
+ 0.27418837130069734,
39
+ -1.9999565029144286
40
+ ],
41
+ "q99": [
42
+ 1.9999998807907104,
43
+ 2.6177230525016784,
44
+ -0.3294296935200691,
45
+ 1.2000004053115845,
46
+ 1.4374910080432892
47
+ ]
48
+ },
49
+ "gripper": {
50
+ "max": [
51
+ 0.7455036640167236
52
+ ],
53
+ "min": [
54
+ -0.09471277892589569
55
+ ],
56
+ "mean": [
57
+ 0.000568521732930094
58
+ ],
59
+ "std": [
60
+ 0.14351940155029297
61
+ ],
62
+ "q01": [
63
+ -0.05334045834839344
64
+ ],
65
+ "q99": [
66
+ 0.5978925746679306
67
+ ]
68
+ }
69
+ },
70
+ "action": {
71
+ "single_arm": {
72
+ "max": [
73
+ 2.0000085830688477,
74
+ 2.9173049926757812,
75
+ -0.0662246122956276,
76
+ 1.2000755071640015,
77
+ 1.549481987953186
78
+ ],
79
+ "min": [
80
+ -1.5781737565994263,
81
+ -0.0013021699851378798,
82
+ -2.8603312969207764,
83
+ -0.24804846942424774,
84
+ -2.4347522258758545
85
+ ],
86
+ "mean": [
87
+ 0.5763245820999146,
88
+ 1.1609699726104736,
89
+ -1.2109930515289307,
90
+ 0.7210843563079834,
91
+ -0.6085058450698853
92
+ ],
93
+ "std": [
94
+ 0.9860737323760986,
95
+ 0.7361332774162292,
96
+ 0.6108555197715759,
97
+ 0.23312774300575256,
98
+ 0.9712501168251038
99
+ ],
100
+ "q01": [
101
+ -1.4462056386470794,
102
+ -1.4080941818974679e-05,
103
+ -2.7542264652252197,
104
+ 0.27880221664905547,
105
+ -1.9997829270362855
106
+ ],
107
+ "q99": [
108
+ 1.9999998807907104,
109
+ 2.6269060873985284,
110
+ -0.33764528483152395,
111
+ 1.200000286102295,
112
+ 1.4436372423171995
113
+ ]
114
+ },
115
+ "gripper": {
116
+ "max": [
117
+ 0.7455036640167236
118
+ ],
119
+ "min": [
120
+ -0.07934218645095825
121
+ ],
122
+ "mean": [
123
+ -0.0036394144408404827
124
+ ],
125
+ "std": [
126
+ 0.13472281396389008
127
+ ],
128
+ "q01": [
129
+ -0.05393668111413717
130
+ ],
131
+ "q99": [
132
+ 0.56789559841156
133
+ ]
134
+ }
135
+ }
136
+ },
137
+ "modalities": {
138
+ "video": {
139
+ "front": {
140
+ "resolution": [
141
+ 640,
142
+ 480
143
+ ],
144
+ "channels": 3,
145
+ "fps": 30.0
146
+ },
147
+ "wrist": {
148
+ "resolution": [
149
+ 640,
150
+ 480
151
+ ],
152
+ "channels": 3,
153
+ "fps": 30.0
154
+ }
155
+ },
156
+ "state": {
157
+ "single_arm": {
158
+ "absolute": true,
159
+ "rotation_type": null,
160
+ "shape": [
161
+ 5
162
+ ],
163
+ "continuous": true
164
+ },
165
+ "gripper": {
166
+ "absolute": true,
167
+ "rotation_type": null,
168
+ "shape": [
169
+ 1
170
+ ],
171
+ "continuous": true
172
+ }
173
+ },
174
+ "action": {
175
+ "single_arm": {
176
+ "absolute": true,
177
+ "rotation_type": null,
178
+ "shape": [
179
+ 5
180
+ ],
181
+ "continuous": true
182
+ },
183
+ "gripper": {
184
+ "absolute": true,
185
+ "rotation_type": null,
186
+ "shape": [
187
+ 1
188
+ ],
189
+ "continuous": true
190
+ }
191
+ }
192
+ },
193
+ "embodiment_tag": "new_embodiment"
194
+ }
195
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c90564297b150a222847e1a1cb3b76645f5a645d14c88a22cda9b63a692768e8
3
+ size 4999367032
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d32a802c118a4bfcc906e341d99cc23fe780b40988d7b776b59df927e1eb40b
3
+ size 2586705312
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
trainer_state.json ADDED
@@ -0,0 +1,1443 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.8396305625524769,
6
+ "eval_steps": 500,
7
+ "global_step": 2000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0041981528127623844,
14
+ "grad_norm": 5.267488956451416,
15
+ "learning_rate": 9e-06,
16
+ "loss": 1.0193,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.008396305625524769,
21
+ "grad_norm": 2.3370044231414795,
22
+ "learning_rate": 1.9e-05,
23
+ "loss": 0.3362,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.012594458438287154,
28
+ "grad_norm": 2.1865203380584717,
29
+ "learning_rate": 2.9e-05,
30
+ "loss": 0.2021,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.016792611251049538,
35
+ "grad_norm": 1.625726342201233,
36
+ "learning_rate": 3.9000000000000006e-05,
37
+ "loss": 0.1466,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.020990764063811923,
42
+ "grad_norm": 1.1913143396377563,
43
+ "learning_rate": 4.9e-05,
44
+ "loss": 0.1184,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.02518891687657431,
49
+ "grad_norm": 1.6674305200576782,
50
+ "learning_rate": 5.9e-05,
51
+ "loss": 0.1028,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.02938706968933669,
56
+ "grad_norm": 0.8989318013191223,
57
+ "learning_rate": 6.9e-05,
58
+ "loss": 0.0836,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.033585222502099076,
63
+ "grad_norm": 0.8966989517211914,
64
+ "learning_rate": 7.900000000000001e-05,
65
+ "loss": 0.0851,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.037783375314861464,
70
+ "grad_norm": 0.7644183039665222,
71
+ "learning_rate": 8.900000000000001e-05,
72
+ "loss": 0.0717,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.041981528127623846,
77
+ "grad_norm": 1.272506833076477,
78
+ "learning_rate": 9.900000000000001e-05,
79
+ "loss": 0.0774,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.04617968094038623,
84
+ "grad_norm": 1.4885015487670898,
85
+ "learning_rate": 9.999446382823013e-05,
86
+ "loss": 0.0712,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.05037783375314862,
91
+ "grad_norm": 1.201401710510254,
92
+ "learning_rate": 9.997532801828658e-05,
93
+ "loss": 0.072,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.054575986565911,
98
+ "grad_norm": 1.367532730102539,
99
+ "learning_rate": 9.99425294526634e-05,
100
+ "loss": 0.0682,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.05877413937867338,
105
+ "grad_norm": 0.9127843976020813,
106
+ "learning_rate": 9.989607709816091e-05,
107
+ "loss": 0.0659,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.06297229219143577,
112
+ "grad_norm": 0.7713141441345215,
113
+ "learning_rate": 9.983598365438902e-05,
114
+ "loss": 0.0528,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.06717044500419815,
119
+ "grad_norm": 0.9186705946922302,
120
+ "learning_rate": 9.976226555029522e-05,
121
+ "loss": 0.0495,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.07136859781696053,
126
+ "grad_norm": 0.5967724919319153,
127
+ "learning_rate": 9.967494293967312e-05,
128
+ "loss": 0.0522,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.07556675062972293,
133
+ "grad_norm": 1.0857634544372559,
134
+ "learning_rate": 9.95740396956525e-05,
135
+ "loss": 0.0578,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.07976490344248531,
140
+ "grad_norm": 0.5992318987846375,
141
+ "learning_rate": 9.945958340417283e-05,
142
+ "loss": 0.0478,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.08396305625524769,
147
+ "grad_norm": 0.5689753293991089,
148
+ "learning_rate": 9.93316053564413e-05,
149
+ "loss": 0.0478,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.08816120906801007,
154
+ "grad_norm": 0.9272230863571167,
155
+ "learning_rate": 9.919014054037836e-05,
156
+ "loss": 0.0527,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.09235936188077246,
161
+ "grad_norm": 1.0262973308563232,
162
+ "learning_rate": 9.903522763105218e-05,
163
+ "loss": 0.051,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.09655751469353484,
168
+ "grad_norm": 0.6636225581169128,
169
+ "learning_rate": 9.886690898010535e-05,
170
+ "loss": 0.0429,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.10075566750629723,
175
+ "grad_norm": 0.6513682007789612,
176
+ "learning_rate": 9.868523060417646e-05,
177
+ "loss": 0.0443,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.10495382031905962,
182
+ "grad_norm": 0.7679399251937866,
183
+ "learning_rate": 9.849024217231935e-05,
184
+ "loss": 0.0406,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.109151973131822,
189
+ "grad_norm": 0.5520310401916504,
190
+ "learning_rate": 9.82819969924244e-05,
191
+ "loss": 0.0345,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.11335012594458438,
196
+ "grad_norm": 0.7136860489845276,
197
+ "learning_rate": 9.806055199664446e-05,
198
+ "loss": 0.0487,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.11754827875734676,
203
+ "grad_norm": 0.5643390417098999,
204
+ "learning_rate": 9.782596772583026e-05,
205
+ "loss": 0.0419,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.12174643157010916,
210
+ "grad_norm": 0.6041691899299622,
211
+ "learning_rate": 9.757830831297914e-05,
212
+ "loss": 0.0479,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.12594458438287154,
217
+ "grad_norm": 0.5313841700553894,
218
+ "learning_rate": 9.731764146570173e-05,
219
+ "loss": 0.037,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.13014273719563393,
224
+ "grad_norm": 0.4854077100753784,
225
+ "learning_rate": 9.704403844771128e-05,
226
+ "loss": 0.0368,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.1343408900083963,
231
+ "grad_norm": 0.44807952642440796,
232
+ "learning_rate": 9.675757405934103e-05,
233
+ "loss": 0.0363,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.1385390428211587,
238
+ "grad_norm": 0.51119464635849,
239
+ "learning_rate": 9.645832661709444e-05,
240
+ "loss": 0.0308,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.14273719563392107,
245
+ "grad_norm": 0.5140641927719116,
246
+ "learning_rate": 9.614637793223425e-05,
247
+ "loss": 0.0316,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.14693534844668346,
252
+ "grad_norm": 0.5101410150527954,
253
+ "learning_rate": 9.582181328841611e-05,
254
+ "loss": 0.0307,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.15113350125944586,
259
+ "grad_norm": 0.6274232864379883,
260
+ "learning_rate": 9.548472141837286e-05,
261
+ "loss": 0.0324,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.15533165407220823,
266
+ "grad_norm": 0.6747106313705444,
267
+ "learning_rate": 9.513519447965595e-05,
268
+ "loss": 0.0353,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.15952980688497062,
273
+ "grad_norm": 0.38412219285964966,
274
+ "learning_rate": 9.477332802944044e-05,
275
+ "loss": 0.0311,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.163727959697733,
280
+ "grad_norm": 0.3586186468601227,
281
+ "learning_rate": 9.439922099840054e-05,
282
+ "loss": 0.032,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.16792611251049538,
287
+ "grad_norm": 0.38126131892204285,
288
+ "learning_rate": 9.401297566366318e-05,
289
+ "loss": 0.0266,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.17212426532325778,
294
+ "grad_norm": 0.3684011995792389,
295
+ "learning_rate": 9.36146976208462e-05,
296
+ "loss": 0.0375,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.17632241813602015,
301
+ "grad_norm": 0.5006753206253052,
302
+ "learning_rate": 9.320449575518972e-05,
303
+ "loss": 0.028,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.18052057094878254,
308
+ "grad_norm": 0.5954798460006714,
309
+ "learning_rate": 9.278248221178798e-05,
310
+ "loss": 0.0329,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.1847187237615449,
315
+ "grad_norm": 0.745754063129425,
316
+ "learning_rate": 9.234877236492997e-05,
317
+ "loss": 0.0339,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.1889168765743073,
322
+ "grad_norm": 0.6667306423187256,
323
+ "learning_rate": 9.190348478655724e-05,
324
+ "loss": 0.0316,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.19311502938706968,
329
+ "grad_norm": 0.7456135153770447,
330
+ "learning_rate": 9.144674121384757e-05,
331
+ "loss": 0.0394,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.19731318219983207,
336
+ "grad_norm": 0.4713476002216339,
337
+ "learning_rate": 9.097866651593317e-05,
338
+ "loss": 0.031,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.20151133501259447,
343
+ "grad_norm": 0.9486730098724365,
344
+ "learning_rate": 9.049938865976275e-05,
345
+ "loss": 0.0335,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.20570948782535683,
350
+ "grad_norm": 0.8080450296401978,
351
+ "learning_rate": 9.000903867511666e-05,
352
+ "loss": 0.0326,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.20990764063811923,
357
+ "grad_norm": 0.6074867844581604,
358
+ "learning_rate": 8.950775061878453e-05,
359
+ "loss": 0.0326,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.2141057934508816,
364
+ "grad_norm": 0.4722858965396881,
365
+ "learning_rate": 8.899566153791566e-05,
366
+ "loss": 0.0362,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.218303946263644,
371
+ "grad_norm": 0.3163154721260071,
372
+ "learning_rate": 8.84729114325516e-05,
373
+ "loss": 0.0312,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.2225020990764064,
378
+ "grad_norm": 0.45854002237319946,
379
+ "learning_rate": 8.79396432173515e-05,
380
+ "loss": 0.0319,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.22670025188916876,
385
+ "grad_norm": 0.5331682562828064,
386
+ "learning_rate": 8.739600268252078e-05,
387
+ "loss": 0.0302,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.23089840470193115,
392
+ "grad_norm": 0.44824129343032837,
393
+ "learning_rate": 8.684213845395339e-05,
394
+ "loss": 0.0256,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.23509655751469352,
399
+ "grad_norm": 0.38797256350517273,
400
+ "learning_rate": 8.627820195259918e-05,
401
+ "loss": 0.024,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.23929471032745592,
406
+ "grad_norm": 0.44604647159576416,
407
+ "learning_rate": 8.570434735306671e-05,
408
+ "loss": 0.0265,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.2434928631402183,
413
+ "grad_norm": 0.46292269229888916,
414
+ "learning_rate": 8.512073154147362e-05,
415
+ "loss": 0.0238,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.24769101595298068,
420
+ "grad_norm": 0.6517325043678284,
421
+ "learning_rate": 8.452751407255541e-05,
422
+ "loss": 0.0283,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.2518891687657431,
427
+ "grad_norm": 0.3638853430747986,
428
+ "learning_rate": 8.392485712604483e-05,
429
+ "loss": 0.0269,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.25608732157850544,
434
+ "grad_norm": 0.3911242187023163,
435
+ "learning_rate": 8.331292546233362e-05,
436
+ "loss": 0.0298,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.26028547439126787,
441
+ "grad_norm": 0.4468225836753845,
442
+ "learning_rate": 8.269188637742846e-05,
443
+ "loss": 0.0235,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.26448362720403024,
448
+ "grad_norm": 0.4399245083332062,
449
+ "learning_rate": 8.206190965721419e-05,
450
+ "loss": 0.0217,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.2686817800167926,
455
+ "grad_norm": 0.34588077664375305,
456
+ "learning_rate": 8.14231675310358e-05,
457
+ "loss": 0.0212,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.27287993282955497,
462
+ "grad_norm": 0.3927046060562134,
463
+ "learning_rate": 8.077583462461283e-05,
464
+ "loss": 0.0228,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.2770780856423174,
469
+ "grad_norm": 0.42579185962677,
470
+ "learning_rate": 8.012008791229826e-05,
471
+ "loss": 0.0239,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.28127623845507976,
476
+ "grad_norm": 0.37947288155555725,
477
+ "learning_rate": 7.945610666869568e-05,
478
+ "loss": 0.0302,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.28547439126784213,
483
+ "grad_norm": 0.46608966588974,
484
+ "learning_rate": 7.878407241964729e-05,
485
+ "loss": 0.0269,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.28967254408060455,
490
+ "grad_norm": 0.4380934536457062,
491
+ "learning_rate": 7.810416889260653e-05,
492
+ "loss": 0.0259,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.2938706968933669,
497
+ "grad_norm": 0.4817158281803131,
498
+ "learning_rate": 7.741658196640892e-05,
499
+ "loss": 0.0246,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.2980688497061293,
504
+ "grad_norm": 0.3391236960887909,
505
+ "learning_rate": 7.672149962045457e-05,
506
+ "loss": 0.0285,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.3022670025188917,
511
+ "grad_norm": 0.5921253561973572,
512
+ "learning_rate": 7.60191118833165e-05,
513
+ "loss": 0.0192,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.3064651553316541,
518
+ "grad_norm": 0.3677942156791687,
519
+ "learning_rate": 7.530961078078873e-05,
520
+ "loss": 0.0259,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.31066330814441645,
525
+ "grad_norm": 0.4081045985221863,
526
+ "learning_rate": 7.45931902833884e-05,
527
+ "loss": 0.025,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.3148614609571788,
532
+ "grad_norm": 0.4675191342830658,
533
+ "learning_rate": 7.387004625332608e-05,
534
+ "loss": 0.0223,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.31905961376994124,
539
+ "grad_norm": 0.40352970361709595,
540
+ "learning_rate": 7.3140376390959e-05,
541
+ "loss": 0.0177,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.3232577665827036,
546
+ "grad_norm": 0.4271659851074219,
547
+ "learning_rate": 7.240438018074189e-05,
548
+ "loss": 0.0187,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.327455919395466,
553
+ "grad_norm": 0.4077198803424835,
554
+ "learning_rate": 7.166225883668969e-05,
555
+ "loss": 0.0198,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.3316540722082284,
560
+ "grad_norm": 0.24861399829387665,
561
+ "learning_rate": 7.091421524736784e-05,
562
+ "loss": 0.0182,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.33585222502099077,
567
+ "grad_norm": 0.3340529501438141,
568
+ "learning_rate": 7.016045392042452e-05,
569
+ "loss": 0.0195,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.34005037783375314,
574
+ "grad_norm": 0.30124542117118835,
575
+ "learning_rate": 6.940118092668022e-05,
576
+ "loss": 0.0202,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.34424853064651556,
581
+ "grad_norm": 0.28010880947113037,
582
+ "learning_rate": 6.863660384379017e-05,
583
+ "loss": 0.0192,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.34844668345927793,
588
+ "grad_norm": 0.34497565031051636,
589
+ "learning_rate": 6.786693169949455e-05,
590
+ "loss": 0.0188,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.3526448362720403,
595
+ "grad_norm": 0.38020557165145874,
596
+ "learning_rate": 6.709237491447249e-05,
597
+ "loss": 0.0159,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.35684298908480266,
602
+ "grad_norm": 0.677865207195282,
603
+ "learning_rate": 6.631314524481513e-05,
604
+ "loss": 0.0214,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.3610411418975651,
609
+ "grad_norm": 0.741203784942627,
610
+ "learning_rate": 6.552945572413358e-05,
611
+ "loss": 0.0203,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.36523929471032746,
616
+ "grad_norm": 0.35501229763031006,
617
+ "learning_rate": 6.474152060531768e-05,
618
+ "loss": 0.0193,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.3694374475230898,
623
+ "grad_norm": 0.3309725522994995,
624
+ "learning_rate": 6.394955530196147e-05,
625
+ "loss": 0.0152,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.37363560033585225,
630
+ "grad_norm": 0.36004316806793213,
631
+ "learning_rate": 6.315377632947115e-05,
632
+ "loss": 0.0151,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.3778337531486146,
637
+ "grad_norm": 0.35332006216049194,
638
+ "learning_rate": 6.235440124587198e-05,
639
+ "loss": 0.0174,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.382031905961377,
644
+ "grad_norm": 0.3157385587692261,
645
+ "learning_rate": 6.155164859233012e-05,
646
+ "loss": 0.0125,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.38623005877413935,
651
+ "grad_norm": 0.2917327284812927,
652
+ "learning_rate": 6.074573783340562e-05,
653
+ "loss": 0.0185,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.3904282115869018,
658
+ "grad_norm": 0.20623797178268433,
659
+ "learning_rate": 5.9936889297052986e-05,
660
+ "loss": 0.0141,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.39462636439966414,
665
+ "grad_norm": 0.30338916182518005,
666
+ "learning_rate": 5.912532411438576e-05,
667
+ "loss": 0.0124,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.3988245172124265,
672
+ "grad_norm": 0.46562105417251587,
673
+ "learning_rate": 5.831126415922148e-05,
674
+ "loss": 0.0187,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.40302267002518893,
679
+ "grad_norm": 0.303715318441391,
680
+ "learning_rate": 5.74949319874235e-05,
681
+ "loss": 0.013,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.4072208228379513,
686
+ "grad_norm": 0.33936807513237,
687
+ "learning_rate": 5.667655077605659e-05,
688
+ "loss": 0.0147,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.41141897565071367,
693
+ "grad_norm": 0.26057168841362,
694
+ "learning_rate": 5.585634426237246e-05,
695
+ "loss": 0.0147,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.4156171284634761,
700
+ "grad_norm": 0.30878400802612305,
701
+ "learning_rate": 5.5034536682642224e-05,
702
+ "loss": 0.0208,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.41981528127623846,
707
+ "grad_norm": 0.32619133591651917,
708
+ "learning_rate": 5.4211352710852495e-05,
709
+ "loss": 0.0147,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.42401343408900083,
714
+ "grad_norm": 0.4049901068210602,
715
+ "learning_rate": 5.3387017397281704e-05,
716
+ "loss": 0.0136,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.4282115869017632,
721
+ "grad_norm": 0.263507217168808,
722
+ "learning_rate": 5.2561756106973656e-05,
723
+ "loss": 0.0142,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.4324097397145256,
728
+ "grad_norm": 0.3729311525821686,
729
+ "learning_rate": 5.1735794458124956e-05,
730
+ "loss": 0.0143,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.436607892527288,
735
+ "grad_norm": 0.433525413274765,
736
+ "learning_rate": 5.0909358260403186e-05,
737
+ "loss": 0.0159,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.44080604534005036,
742
+ "grad_norm": 0.20972338318824768,
743
+ "learning_rate": 5.0082673453212914e-05,
744
+ "loss": 0.0125,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.4450041981528128,
749
+ "grad_norm": 0.44978395104408264,
750
+ "learning_rate": 4.925596604392603e-05,
751
+ "loss": 0.0201,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.44920235096557515,
756
+ "grad_norm": 0.33291032910346985,
757
+ "learning_rate": 4.8429462046093585e-05,
758
+ "loss": 0.0145,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.4534005037783375,
763
+ "grad_norm": 0.3551059663295746,
764
+ "learning_rate": 4.7603387417656026e-05,
765
+ "loss": 0.0164,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.45759865659109994,
770
+ "grad_norm": 0.2902749478816986,
771
+ "learning_rate": 4.677796799916845e-05,
772
+ "loss": 0.0088,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.4617968094038623,
777
+ "grad_norm": 0.4253701865673065,
778
+ "learning_rate": 4.5953429452058135e-05,
779
+ "loss": 0.0149,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.4659949622166247,
784
+ "grad_norm": 0.25408294796943665,
785
+ "learning_rate": 4.5129997196930845e-05,
786
+ "loss": 0.014,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.47019311502938704,
791
+ "grad_norm": 0.3801613450050354,
792
+ "learning_rate": 4.430789635194324e-05,
793
+ "loss": 0.0135,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.47439126784214947,
798
+ "grad_norm": 0.3434455990791321,
799
+ "learning_rate": 4.348735167125771e-05,
800
+ "loss": 0.0175,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.47858942065491183,
805
+ "grad_norm": 0.38031941652297974,
806
+ "learning_rate": 4.2668587483596864e-05,
807
+ "loss": 0.0131,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.4827875734676742,
812
+ "grad_norm": 0.2573353350162506,
813
+ "learning_rate": 4.1851827630914305e-05,
814
+ "loss": 0.0137,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.4869857262804366,
819
+ "grad_norm": 0.31360694766044617,
820
+ "learning_rate": 4.103729540719847e-05,
821
+ "loss": 0.0146,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.491183879093199,
826
+ "grad_norm": 0.32134002447128296,
827
+ "learning_rate": 4.0225213497426276e-05,
828
+ "loss": 0.0133,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.49538203190596136,
833
+ "grad_norm": 0.31462499499320984,
834
+ "learning_rate": 3.9415803916683224e-05,
835
+ "loss": 0.0124,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.4995801847187238,
840
+ "grad_norm": 0.3015807569026947,
841
+ "learning_rate": 3.860928794946682e-05,
842
+ "loss": 0.0147,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.5037783375314862,
847
+ "grad_norm": 0.26400843262672424,
848
+ "learning_rate": 3.780588608918947e-05,
849
+ "loss": 0.0125,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.5079764903442485,
854
+ "grad_norm": 0.27983877062797546,
855
+ "learning_rate": 3.700581797789786e-05,
856
+ "loss": 0.0139,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.5121746431570109,
861
+ "grad_norm": 0.34057655930519104,
862
+ "learning_rate": 3.6209302346225006e-05,
863
+ "loss": 0.0125,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.5163727959697733,
868
+ "grad_norm": 0.3145323097705841,
869
+ "learning_rate": 3.541655695359142e-05,
870
+ "loss": 0.0141,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.5205709487825357,
875
+ "grad_norm": 0.2706829905509949,
876
+ "learning_rate": 3.462779852867197e-05,
877
+ "loss": 0.0112,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.5247691015952981,
882
+ "grad_norm": 0.3848113417625427,
883
+ "learning_rate": 3.384324271014429e-05,
884
+ "loss": 0.0147,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.5289672544080605,
889
+ "grad_norm": 0.32747682929039,
890
+ "learning_rate": 3.3063103987735433e-05,
891
+ "loss": 0.0114,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.5331654072208228,
896
+ "grad_norm": 0.2991146445274353,
897
+ "learning_rate": 3.228759564358248e-05,
898
+ "loss": 0.0141,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.5373635600335852,
903
+ "grad_norm": 0.3067378103733063,
904
+ "learning_rate": 3.1516929693923315e-05,
905
+ "loss": 0.0129,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.5415617128463476,
910
+ "grad_norm": 0.15431927144527435,
911
+ "learning_rate": 3.075131683113352e-05,
912
+ "loss": 0.013,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.5457598656591099,
917
+ "grad_norm": 0.1479986160993576,
918
+ "learning_rate": 2.999096636612518e-05,
919
+ "loss": 0.0121,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.5499580184718724,
924
+ "grad_norm": 0.1798468679189682,
925
+ "learning_rate": 2.9236086171123404e-05,
926
+ "loss": 0.0107,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.5541561712846348,
931
+ "grad_norm": 0.21686813235282898,
932
+ "learning_rate": 2.8486882622836026e-05,
933
+ "loss": 0.0124,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.5583543240973972,
938
+ "grad_norm": 0.40006551146507263,
939
+ "learning_rate": 2.774356054603243e-05,
940
+ "loss": 0.0103,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.5625524769101595,
945
+ "grad_norm": 0.39905092120170593,
946
+ "learning_rate": 2.7006323157546386e-05,
947
+ "loss": 0.0146,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.5667506297229219,
952
+ "grad_norm": 0.156532883644104,
953
+ "learning_rate": 2.6275372010718635e-05,
954
+ "loss": 0.0083,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.5709487825356843,
959
+ "grad_norm": 0.418869286775589,
960
+ "learning_rate": 2.555090694029421e-05,
961
+ "loss": 0.0112,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.5751469353484466,
966
+ "grad_norm": 0.3597627878189087,
967
+ "learning_rate": 2.4833126007789653e-05,
968
+ "loss": 0.0071,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.5793450881612091,
973
+ "grad_norm": 0.3350473940372467,
974
+ "learning_rate": 2.4122225447344875e-05,
975
+ "loss": 0.0094,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.5835432409739715,
980
+ "grad_norm": 0.2572272717952728,
981
+ "learning_rate": 2.341839961207482e-05,
982
+ "loss": 0.0086,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.5877413937867338,
987
+ "grad_norm": 0.44453227519989014,
988
+ "learning_rate": 2.2721840920935196e-05,
989
+ "loss": 0.0104,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.5919395465994962,
994
+ "grad_norm": 0.23089422285556793,
995
+ "learning_rate": 2.2032739806117058e-05,
996
+ "loss": 0.0075,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.5961376994122586,
1001
+ "grad_norm": 0.2570338547229767,
1002
+ "learning_rate": 2.1351284660984572e-05,
1003
+ "loss": 0.0128,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.600335852225021,
1008
+ "grad_norm": 0.4198639690876007,
1009
+ "learning_rate": 2.067766178857013e-05,
1010
+ "loss": 0.0133,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 0.6045340050377834,
1015
+ "grad_norm": 0.4144093990325928,
1016
+ "learning_rate": 2.0012055350640986e-05,
1017
+ "loss": 0.0095,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 0.6087321578505458,
1022
+ "grad_norm": 0.25941231846809387,
1023
+ "learning_rate": 1.9354647317351188e-05,
1024
+ "loss": 0.0105,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 0.6129303106633082,
1029
+ "grad_norm": 0.25724703073501587,
1030
+ "learning_rate": 1.8705617417492883e-05,
1031
+ "loss": 0.0096,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 0.6171284634760705,
1036
+ "grad_norm": 0.2537188231945038,
1037
+ "learning_rate": 1.8065143089360172e-05,
1038
+ "loss": 0.0094,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 0.6213266162888329,
1043
+ "grad_norm": 0.25936058163642883,
1044
+ "learning_rate": 1.743339943223926e-05,
1045
+ "loss": 0.0132,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 0.6255247691015953,
1050
+ "grad_norm": 0.27446359395980835,
1051
+ "learning_rate": 1.6810559158538092e-05,
1052
+ "loss": 0.0084,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 0.6297229219143576,
1057
+ "grad_norm": 0.27379536628723145,
1058
+ "learning_rate": 1.6196792546568472e-05,
1059
+ "loss": 0.0126,
1060
+ "step": 1500
1061
+ },
1062
+ {
1063
+ "epoch": 0.6339210747271201,
1064
+ "grad_norm": 0.16973455250263214,
1065
+ "learning_rate": 1.5592267393993716e-05,
1066
+ "loss": 0.0082,
1067
+ "step": 1510
1068
+ },
1069
+ {
1070
+ "epoch": 0.6381192275398825,
1071
+ "grad_norm": 0.19682660698890686,
1072
+ "learning_rate": 1.4997148971954344e-05,
1073
+ "loss": 0.0119,
1074
+ "step": 1520
1075
+ },
1076
+ {
1077
+ "epoch": 0.6423173803526449,
1078
+ "grad_norm": 0.16340869665145874,
1079
+ "learning_rate": 1.4411599979884744e-05,
1080
+ "loss": 0.0101,
1081
+ "step": 1530
1082
+ },
1083
+ {
1084
+ "epoch": 0.6465155331654072,
1085
+ "grad_norm": 0.27933576703071594,
1086
+ "learning_rate": 1.383578050103268e-05,
1087
+ "loss": 0.0117,
1088
+ "step": 1540
1089
+ },
1090
+ {
1091
+ "epoch": 0.6507136859781696,
1092
+ "grad_norm": 0.10343077033758163,
1093
+ "learning_rate": 1.3269847958694148e-05,
1094
+ "loss": 0.0101,
1095
+ "step": 1550
1096
+ },
1097
+ {
1098
+ "epoch": 0.654911838790932,
1099
+ "grad_norm": 0.15868137776851654,
1100
+ "learning_rate": 1.2713957073175425e-05,
1101
+ "loss": 0.0084,
1102
+ "step": 1560
1103
+ },
1104
+ {
1105
+ "epoch": 0.6591099916036943,
1106
+ "grad_norm": 0.22004792094230652,
1107
+ "learning_rate": 1.2168259819494066e-05,
1108
+ "loss": 0.0087,
1109
+ "step": 1570
1110
+ },
1111
+ {
1112
+ "epoch": 0.6633081444164568,
1113
+ "grad_norm": 0.11181829124689102,
1114
+ "learning_rate": 1.1632905385830484e-05,
1115
+ "loss": 0.0098,
1116
+ "step": 1580
1117
+ },
1118
+ {
1119
+ "epoch": 0.6675062972292192,
1120
+ "grad_norm": 0.1479823738336563,
1121
+ "learning_rate": 1.1108040132741354e-05,
1122
+ "loss": 0.0093,
1123
+ "step": 1590
1124
+ },
1125
+ {
1126
+ "epoch": 0.6717044500419815,
1127
+ "grad_norm": 0.2433646023273468,
1128
+ "learning_rate": 1.059380755314613e-05,
1129
+ "loss": 0.0088,
1130
+ "step": 1600
1131
+ },
1132
+ {
1133
+ "epoch": 0.6759026028547439,
1134
+ "grad_norm": 0.212657168507576,
1135
+ "learning_rate": 1.009034823309749e-05,
1136
+ "loss": 0.0061,
1137
+ "step": 1610
1138
+ },
1139
+ {
1140
+ "epoch": 0.6801007556675063,
1141
+ "grad_norm": 0.10653030127286911,
1142
+ "learning_rate": 9.597799813346525e-06,
1143
+ "loss": 0.0068,
1144
+ "step": 1620
1145
+ },
1146
+ {
1147
+ "epoch": 0.6842989084802686,
1148
+ "grad_norm": 0.10436981171369553,
1149
+ "learning_rate": 9.116296951713133e-06,
1150
+ "loss": 0.0064,
1151
+ "step": 1630
1152
+ },
1153
+ {
1154
+ "epoch": 0.6884970612930311,
1155
+ "grad_norm": 0.28674691915512085,
1156
+ "learning_rate": 8.645971286271904e-06,
1157
+ "loss": 0.0083,
1158
+ "step": 1640
1159
+ },
1160
+ {
1161
+ "epoch": 0.6926952141057935,
1162
+ "grad_norm": 0.2074134796857834,
1163
+ "learning_rate": 8.186951399363613e-06,
1164
+ "loss": 0.0103,
1165
+ "step": 1650
1166
+ },
1167
+ {
1168
+ "epoch": 0.6968933669185559,
1169
+ "grad_norm": 0.160618856549263,
1170
+ "learning_rate": 7.739362782442021e-06,
1171
+ "loss": 0.0051,
1172
+ "step": 1660
1173
+ },
1174
+ {
1175
+ "epoch": 0.7010915197313182,
1176
+ "grad_norm": 0.2147001475095749,
1177
+ "learning_rate": 7.30332780176588e-06,
1178
+ "loss": 0.0085,
1179
+ "step": 1670
1180
+ },
1181
+ {
1182
+ "epoch": 0.7052896725440806,
1183
+ "grad_norm": 0.27866339683532715,
1184
+ "learning_rate": 6.878965664945108e-06,
1185
+ "loss": 0.0089,
1186
+ "step": 1680
1187
+ },
1188
+ {
1189
+ "epoch": 0.709487825356843,
1190
+ "grad_norm": 0.22506844997406006,
1191
+ "learning_rate": 6.466392388350695e-06,
1192
+ "loss": 0.009,
1193
+ "step": 1690
1194
+ },
1195
+ {
1196
+ "epoch": 0.7136859781696053,
1197
+ "grad_norm": 0.2676888406276703,
1198
+ "learning_rate": 6.0657207653969315e-06,
1199
+ "loss": 0.0074,
1200
+ "step": 1700
1201
+ },
1202
+ {
1203
+ "epoch": 0.7178841309823678,
1204
+ "grad_norm": 0.19724851846694946,
1205
+ "learning_rate": 5.67706033570487e-06,
1206
+ "loss": 0.0074,
1207
+ "step": 1710
1208
+ },
1209
+ {
1210
+ "epoch": 0.7220822837951302,
1211
+ "grad_norm": 0.18323613703250885,
1212
+ "learning_rate": 5.300517355155215e-06,
1213
+ "loss": 0.0093,
1214
+ "step": 1720
1215
+ },
1216
+ {
1217
+ "epoch": 0.7262804366078925,
1218
+ "grad_norm": 0.2785090208053589,
1219
+ "learning_rate": 4.936194766839103e-06,
1220
+ "loss": 0.0093,
1221
+ "step": 1730
1222
+ },
1223
+ {
1224
+ "epoch": 0.7304785894206549,
1225
+ "grad_norm": 0.16792871057987213,
1226
+ "learning_rate": 4.5841921729144424e-06,
1227
+ "loss": 0.0063,
1228
+ "step": 1740
1229
+ },
1230
+ {
1231
+ "epoch": 0.7346767422334173,
1232
+ "grad_norm": 0.1223788782954216,
1233
+ "learning_rate": 4.244605807375679e-06,
1234
+ "loss": 0.009,
1235
+ "step": 1750
1236
+ },
1237
+ {
1238
+ "epoch": 0.7388748950461796,
1239
+ "grad_norm": 0.13132202625274658,
1240
+ "learning_rate": 3.917528509744412e-06,
1241
+ "loss": 0.0093,
1242
+ "step": 1760
1243
+ },
1244
+ {
1245
+ "epoch": 0.743073047858942,
1246
+ "grad_norm": 0.2555523216724396,
1247
+ "learning_rate": 3.60304969968796e-06,
1248
+ "loss": 0.0078,
1249
+ "step": 1770
1250
+ },
1251
+ {
1252
+ "epoch": 0.7472712006717045,
1253
+ "grad_norm": 0.16517667472362518,
1254
+ "learning_rate": 3.301255352572946e-06,
1255
+ "loss": 0.0074,
1256
+ "step": 1780
1257
+ },
1258
+ {
1259
+ "epoch": 0.7514693534844669,
1260
+ "grad_norm": 0.1717490404844284,
1261
+ "learning_rate": 3.0122279759604745e-06,
1262
+ "loss": 0.0116,
1263
+ "step": 1790
1264
+ },
1265
+ {
1266
+ "epoch": 0.7556675062972292,
1267
+ "grad_norm": 0.1151094138622284,
1268
+ "learning_rate": 2.73604658704939e-06,
1269
+ "loss": 0.0075,
1270
+ "step": 1800
1271
+ },
1272
+ {
1273
+ "epoch": 0.7598656591099916,
1274
+ "grad_norm": 0.19842703640460968,
1275
+ "learning_rate": 2.4727866910737583e-06,
1276
+ "loss": 0.0065,
1277
+ "step": 1810
1278
+ },
1279
+ {
1280
+ "epoch": 0.764063811922754,
1281
+ "grad_norm": 0.2035660296678543,
1282
+ "learning_rate": 2.222520260660521e-06,
1283
+ "loss": 0.0057,
1284
+ "step": 1820
1285
+ },
1286
+ {
1287
+ "epoch": 0.7682619647355163,
1288
+ "grad_norm": 0.16106460988521576,
1289
+ "learning_rate": 1.985315716152847e-06,
1290
+ "loss": 0.0058,
1291
+ "step": 1830
1292
+ },
1293
+ {
1294
+ "epoch": 0.7724601175482787,
1295
+ "grad_norm": 0.20101721584796906,
1296
+ "learning_rate": 1.7612379069047335e-06,
1297
+ "loss": 0.0059,
1298
+ "step": 1840
1299
+ },
1300
+ {
1301
+ "epoch": 0.7766582703610412,
1302
+ "grad_norm": 0.11036290228366852,
1303
+ "learning_rate": 1.550348093551829e-06,
1304
+ "loss": 0.0069,
1305
+ "step": 1850
1306
+ },
1307
+ {
1308
+ "epoch": 0.7808564231738035,
1309
+ "grad_norm": 0.14676934480667114,
1310
+ "learning_rate": 1.3527039312633827e-06,
1311
+ "loss": 0.0082,
1312
+ "step": 1860
1313
+ },
1314
+ {
1315
+ "epoch": 0.7850545759865659,
1316
+ "grad_norm": 0.21743223071098328,
1317
+ "learning_rate": 1.1683594539798893e-06,
1318
+ "loss": 0.0093,
1319
+ "step": 1870
1320
+ },
1321
+ {
1322
+ "epoch": 0.7892527287993283,
1323
+ "grad_norm": 0.09522661566734314,
1324
+ "learning_rate": 9.97365059640787e-07,
1325
+ "loss": 0.0058,
1326
+ "step": 1880
1327
+ },
1328
+ {
1329
+ "epoch": 0.7934508816120907,
1330
+ "grad_norm": 0.12366088479757309,
1331
+ "learning_rate": 8.397674964061075e-07,
1332
+ "loss": 0.0077,
1333
+ "step": 1890
1334
+ },
1335
+ {
1336
+ "epoch": 0.797649034424853,
1337
+ "grad_norm": 0.17567159235477448,
1338
+ "learning_rate": 6.956098498760389e-07,
1339
+ "loss": 0.0105,
1340
+ "step": 1900
1341
+ },
1342
+ {
1343
+ "epoch": 0.8018471872376155,
1344
+ "grad_norm": 0.17301349341869354,
1345
+ "learning_rate": 5.64931531311741e-07,
1346
+ "loss": 0.0072,
1347
+ "step": 1910
1348
+ },
1349
+ {
1350
+ "epoch": 0.8060453400503779,
1351
+ "grad_norm": 0.13477115333080292,
1352
+ "learning_rate": 4.4776826686069305e-07,
1353
+ "loss": 0.008,
1354
+ "step": 1920
1355
+ },
1356
+ {
1357
+ "epoch": 0.8102434928631402,
1358
+ "grad_norm": 0.16665753722190857,
1359
+ "learning_rate": 3.441520877895288e-07,
1360
+ "loss": 0.0078,
1361
+ "step": 1930
1362
+ },
1363
+ {
1364
+ "epoch": 0.8144416456759026,
1365
+ "grad_norm": 0.20697897672653198,
1366
+ "learning_rate": 2.5411132172700194e-07,
1367
+ "loss": 0.0079,
1368
+ "step": 1940
1369
+ },
1370
+ {
1371
+ "epoch": 0.818639798488665,
1372
+ "grad_norm": 0.23099485039710999,
1373
+ "learning_rate": 1.776705849195037e-07,
1374
+ "loss": 0.0104,
1375
+ "step": 1950
1376
+ },
1377
+ {
1378
+ "epoch": 0.8228379513014273,
1379
+ "grad_norm": 0.16362063586711884,
1380
+ "learning_rate": 1.1485077550122402e-07,
1381
+ "loss": 0.0057,
1382
+ "step": 1960
1383
+ },
1384
+ {
1385
+ "epoch": 0.8270361041141897,
1386
+ "grad_norm": 0.12974146008491516,
1387
+ "learning_rate": 6.566906778079917e-08,
1388
+ "loss": 0.0079,
1389
+ "step": 1970
1390
+ },
1391
+ {
1392
+ "epoch": 0.8312342569269522,
1393
+ "grad_norm": 0.1990516036748886,
1394
+ "learning_rate": 3.01389075460512e-08,
1395
+ "loss": 0.0101,
1396
+ "step": 1980
1397
+ },
1398
+ {
1399
+ "epoch": 0.8354324097397146,
1400
+ "grad_norm": 0.15594905614852905,
1401
+ "learning_rate": 8.270008388022721e-09,
1402
+ "loss": 0.0065,
1403
+ "step": 1990
1404
+ },
1405
+ {
1406
+ "epoch": 0.8396305625524769,
1407
+ "grad_norm": 0.07478626072406769,
1408
+ "learning_rate": 6.834904537900144e-11,
1409
+ "loss": 0.0054,
1410
+ "step": 2000
1411
+ },
1412
+ {
1413
+ "epoch": 0.8396305625524769,
1414
+ "step": 2000,
1415
+ "total_flos": 0.0,
1416
+ "train_loss": 0.030227677578106522,
1417
+ "train_runtime": 2758.5082,
1418
+ "train_samples_per_second": 23.201,
1419
+ "train_steps_per_second": 0.725
1420
+ }
1421
+ ],
1422
+ "logging_steps": 10,
1423
+ "max_steps": 2000,
1424
+ "num_input_tokens_seen": 0,
1425
+ "num_train_epochs": 1,
1426
+ "save_steps": 1000,
1427
+ "stateful_callbacks": {
1428
+ "TrainerControl": {
1429
+ "args": {
1430
+ "should_epoch_stop": false,
1431
+ "should_evaluate": false,
1432
+ "should_log": false,
1433
+ "should_save": true,
1434
+ "should_training_stop": true
1435
+ },
1436
+ "attributes": {}
1437
+ }
1438
+ },
1439
+ "total_flos": 0.0,
1440
+ "train_batch_size": 32,
1441
+ "trial_name": null,
1442
+ "trial_params": null
1443
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b470d11cba755fbf9e72e14f85a525dd8987613da7de8d492bcec919d2af9aa
3
+ size 5368