Add files using upload-large-folder tool
Browse files- experiment_cfg/metadata.json +79 -71
- model-00001-of-00002.safetensors +1 -1
- model-00002-of-00002.safetensors +2 -2
- model.safetensors.index.json +2 -1
- optimizer.pt +2 -2
- rng_state_0.pth +1 -1
- rng_state_1.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +1803 -403
experiment_cfg/metadata.json
CHANGED
|
@@ -4,139 +4,147 @@
|
|
| 4 |
"state": {
|
| 5 |
"single_arm": {
|
| 6 |
"max": [
|
| 7 |
-
2.
|
| 8 |
-
2.
|
| 9 |
-
-
|
| 10 |
-
1.
|
| 11 |
-
1.
|
| 12 |
],
|
| 13 |
"min": [
|
| 14 |
-
-1.
|
| 15 |
-
-0.
|
| 16 |
-
-2.
|
| 17 |
-
-0.
|
| 18 |
-
-2.
|
| 19 |
],
|
| 20 |
"mean": [
|
| 21 |
-
0.
|
| 22 |
-
1.
|
| 23 |
-
-1.
|
| 24 |
-
0.
|
| 25 |
-
-0.
|
| 26 |
],
|
| 27 |
"std": [
|
| 28 |
-
0.
|
| 29 |
-
0.
|
| 30 |
-
0.
|
| 31 |
-
0.
|
| 32 |
-
0.
|
| 33 |
],
|
| 34 |
"q01": [
|
| 35 |
-
-1.
|
| 36 |
-
-1.
|
| 37 |
-
-2.
|
| 38 |
-
0.
|
| 39 |
-
-1.
|
| 40 |
],
|
| 41 |
"q99": [
|
| 42 |
-
1.
|
| 43 |
-
2.
|
| 44 |
-
-0.
|
| 45 |
1.2000004053115845,
|
| 46 |
-
1.
|
| 47 |
]
|
| 48 |
},
|
| 49 |
"gripper": {
|
| 50 |
"max": [
|
| 51 |
-
0.
|
| 52 |
],
|
| 53 |
"min": [
|
| 54 |
-
-0.
|
| 55 |
],
|
| 56 |
"mean": [
|
| 57 |
-
0.
|
| 58 |
],
|
| 59 |
"std": [
|
| 60 |
-
0.
|
| 61 |
],
|
| 62 |
"q01": [
|
| 63 |
-
-0.
|
| 64 |
],
|
| 65 |
"q99": [
|
| 66 |
-
0.
|
| 67 |
]
|
| 68 |
}
|
| 69 |
},
|
| 70 |
"action": {
|
| 71 |
"single_arm": {
|
| 72 |
"max": [
|
| 73 |
-
2.
|
| 74 |
-
2.
|
| 75 |
-
-0.
|
| 76 |
-
1.
|
| 77 |
-
1.
|
| 78 |
],
|
| 79 |
"min": [
|
| 80 |
-
-1.
|
| 81 |
-
-0.
|
| 82 |
-
-2.
|
| 83 |
-
-0.
|
| 84 |
-
-2.
|
| 85 |
],
|
| 86 |
"mean": [
|
| 87 |
-
0.
|
| 88 |
-
1.
|
| 89 |
-
-1.
|
| 90 |
-
0.
|
| 91 |
-
-0.
|
| 92 |
],
|
| 93 |
"std": [
|
| 94 |
-
0.
|
| 95 |
-
0.
|
| 96 |
-
0.
|
| 97 |
-
0.
|
| 98 |
-
0.
|
| 99 |
],
|
| 100 |
"q01": [
|
| 101 |
-
-1.
|
| 102 |
-
-
|
| 103 |
-
-2.
|
| 104 |
-
0.
|
| 105 |
-
-1.
|
| 106 |
],
|
| 107 |
"q99": [
|
| 108 |
-
1.
|
| 109 |
-
2.
|
| 110 |
-
-0.
|
| 111 |
1.200000286102295,
|
| 112 |
-
1.
|
| 113 |
]
|
| 114 |
},
|
| 115 |
"gripper": {
|
| 116 |
"max": [
|
| 117 |
-
0.
|
| 118 |
],
|
| 119 |
"min": [
|
| 120 |
-
-0.
|
| 121 |
],
|
| 122 |
"mean": [
|
| 123 |
-
0.
|
| 124 |
],
|
| 125 |
"std": [
|
| 126 |
-
0.
|
| 127 |
],
|
| 128 |
"q01": [
|
| 129 |
-
-0.
|
| 130 |
],
|
| 131 |
"q99": [
|
| 132 |
-
0.
|
| 133 |
]
|
| 134 |
}
|
| 135 |
}
|
| 136 |
},
|
| 137 |
"modalities": {
|
| 138 |
"video": {
|
| 139 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 140 |
"resolution": [
|
| 141 |
640,
|
| 142 |
480
|
|
|
|
| 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
|
model-00001-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999367032
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e14836ce74a4e9b04da62ca4114faf735a2809c0b08a9108f38af75d77cd30d7
|
| 3 |
size 4999367032
|
model-00002-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0453b04145be502d125d4232207a885382b890f223ea28e76aaa9b9dec0b7a7d
|
| 3 |
+
size 2586705312
|
model.safetensors.index.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"metadata": {
|
| 3 |
-
"total_size":
|
| 4 |
},
|
| 5 |
"weight_map": {
|
| 6 |
"action_head.action_decoder.layer1.W": "model-00002-of-00002.safetensors",
|
|
@@ -13,6 +13,7 @@
|
|
| 13 |
"action_head.action_encoder.W2.b": "model-00002-of-00002.safetensors",
|
| 14 |
"action_head.action_encoder.W3.W": "model-00002-of-00002.safetensors",
|
| 15 |
"action_head.action_encoder.W3.b": "model-00002-of-00002.safetensors",
|
|
|
|
| 16 |
"action_head.model.proj_out_1.bias": "model-00002-of-00002.safetensors",
|
| 17 |
"action_head.model.proj_out_1.weight": "model-00002-of-00002.safetensors",
|
| 18 |
"action_head.model.proj_out_2.bias": "model-00002-of-00002.safetensors",
|
|
|
|
| 1 |
{
|
| 2 |
"metadata": {
|
| 3 |
+
"total_size": 7585939328
|
| 4 |
},
|
| 5 |
"weight_map": {
|
| 6 |
"action_head.action_decoder.layer1.W": "model-00002-of-00002.safetensors",
|
|
|
|
| 13 |
"action_head.action_encoder.W2.b": "model-00002-of-00002.safetensors",
|
| 14 |
"action_head.action_encoder.W3.W": "model-00002-of-00002.safetensors",
|
| 15 |
"action_head.action_encoder.W3.b": "model-00002-of-00002.safetensors",
|
| 16 |
+
"action_head.future_tokens.weight": "model-00002-of-00002.safetensors",
|
| 17 |
"action_head.model.proj_out_1.bias": "model-00002-of-00002.safetensors",
|
| 18 |
"action_head.model.proj_out_1.weight": "model-00002-of-00002.safetensors",
|
| 19 |
"action_head.model.proj_out_2.bias": "model-00002-of-00002.safetensors",
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4fff16306a8567006573dec88809781da838424eb8893b8ac50c99c1707d8bf
|
| 3 |
+
size 8550720062
|
rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa987f049de3910e95fa2a955bc4440367b40904739310b87f0dcc5b65565405
|
| 3 |
size 14512
|
rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14512
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a53f5bc2d14df97b7c41aafe1a226f125dc1e3b00e93b40e785ab358650080f3
|
| 3 |
size 14512
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f3bbae53c08b86f610006e2dd1b837f5d64ea03e1ff2ff6562b0e6794c52c092
|
| 3 |
size 1064
|
trainer_state.json
CHANGED
|
@@ -2,718 +2,2118 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 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.
|
| 14 |
-
"grad_norm":
|
| 15 |
-
"learning_rate":
|
| 16 |
-
"loss":
|
| 17 |
"step": 10
|
| 18 |
},
|
| 19 |
{
|
| 20 |
-
"epoch": 0.
|
| 21 |
-
"grad_norm": 2.
|
| 22 |
-
"learning_rate":
|
| 23 |
-
"loss": 0.
|
| 24 |
"step": 20
|
| 25 |
},
|
| 26 |
{
|
| 27 |
-
"epoch": 0.
|
| 28 |
-
"grad_norm": 1.
|
| 29 |
-
"learning_rate": 1.
|
| 30 |
-
"loss": 0.
|
| 31 |
"step": 30
|
| 32 |
},
|
| 33 |
{
|
| 34 |
-
"epoch": 0.
|
| 35 |
-
"grad_norm": 1.
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss": 0.
|
| 38 |
"step": 40
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 0.
|
| 42 |
-
"grad_norm": 1.
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 50
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"grad_norm":
|
| 50 |
-
"learning_rate":
|
| 51 |
-
"loss": 0.
|
| 52 |
"step": 60
|
| 53 |
},
|
| 54 |
{
|
| 55 |
-
"epoch": 0.
|
| 56 |
-
"grad_norm":
|
| 57 |
-
"learning_rate":
|
| 58 |
-
"loss": 0.
|
| 59 |
"step": 70
|
| 60 |
},
|
| 61 |
{
|
| 62 |
-
"epoch": 0.
|
| 63 |
-
"grad_norm":
|
| 64 |
-
"learning_rate":
|
| 65 |
-
"loss": 0.
|
| 66 |
"step": 80
|
| 67 |
},
|
| 68 |
{
|
| 69 |
-
"epoch": 0.
|
| 70 |
-
"grad_norm":
|
| 71 |
-
"learning_rate":
|
| 72 |
-
"loss": 0.
|
| 73 |
"step": 90
|
| 74 |
},
|
| 75 |
{
|
| 76 |
-
"epoch": 0.
|
| 77 |
-
"grad_norm": 0.
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 0.
|
| 80 |
"step": 100
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0.
|
| 84 |
-
"grad_norm": 0.
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 110
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch": 0.
|
| 91 |
-
"grad_norm": 0.
|
| 92 |
-
"learning_rate":
|
| 93 |
-
"loss": 0.
|
| 94 |
"step": 120
|
| 95 |
},
|
| 96 |
{
|
| 97 |
-
"epoch": 0.
|
| 98 |
-
"grad_norm": 0.
|
| 99 |
-
"learning_rate":
|
| 100 |
-
"loss": 0.
|
| 101 |
"step": 130
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch": 0.
|
| 105 |
-
"grad_norm": 0.
|
| 106 |
-
"learning_rate":
|
| 107 |
-
"loss": 0.
|
| 108 |
"step": 140
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch": 0.
|
| 112 |
-
"grad_norm": 0.
|
| 113 |
-
"learning_rate":
|
| 114 |
-
"loss": 0.
|
| 115 |
"step": 150
|
| 116 |
},
|
| 117 |
{
|
| 118 |
-
"epoch": 0.
|
| 119 |
-
"grad_norm": 0.
|
| 120 |
-
"learning_rate":
|
| 121 |
-
"loss": 0.
|
| 122 |
"step": 160
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 0.
|
| 126 |
-
"grad_norm": 0.
|
| 127 |
-
"learning_rate":
|
| 128 |
-
"loss": 0.
|
| 129 |
"step": 170
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
-
"grad_norm": 0.
|
| 134 |
-
"learning_rate":
|
| 135 |
-
"loss": 0.
|
| 136 |
"step": 180
|
| 137 |
},
|
| 138 |
{
|
| 139 |
-
"epoch": 0.
|
| 140 |
-
"grad_norm":
|
| 141 |
-
"learning_rate": 9.
|
| 142 |
-
"loss": 0.
|
| 143 |
"step": 190
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch": 0.
|
| 147 |
-
"grad_norm": 0.
|
| 148 |
-
"learning_rate": 9.
|
| 149 |
-
"loss": 0.
|
| 150 |
"step": 200
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"epoch": 0.
|
| 154 |
-
"grad_norm": 0.
|
| 155 |
-
"learning_rate": 9.
|
| 156 |
-
"loss": 0.
|
| 157 |
"step": 210
|
| 158 |
},
|
| 159 |
{
|
| 160 |
-
"epoch": 0.
|
| 161 |
-
"grad_norm": 0.
|
| 162 |
-
"learning_rate": 9.
|
| 163 |
-
"loss": 0.
|
| 164 |
"step": 220
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch": 0.
|
| 168 |
-
"grad_norm": 0.
|
| 169 |
-
"learning_rate": 9.
|
| 170 |
-
"loss": 0.
|
| 171 |
"step": 230
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch": 0.
|
| 175 |
-
"grad_norm": 0.
|
| 176 |
-
"learning_rate": 9.
|
| 177 |
-
"loss": 0.
|
| 178 |
"step": 240
|
| 179 |
},
|
| 180 |
{
|
| 181 |
-
"epoch": 0.
|
| 182 |
-
"grad_norm": 0.
|
| 183 |
-
"learning_rate": 9.
|
| 184 |
-
"loss": 0.
|
| 185 |
"step": 250
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch": 0.
|
| 189 |
-
"grad_norm": 0.
|
| 190 |
-
"learning_rate": 9.
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 260
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch": 0.
|
| 196 |
-
"grad_norm": 0.
|
| 197 |
-
"learning_rate": 9.
|
| 198 |
-
"loss": 0.
|
| 199 |
"step": 270
|
| 200 |
},
|
| 201 |
{
|
| 202 |
-
"epoch": 0.
|
| 203 |
-
"grad_norm": 0.
|
| 204 |
-
"learning_rate": 9.
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 280
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch": 0.
|
| 210 |
-
"grad_norm": 0.
|
| 211 |
-
"learning_rate": 9.
|
| 212 |
-
"loss": 0.
|
| 213 |
"step": 290
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 0.
|
| 217 |
-
"grad_norm": 0.
|
| 218 |
-
"learning_rate": 9.
|
| 219 |
-
"loss": 0.
|
| 220 |
"step": 300
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"epoch": 0.
|
| 224 |
-
"grad_norm": 0.
|
| 225 |
-
"learning_rate": 9.
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 310
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch": 0.
|
| 231 |
-
"grad_norm": 0.
|
| 232 |
-
"learning_rate": 9.
|
| 233 |
-
"loss": 0.
|
| 234 |
"step": 320
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 0.
|
| 238 |
-
"grad_norm": 0.
|
| 239 |
-
"learning_rate": 9.
|
| 240 |
-
"loss": 0.
|
| 241 |
"step": 330
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch": 0.
|
| 245 |
-
"grad_norm": 0.
|
| 246 |
-
"learning_rate": 9.
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 340
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch": 0.
|
| 252 |
-
"grad_norm": 0.
|
| 253 |
-
"learning_rate": 9.
|
| 254 |
-
"loss": 0.
|
| 255 |
"step": 350
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch": 0.
|
| 259 |
-
"grad_norm": 0.
|
| 260 |
-
"learning_rate": 9.
|
| 261 |
-
"loss": 0.
|
| 262 |
"step": 360
|
| 263 |
},
|
| 264 |
{
|
| 265 |
-
"epoch": 0.
|
| 266 |
-
"grad_norm": 0.
|
| 267 |
-
"learning_rate": 9.
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 370
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch": 0.
|
| 273 |
-
"grad_norm": 0.
|
| 274 |
-
"learning_rate": 9.
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 380
|
| 277 |
},
|
| 278 |
{
|
| 279 |
-
"epoch": 0.
|
| 280 |
-
"grad_norm": 0.
|
| 281 |
-
"learning_rate": 9.
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 390
|
| 284 |
},
|
| 285 |
{
|
| 286 |
-
"epoch": 0.
|
| 287 |
-
"grad_norm": 0.
|
| 288 |
-
"learning_rate": 9.
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 400
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch": 0.
|
| 294 |
-
"grad_norm": 0.
|
| 295 |
-
"learning_rate": 9.
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 410
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch": 0.
|
| 301 |
-
"grad_norm": 0.
|
| 302 |
-
"learning_rate": 9.
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 420
|
| 305 |
},
|
| 306 |
{
|
| 307 |
-
"epoch": 0.
|
| 308 |
-
"grad_norm": 0.
|
| 309 |
-
"learning_rate": 9.
|
| 310 |
-
"loss": 0.
|
| 311 |
"step": 430
|
| 312 |
},
|
| 313 |
{
|
| 314 |
-
"epoch": 0.
|
| 315 |
-
"grad_norm": 0.
|
| 316 |
-
"learning_rate": 9.
|
| 317 |
-
"loss": 0.
|
| 318 |
"step": 440
|
| 319 |
},
|
| 320 |
{
|
| 321 |
-
"epoch": 0.
|
| 322 |
-
"grad_norm": 0.
|
| 323 |
-
"learning_rate": 9.
|
| 324 |
-
"loss": 0.
|
| 325 |
"step": 450
|
| 326 |
},
|
| 327 |
{
|
| 328 |
-
"epoch": 0.
|
| 329 |
-
"grad_norm": 0.
|
| 330 |
-
"learning_rate": 9.
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 460
|
| 333 |
},
|
| 334 |
{
|
| 335 |
-
"epoch": 0.
|
| 336 |
-
"grad_norm": 0.
|
| 337 |
-
"learning_rate": 9.
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 470
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch": 0.
|
| 343 |
-
"grad_norm": 0.
|
| 344 |
-
"learning_rate": 9.
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 480
|
| 347 |
},
|
| 348 |
{
|
| 349 |
-
"epoch": 0.
|
| 350 |
-
"grad_norm": 0.
|
| 351 |
-
"learning_rate": 9.
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 490
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch": 0.
|
| 357 |
-
"grad_norm": 0.
|
| 358 |
-
"learning_rate": 9.
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 500
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch": 0.
|
| 364 |
-
"grad_norm": 0.
|
| 365 |
-
"learning_rate": 9.
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 510
|
| 368 |
},
|
| 369 |
{
|
| 370 |
-
"epoch": 0.
|
| 371 |
-
"grad_norm": 0.
|
| 372 |
-
"learning_rate": 9.
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 520
|
| 375 |
},
|
| 376 |
{
|
| 377 |
-
"epoch": 0.
|
| 378 |
-
"grad_norm": 0.
|
| 379 |
-
"learning_rate": 9.
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 530
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch": 0.
|
| 385 |
-
"grad_norm": 0.
|
| 386 |
-
"learning_rate": 9.
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 540
|
| 389 |
},
|
| 390 |
{
|
| 391 |
-
"epoch": 0.
|
| 392 |
-
"grad_norm": 0.
|
| 393 |
-
"learning_rate": 9.
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 550
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch": 0.
|
| 399 |
-
"grad_norm": 0.
|
| 400 |
-
"learning_rate": 9.
|
| 401 |
-
"loss": 0.
|
| 402 |
"step": 560
|
| 403 |
},
|
| 404 |
{
|
| 405 |
-
"epoch": 0.
|
| 406 |
-
"grad_norm": 0.
|
| 407 |
-
"learning_rate": 9.
|
| 408 |
-
"loss": 0.
|
| 409 |
"step": 570
|
| 410 |
},
|
| 411 |
{
|
| 412 |
-
"epoch": 0.
|
| 413 |
-
"grad_norm": 0.
|
| 414 |
-
"learning_rate": 9.
|
| 415 |
-
"loss": 0.
|
| 416 |
"step": 580
|
| 417 |
},
|
| 418 |
{
|
| 419 |
-
"epoch": 0.
|
| 420 |
-
"grad_norm": 0.
|
| 421 |
-
"learning_rate": 9.
|
| 422 |
-
"loss": 0.
|
| 423 |
"step": 590
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch": 0.
|
| 427 |
-
"grad_norm": 0.
|
| 428 |
-
"learning_rate": 9.
|
| 429 |
-
"loss": 0.
|
| 430 |
"step": 600
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"epoch": 0.
|
| 434 |
-
"grad_norm": 0.
|
| 435 |
-
"learning_rate": 9.
|
| 436 |
-
"loss": 0.
|
| 437 |
"step": 610
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch": 0.
|
| 441 |
-
"grad_norm": 0.
|
| 442 |
-
"learning_rate": 9.
|
| 443 |
-
"loss": 0.
|
| 444 |
"step": 620
|
| 445 |
},
|
| 446 |
{
|
| 447 |
-
"epoch": 0.
|
| 448 |
-
"grad_norm": 0.
|
| 449 |
-
"learning_rate": 9.
|
| 450 |
-
"loss": 0.
|
| 451 |
"step": 630
|
| 452 |
},
|
| 453 |
{
|
| 454 |
-
"epoch": 0.
|
| 455 |
-
"grad_norm": 0.
|
| 456 |
-
"learning_rate": 9.
|
| 457 |
-
"loss": 0.
|
| 458 |
"step": 640
|
| 459 |
},
|
| 460 |
{
|
| 461 |
-
"epoch": 0.
|
| 462 |
-
"grad_norm": 0.
|
| 463 |
-
"learning_rate": 9.
|
| 464 |
-
"loss": 0.
|
| 465 |
"step": 650
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch": 0.
|
| 469 |
-
"grad_norm": 0.
|
| 470 |
-
"learning_rate": 9.
|
| 471 |
-
"loss": 0.
|
| 472 |
"step": 660
|
| 473 |
},
|
| 474 |
{
|
| 475 |
-
"epoch": 0.
|
| 476 |
-
"grad_norm": 0.
|
| 477 |
-
"learning_rate": 9.
|
| 478 |
-
"loss": 0.
|
| 479 |
"step": 670
|
| 480 |
},
|
| 481 |
{
|
| 482 |
-
"epoch": 0.
|
| 483 |
-
"grad_norm": 0.
|
| 484 |
-
"learning_rate": 9.
|
| 485 |
-
"loss": 0.
|
| 486 |
"step": 680
|
| 487 |
},
|
| 488 |
{
|
| 489 |
-
"epoch": 0.
|
| 490 |
-
"grad_norm": 0.
|
| 491 |
-
"learning_rate": 9.
|
| 492 |
-
"loss": 0.
|
| 493 |
"step": 690
|
| 494 |
},
|
| 495 |
{
|
| 496 |
-
"epoch": 0.
|
| 497 |
-
"grad_norm": 0.
|
| 498 |
-
"learning_rate": 9.
|
| 499 |
-
"loss": 0.
|
| 500 |
"step": 700
|
| 501 |
},
|
| 502 |
{
|
| 503 |
-
"epoch": 0.
|
| 504 |
-
"grad_norm": 0.
|
| 505 |
-
"learning_rate": 9.
|
| 506 |
-
"loss": 0.
|
| 507 |
"step": 710
|
| 508 |
},
|
| 509 |
{
|
| 510 |
-
"epoch": 0.
|
| 511 |
-
"grad_norm": 0.
|
| 512 |
-
"learning_rate": 9.
|
| 513 |
-
"loss": 0.
|
| 514 |
"step": 720
|
| 515 |
},
|
| 516 |
{
|
| 517 |
-
"epoch": 0.
|
| 518 |
-
"grad_norm": 0.
|
| 519 |
-
"learning_rate": 9.
|
| 520 |
-
"loss": 0.
|
| 521 |
"step": 730
|
| 522 |
},
|
| 523 |
{
|
| 524 |
-
"epoch": 0.
|
| 525 |
-
"grad_norm": 0.
|
| 526 |
-
"learning_rate":
|
| 527 |
-
"loss": 0.
|
| 528 |
"step": 740
|
| 529 |
},
|
| 530 |
{
|
| 531 |
-
"epoch": 0.
|
| 532 |
-
"grad_norm": 0.
|
| 533 |
-
"learning_rate":
|
| 534 |
-
"loss": 0.
|
| 535 |
"step": 750
|
| 536 |
},
|
| 537 |
{
|
| 538 |
-
"epoch": 0.
|
| 539 |
-
"grad_norm": 0.
|
| 540 |
-
"learning_rate":
|
| 541 |
-
"loss": 0.
|
| 542 |
"step": 760
|
| 543 |
},
|
| 544 |
{
|
| 545 |
-
"epoch": 0.
|
| 546 |
-
"grad_norm": 0.
|
| 547 |
-
"learning_rate":
|
| 548 |
-
"loss": 0.
|
| 549 |
"step": 770
|
| 550 |
},
|
| 551 |
{
|
| 552 |
-
"epoch": 0.
|
| 553 |
-
"grad_norm": 0.
|
| 554 |
-
"learning_rate":
|
| 555 |
-
"loss": 0.
|
| 556 |
"step": 780
|
| 557 |
},
|
| 558 |
{
|
| 559 |
-
"epoch": 0.
|
| 560 |
-
"grad_norm": 0.
|
| 561 |
-
"learning_rate":
|
| 562 |
-
"loss": 0.
|
| 563 |
"step": 790
|
| 564 |
},
|
| 565 |
{
|
| 566 |
-
"epoch": 0.
|
| 567 |
-
"grad_norm": 0.
|
| 568 |
-
"learning_rate":
|
| 569 |
"loss": 0.0173,
|
| 570 |
"step": 800
|
| 571 |
},
|
| 572 |
{
|
| 573 |
-
"epoch": 0.
|
| 574 |
-
"grad_norm": 0.
|
| 575 |
-
"learning_rate":
|
| 576 |
-
"loss": 0.
|
| 577 |
"step": 810
|
| 578 |
},
|
| 579 |
{
|
| 580 |
-
"epoch": 0.
|
| 581 |
-
"grad_norm": 0.
|
| 582 |
-
"learning_rate":
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 820
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch": 0.
|
| 588 |
-
"grad_norm": 0.
|
| 589 |
-
"learning_rate":
|
| 590 |
-
"loss": 0.
|
| 591 |
"step": 830
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch": 0.
|
| 595 |
-
"grad_norm": 0.
|
| 596 |
-
"learning_rate":
|
| 597 |
-
"loss": 0.
|
| 598 |
"step": 840
|
| 599 |
},
|
| 600 |
{
|
| 601 |
-
"epoch": 0.
|
| 602 |
-
"grad_norm": 0.
|
| 603 |
-
"learning_rate":
|
| 604 |
-
"loss": 0.
|
| 605 |
"step": 850
|
| 606 |
},
|
| 607 |
{
|
| 608 |
-
"epoch": 0.
|
| 609 |
-
"grad_norm": 0.
|
| 610 |
-
"learning_rate":
|
| 611 |
-
"loss": 0.
|
| 612 |
"step": 860
|
| 613 |
},
|
| 614 |
{
|
| 615 |
-
"epoch": 0.
|
| 616 |
-
"grad_norm": 0.
|
| 617 |
-
"learning_rate":
|
| 618 |
-
"loss": 0.
|
| 619 |
"step": 870
|
| 620 |
},
|
| 621 |
{
|
| 622 |
-
"epoch": 0.
|
| 623 |
-
"grad_norm": 0.
|
| 624 |
-
"learning_rate":
|
| 625 |
-
"loss": 0.
|
| 626 |
"step": 880
|
| 627 |
},
|
| 628 |
{
|
| 629 |
-
"epoch": 0.
|
| 630 |
-
"grad_norm": 0.
|
| 631 |
-
"learning_rate":
|
| 632 |
-
"loss": 0.
|
| 633 |
"step": 890
|
| 634 |
},
|
| 635 |
{
|
| 636 |
-
"epoch": 0.
|
| 637 |
-
"grad_norm": 0.
|
| 638 |
-
"learning_rate":
|
| 639 |
-
"loss": 0.
|
| 640 |
"step": 900
|
| 641 |
},
|
| 642 |
{
|
| 643 |
-
"epoch": 0.
|
| 644 |
-
"grad_norm": 0.
|
| 645 |
-
"learning_rate":
|
| 646 |
-
"loss": 0.
|
| 647 |
"step": 910
|
| 648 |
},
|
| 649 |
{
|
| 650 |
-
"epoch": 0.
|
| 651 |
-
"grad_norm": 0.
|
| 652 |
-
"learning_rate":
|
| 653 |
-
"loss": 0.
|
| 654 |
"step": 920
|
| 655 |
},
|
| 656 |
{
|
| 657 |
-
"epoch": 0.
|
| 658 |
-
"grad_norm": 0.
|
| 659 |
-
"learning_rate":
|
| 660 |
-
"loss": 0.
|
| 661 |
"step": 930
|
| 662 |
},
|
| 663 |
{
|
| 664 |
-
"epoch": 0.
|
| 665 |
-
"grad_norm": 0.
|
| 666 |
-
"learning_rate":
|
| 667 |
-
"loss": 0.
|
| 668 |
"step": 940
|
| 669 |
},
|
| 670 |
{
|
| 671 |
-
"epoch": 0.
|
| 672 |
-
"grad_norm": 0.
|
| 673 |
-
"learning_rate":
|
| 674 |
-
"loss": 0.
|
| 675 |
"step": 950
|
| 676 |
},
|
| 677 |
{
|
| 678 |
-
"epoch": 0.
|
| 679 |
-
"grad_norm": 0.
|
| 680 |
-
"learning_rate":
|
| 681 |
-
"loss": 0.
|
| 682 |
"step": 960
|
| 683 |
},
|
| 684 |
{
|
| 685 |
-
"epoch": 0.
|
| 686 |
-
"grad_norm": 0.
|
| 687 |
-
"learning_rate":
|
| 688 |
-
"loss": 0.
|
| 689 |
"step": 970
|
| 690 |
},
|
| 691 |
{
|
| 692 |
-
"epoch": 0.
|
| 693 |
-
"grad_norm": 0.
|
| 694 |
-
"learning_rate": 8.
|
| 695 |
-
"loss": 0.
|
| 696 |
"step": 980
|
| 697 |
},
|
| 698 |
{
|
| 699 |
-
"epoch": 0.
|
| 700 |
-
"grad_norm": 0.
|
| 701 |
-
"learning_rate": 8.
|
| 702 |
-
"loss": 0.
|
| 703 |
"step": 990
|
| 704 |
},
|
| 705 |
{
|
| 706 |
-
"epoch": 0.
|
| 707 |
-
"grad_norm": 0.
|
| 708 |
-
"learning_rate":
|
| 709 |
-
"loss": 0.
|
| 710 |
"step": 1000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 711 |
}
|
| 712 |
],
|
| 713 |
"logging_steps": 10,
|
| 714 |
-
"max_steps":
|
| 715 |
"num_input_tokens_seen": 0,
|
| 716 |
-
"num_train_epochs":
|
| 717 |
"save_steps": 1000,
|
| 718 |
"stateful_callbacks": {
|
| 719 |
"TrainerControl": {
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.5188916876574305,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 3000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
+
"epoch": 0.008396305625524769,
|
| 14 |
+
"grad_norm": 10.171950340270996,
|
| 15 |
+
"learning_rate": 4.5e-06,
|
| 16 |
+
"loss": 1.1261,
|
| 17 |
"step": 10
|
| 18 |
},
|
| 19 |
{
|
| 20 |
+
"epoch": 0.016792611251049538,
|
| 21 |
+
"grad_norm": 2.374096632003784,
|
| 22 |
+
"learning_rate": 9.5e-06,
|
| 23 |
+
"loss": 0.5171,
|
| 24 |
"step": 20
|
| 25 |
},
|
| 26 |
{
|
| 27 |
+
"epoch": 0.02518891687657431,
|
| 28 |
+
"grad_norm": 1.9998104572296143,
|
| 29 |
+
"learning_rate": 1.45e-05,
|
| 30 |
+
"loss": 0.2565,
|
| 31 |
"step": 30
|
| 32 |
},
|
| 33 |
{
|
| 34 |
+
"epoch": 0.033585222502099076,
|
| 35 |
+
"grad_norm": 1.4298408031463623,
|
| 36 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 37 |
+
"loss": 0.1707,
|
| 38 |
"step": 40
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.041981528127623846,
|
| 42 |
+
"grad_norm": 1.2781989574432373,
|
| 43 |
+
"learning_rate": 2.45e-05,
|
| 44 |
+
"loss": 0.1302,
|
| 45 |
"step": 50
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.05037783375314862,
|
| 49 |
+
"grad_norm": 1.4599987268447876,
|
| 50 |
+
"learning_rate": 2.95e-05,
|
| 51 |
+
"loss": 0.1016,
|
| 52 |
"step": 60
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"epoch": 0.05877413937867338,
|
| 56 |
+
"grad_norm": 0.7745999097824097,
|
| 57 |
+
"learning_rate": 3.45e-05,
|
| 58 |
+
"loss": 0.0848,
|
| 59 |
"step": 70
|
| 60 |
},
|
| 61 |
{
|
| 62 |
+
"epoch": 0.06717044500419815,
|
| 63 |
+
"grad_norm": 1.022971272468567,
|
| 64 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 65 |
+
"loss": 0.0818,
|
| 66 |
"step": 80
|
| 67 |
},
|
| 68 |
{
|
| 69 |
+
"epoch": 0.07556675062972293,
|
| 70 |
+
"grad_norm": 0.4050329923629761,
|
| 71 |
+
"learning_rate": 4.4500000000000004e-05,
|
| 72 |
+
"loss": 0.0635,
|
| 73 |
"step": 90
|
| 74 |
},
|
| 75 |
{
|
| 76 |
+
"epoch": 0.08396305625524769,
|
| 77 |
+
"grad_norm": 0.7355945110321045,
|
| 78 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 79 |
+
"loss": 0.0647,
|
| 80 |
"step": 100
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.09235936188077246,
|
| 84 |
+
"grad_norm": 0.9530194997787476,
|
| 85 |
+
"learning_rate": 5.45e-05,
|
| 86 |
+
"loss": 0.0634,
|
| 87 |
"step": 110
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"epoch": 0.10075566750629723,
|
| 91 |
+
"grad_norm": 0.8249333500862122,
|
| 92 |
+
"learning_rate": 5.95e-05,
|
| 93 |
+
"loss": 0.0556,
|
| 94 |
"step": 120
|
| 95 |
},
|
| 96 |
{
|
| 97 |
+
"epoch": 0.109151973131822,
|
| 98 |
+
"grad_norm": 0.6259945631027222,
|
| 99 |
+
"learning_rate": 6.450000000000001e-05,
|
| 100 |
+
"loss": 0.0554,
|
| 101 |
"step": 130
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 0.11754827875734676,
|
| 105 |
+
"grad_norm": 0.8876557350158691,
|
| 106 |
+
"learning_rate": 6.95e-05,
|
| 107 |
+
"loss": 0.0495,
|
| 108 |
"step": 140
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 0.12594458438287154,
|
| 112 |
+
"grad_norm": 0.777101993560791,
|
| 113 |
+
"learning_rate": 7.450000000000001e-05,
|
| 114 |
+
"loss": 0.0518,
|
| 115 |
"step": 150
|
| 116 |
},
|
| 117 |
{
|
| 118 |
+
"epoch": 0.1343408900083963,
|
| 119 |
+
"grad_norm": 0.9956705570220947,
|
| 120 |
+
"learning_rate": 7.950000000000001e-05,
|
| 121 |
+
"loss": 0.0456,
|
| 122 |
"step": 160
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.14273719563392107,
|
| 126 |
+
"grad_norm": 0.5946229696273804,
|
| 127 |
+
"learning_rate": 8.450000000000001e-05,
|
| 128 |
+
"loss": 0.042,
|
| 129 |
"step": 170
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 0.15113350125944586,
|
| 133 |
+
"grad_norm": 0.7513177990913391,
|
| 134 |
+
"learning_rate": 8.950000000000001e-05,
|
| 135 |
+
"loss": 0.0416,
|
| 136 |
"step": 180
|
| 137 |
},
|
| 138 |
{
|
| 139 |
+
"epoch": 0.15952980688497062,
|
| 140 |
+
"grad_norm": 1.2336539030075073,
|
| 141 |
+
"learning_rate": 9.449999999999999e-05,
|
| 142 |
+
"loss": 0.04,
|
| 143 |
"step": 190
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 0.16792611251049538,
|
| 147 |
+
"grad_norm": 0.6550417542457581,
|
| 148 |
+
"learning_rate": 9.95e-05,
|
| 149 |
+
"loss": 0.0419,
|
| 150 |
"step": 200
|
| 151 |
},
|
| 152 |
{
|
| 153 |
+
"epoch": 0.17632241813602015,
|
| 154 |
+
"grad_norm": 0.7052122354507446,
|
| 155 |
+
"learning_rate": 9.999861593790126e-05,
|
| 156 |
+
"loss": 0.0444,
|
| 157 |
"step": 210
|
| 158 |
},
|
| 159 |
{
|
| 160 |
+
"epoch": 0.1847187237615449,
|
| 161 |
+
"grad_norm": 0.5741733908653259,
|
| 162 |
+
"learning_rate": 9.999383162408304e-05,
|
| 163 |
+
"loss": 0.043,
|
| 164 |
"step": 220
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 0.19311502938706968,
|
| 168 |
+
"grad_norm": 0.7300972938537598,
|
| 169 |
+
"learning_rate": 9.998563029828259e-05,
|
| 170 |
+
"loss": 0.0396,
|
| 171 |
"step": 230
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 0.20151133501259447,
|
| 175 |
+
"grad_norm": 0.35928529500961304,
|
| 176 |
+
"learning_rate": 9.997401252104962e-05,
|
| 177 |
+
"loss": 0.0382,
|
| 178 |
"step": 240
|
| 179 |
},
|
| 180 |
{
|
| 181 |
+
"epoch": 0.20990764063811923,
|
| 182 |
+
"grad_norm": 0.5260980129241943,
|
| 183 |
+
"learning_rate": 9.995897908644378e-05,
|
| 184 |
+
"loss": 0.0332,
|
| 185 |
"step": 250
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 0.218303946263644,
|
| 189 |
+
"grad_norm": 0.4173150956630707,
|
| 190 |
+
"learning_rate": 9.994053102198034e-05,
|
| 191 |
+
"loss": 0.0314,
|
| 192 |
"step": 260
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 0.22670025188916876,
|
| 196 |
+
"grad_norm": 0.6619486808776855,
|
| 197 |
+
"learning_rate": 9.991866958856003e-05,
|
| 198 |
+
"loss": 0.0375,
|
| 199 |
"step": 270
|
| 200 |
},
|
| 201 |
{
|
| 202 |
+
"epoch": 0.23509655751469352,
|
| 203 |
+
"grad_norm": 0.6853200197219849,
|
| 204 |
+
"learning_rate": 9.989339628038276e-05,
|
| 205 |
+
"loss": 0.0348,
|
| 206 |
"step": 280
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 0.2434928631402183,
|
| 210 |
+
"grad_norm": 0.7677571177482605,
|
| 211 |
+
"learning_rate": 9.98647128248456e-05,
|
| 212 |
+
"loss": 0.0434,
|
| 213 |
"step": 290
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 0.2518891687657431,
|
| 217 |
+
"grad_norm": 0.47874388098716736,
|
| 218 |
+
"learning_rate": 9.98326211824246e-05,
|
| 219 |
+
"loss": 0.0339,
|
| 220 |
"step": 300
|
| 221 |
},
|
| 222 |
{
|
| 223 |
+
"epoch": 0.26028547439126787,
|
| 224 |
+
"grad_norm": 0.5458245277404785,
|
| 225 |
+
"learning_rate": 9.979712354654091e-05,
|
| 226 |
+
"loss": 0.0365,
|
| 227 |
"step": 310
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 0.2686817800167926,
|
| 231 |
+
"grad_norm": 0.7390539646148682,
|
| 232 |
+
"learning_rate": 9.975822234341079e-05,
|
| 233 |
+
"loss": 0.0319,
|
| 234 |
"step": 320
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 0.2770780856423174,
|
| 238 |
+
"grad_norm": 0.9258743524551392,
|
| 239 |
+
"learning_rate": 9.97159202318798e-05,
|
| 240 |
+
"loss": 0.0313,
|
| 241 |
"step": 330
|
| 242 |
},
|
| 243 |
{
|
| 244 |
+
"epoch": 0.28547439126784213,
|
| 245 |
+
"grad_norm": 0.4790763556957245,
|
| 246 |
+
"learning_rate": 9.967022010324105e-05,
|
| 247 |
+
"loss": 0.0308,
|
| 248 |
"step": 340
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 0.2938706968933669,
|
| 252 |
+
"grad_norm": 0.5265215635299683,
|
| 253 |
+
"learning_rate": 9.962112508103765e-05,
|
| 254 |
+
"loss": 0.0292,
|
| 255 |
"step": 350
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 0.3022670025188917,
|
| 259 |
+
"grad_norm": 0.5428227782249451,
|
| 260 |
+
"learning_rate": 9.956863852084914e-05,
|
| 261 |
+
"loss": 0.0286,
|
| 262 |
"step": 360
|
| 263 |
},
|
| 264 |
{
|
| 265 |
+
"epoch": 0.31066330814441645,
|
| 266 |
+
"grad_norm": 0.5066298246383667,
|
| 267 |
+
"learning_rate": 9.951276401006221e-05,
|
| 268 |
+
"loss": 0.0316,
|
| 269 |
"step": 370
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 0.31905961376994124,
|
| 273 |
+
"grad_norm": 0.3620965778827667,
|
| 274 |
+
"learning_rate": 9.945350536762543e-05,
|
| 275 |
+
"loss": 0.0251,
|
| 276 |
"step": 380
|
| 277 |
},
|
| 278 |
{
|
| 279 |
+
"epoch": 0.327455919395466,
|
| 280 |
+
"grad_norm": 0.44671374559402466,
|
| 281 |
+
"learning_rate": 9.939086664378829e-05,
|
| 282 |
+
"loss": 0.0271,
|
| 283 |
"step": 390
|
| 284 |
},
|
| 285 |
{
|
| 286 |
+
"epoch": 0.33585222502099077,
|
| 287 |
+
"grad_norm": 0.4878353178501129,
|
| 288 |
+
"learning_rate": 9.932485211982437e-05,
|
| 289 |
+
"loss": 0.0228,
|
| 290 |
"step": 400
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 0.34424853064651556,
|
| 294 |
+
"grad_norm": 0.43716397881507874,
|
| 295 |
+
"learning_rate": 9.92554663077387e-05,
|
| 296 |
+
"loss": 0.0308,
|
| 297 |
"step": 410
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 0.3526448362720403,
|
| 301 |
+
"grad_norm": 0.5659756064414978,
|
| 302 |
+
"learning_rate": 9.918271394995935e-05,
|
| 303 |
+
"loss": 0.0244,
|
| 304 |
"step": 420
|
| 305 |
},
|
| 306 |
{
|
| 307 |
+
"epoch": 0.3610411418975651,
|
| 308 |
+
"grad_norm": 0.4486585259437561,
|
| 309 |
+
"learning_rate": 9.910660001901335e-05,
|
| 310 |
+
"loss": 0.028,
|
| 311 |
"step": 430
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 0.3694374475230898,
|
| 315 |
+
"grad_norm": 0.4010409116744995,
|
| 316 |
+
"learning_rate": 9.902712971718675e-05,
|
| 317 |
+
"loss": 0.0268,
|
| 318 |
"step": 440
|
| 319 |
},
|
| 320 |
{
|
| 321 |
+
"epoch": 0.3778337531486146,
|
| 322 |
+
"grad_norm": 0.48508334159851074,
|
| 323 |
+
"learning_rate": 9.894430847616915e-05,
|
| 324 |
+
"loss": 0.0258,
|
| 325 |
"step": 450
|
| 326 |
},
|
| 327 |
{
|
| 328 |
+
"epoch": 0.38623005877413935,
|
| 329 |
+
"grad_norm": 0.5513168573379517,
|
| 330 |
+
"learning_rate": 9.885814195668232e-05,
|
| 331 |
+
"loss": 0.0288,
|
| 332 |
"step": 460
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 0.39462636439966414,
|
| 336 |
+
"grad_norm": 0.3650225102901459,
|
| 337 |
+
"learning_rate": 9.876863604809344e-05,
|
| 338 |
+
"loss": 0.0284,
|
| 339 |
"step": 470
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 0.40302267002518893,
|
| 343 |
+
"grad_norm": 0.35296711325645447,
|
| 344 |
+
"learning_rate": 9.867579686801245e-05,
|
| 345 |
+
"loss": 0.0246,
|
| 346 |
"step": 480
|
| 347 |
},
|
| 348 |
{
|
| 349 |
+
"epoch": 0.41141897565071367,
|
| 350 |
+
"grad_norm": 0.3748459219932556,
|
| 351 |
+
"learning_rate": 9.8579630761874e-05,
|
| 352 |
+
"loss": 0.0233,
|
| 353 |
"step": 490
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 0.41981528127623846,
|
| 357 |
+
"grad_norm": 0.41755491495132446,
|
| 358 |
+
"learning_rate": 9.848014430250367e-05,
|
| 359 |
+
"loss": 0.023,
|
| 360 |
"step": 500
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 0.4282115869017632,
|
| 364 |
+
"grad_norm": 0.43207406997680664,
|
| 365 |
+
"learning_rate": 9.837734428966885e-05,
|
| 366 |
+
"loss": 0.0245,
|
| 367 |
"step": 510
|
| 368 |
},
|
| 369 |
{
|
| 370 |
+
"epoch": 0.436607892527288,
|
| 371 |
+
"grad_norm": 0.27464979887008667,
|
| 372 |
+
"learning_rate": 9.827123774961383e-05,
|
| 373 |
+
"loss": 0.0216,
|
| 374 |
"step": 520
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 0.4450041981528128,
|
| 378 |
+
"grad_norm": 0.5782731771469116,
|
| 379 |
+
"learning_rate": 9.816183193457968e-05,
|
| 380 |
+
"loss": 0.0245,
|
| 381 |
"step": 530
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 0.4534005037783375,
|
| 385 |
+
"grad_norm": 0.559138834476471,
|
| 386 |
+
"learning_rate": 9.804913432230856e-05,
|
| 387 |
+
"loss": 0.024,
|
| 388 |
"step": 540
|
| 389 |
},
|
| 390 |
{
|
| 391 |
+
"epoch": 0.4617968094038623,
|
| 392 |
+
"grad_norm": 0.6747875809669495,
|
| 393 |
+
"learning_rate": 9.793315261553252e-05,
|
| 394 |
+
"loss": 0.0256,
|
| 395 |
"step": 550
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 0.47019311502938704,
|
| 399 |
+
"grad_norm": 0.55149245262146,
|
| 400 |
+
"learning_rate": 9.781389474144717e-05,
|
| 401 |
+
"loss": 0.0234,
|
| 402 |
"step": 560
|
| 403 |
},
|
| 404 |
{
|
| 405 |
+
"epoch": 0.47858942065491183,
|
| 406 |
+
"grad_norm": 0.32782289385795593,
|
| 407 |
+
"learning_rate": 9.76913688511698e-05,
|
| 408 |
+
"loss": 0.0224,
|
| 409 |
"step": 570
|
| 410 |
},
|
| 411 |
{
|
| 412 |
+
"epoch": 0.4869857262804366,
|
| 413 |
+
"grad_norm": 0.5114140510559082,
|
| 414 |
+
"learning_rate": 9.756558331918227e-05,
|
| 415 |
+
"loss": 0.0236,
|
| 416 |
"step": 580
|
| 417 |
},
|
| 418 |
{
|
| 419 |
+
"epoch": 0.49538203190596136,
|
| 420 |
+
"grad_norm": 0.6146140098571777,
|
| 421 |
+
"learning_rate": 9.743654674275855e-05,
|
| 422 |
+
"loss": 0.0255,
|
| 423 |
"step": 590
|
| 424 |
},
|
| 425 |
{
|
| 426 |
+
"epoch": 0.5037783375314862,
|
| 427 |
+
"grad_norm": 0.7248744964599609,
|
| 428 |
+
"learning_rate": 9.730426794137727e-05,
|
| 429 |
+
"loss": 0.0266,
|
| 430 |
"step": 600
|
| 431 |
},
|
| 432 |
{
|
| 433 |
+
"epoch": 0.5121746431570109,
|
| 434 |
+
"grad_norm": 0.389180451631546,
|
| 435 |
+
"learning_rate": 9.716875595611879e-05,
|
| 436 |
+
"loss": 0.0267,
|
| 437 |
"step": 610
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 0.5205709487825357,
|
| 441 |
+
"grad_norm": 0.49693188071250916,
|
| 442 |
+
"learning_rate": 9.703002004904729e-05,
|
| 443 |
+
"loss": 0.0227,
|
| 444 |
"step": 620
|
| 445 |
},
|
| 446 |
{
|
| 447 |
+
"epoch": 0.5289672544080605,
|
| 448 |
+
"grad_norm": 0.5026288628578186,
|
| 449 |
+
"learning_rate": 9.688806970257773e-05,
|
| 450 |
+
"loss": 0.0197,
|
| 451 |
"step": 630
|
| 452 |
},
|
| 453 |
{
|
| 454 |
+
"epoch": 0.5373635600335852,
|
| 455 |
+
"grad_norm": 0.35680270195007324,
|
| 456 |
+
"learning_rate": 9.674291461882774e-05,
|
| 457 |
+
"loss": 0.0195,
|
| 458 |
"step": 640
|
| 459 |
},
|
| 460 |
{
|
| 461 |
+
"epoch": 0.5457598656591099,
|
| 462 |
+
"grad_norm": 0.3147616386413574,
|
| 463 |
+
"learning_rate": 9.659456471895445e-05,
|
| 464 |
+
"loss": 0.0212,
|
| 465 |
"step": 650
|
| 466 |
},
|
| 467 |
{
|
| 468 |
+
"epoch": 0.5541561712846348,
|
| 469 |
+
"grad_norm": 0.25688570737838745,
|
| 470 |
+
"learning_rate": 9.644303014247648e-05,
|
| 471 |
+
"loss": 0.0184,
|
| 472 |
"step": 660
|
| 473 |
},
|
| 474 |
{
|
| 475 |
+
"epoch": 0.5625524769101595,
|
| 476 |
+
"grad_norm": 0.33005818724632263,
|
| 477 |
+
"learning_rate": 9.628832124658085e-05,
|
| 478 |
+
"loss": 0.0248,
|
| 479 |
"step": 670
|
| 480 |
},
|
| 481 |
{
|
| 482 |
+
"epoch": 0.5709487825356843,
|
| 483 |
+
"grad_norm": 0.3768410384654999,
|
| 484 |
+
"learning_rate": 9.613044860541507e-05,
|
| 485 |
+
"loss": 0.0207,
|
| 486 |
"step": 680
|
| 487 |
},
|
| 488 |
{
|
| 489 |
+
"epoch": 0.5793450881612091,
|
| 490 |
+
"grad_norm": 0.27856314182281494,
|
| 491 |
+
"learning_rate": 9.596942300936445e-05,
|
| 492 |
+
"loss": 0.0174,
|
| 493 |
"step": 690
|
| 494 |
},
|
| 495 |
{
|
| 496 |
+
"epoch": 0.5877413937867338,
|
| 497 |
+
"grad_norm": 0.44634294509887695,
|
| 498 |
+
"learning_rate": 9.580525546431459e-05,
|
| 499 |
+
"loss": 0.0194,
|
| 500 |
"step": 700
|
| 501 |
},
|
| 502 |
{
|
| 503 |
+
"epoch": 0.5961376994122586,
|
| 504 |
+
"grad_norm": 0.35559549927711487,
|
| 505 |
+
"learning_rate": 9.563795719089911e-05,
|
| 506 |
+
"loss": 0.0236,
|
| 507 |
"step": 710
|
| 508 |
},
|
| 509 |
{
|
| 510 |
+
"epoch": 0.6045340050377834,
|
| 511 |
+
"grad_norm": 0.3311902582645416,
|
| 512 |
+
"learning_rate": 9.546753962373281e-05,
|
| 513 |
+
"loss": 0.0175,
|
| 514 |
"step": 720
|
| 515 |
},
|
| 516 |
{
|
| 517 |
+
"epoch": 0.6129303106633082,
|
| 518 |
+
"grad_norm": 0.2235281765460968,
|
| 519 |
+
"learning_rate": 9.529401441062997e-05,
|
| 520 |
+
"loss": 0.018,
|
| 521 |
"step": 730
|
| 522 |
},
|
| 523 |
{
|
| 524 |
+
"epoch": 0.6213266162888329,
|
| 525 |
+
"grad_norm": 0.3235013782978058,
|
| 526 |
+
"learning_rate": 9.511739341180842e-05,
|
| 527 |
+
"loss": 0.0142,
|
| 528 |
"step": 740
|
| 529 |
},
|
| 530 |
{
|
| 531 |
+
"epoch": 0.6297229219143576,
|
| 532 |
+
"grad_norm": 0.4594292640686035,
|
| 533 |
+
"learning_rate": 9.493768869907886e-05,
|
| 534 |
+
"loss": 0.0194,
|
| 535 |
"step": 750
|
| 536 |
},
|
| 537 |
{
|
| 538 |
+
"epoch": 0.6381192275398825,
|
| 539 |
+
"grad_norm": 0.4702000617980957,
|
| 540 |
+
"learning_rate": 9.475491255501968e-05,
|
| 541 |
+
"loss": 0.0149,
|
| 542 |
"step": 760
|
| 543 |
},
|
| 544 |
{
|
| 545 |
+
"epoch": 0.6465155331654072,
|
| 546 |
+
"grad_norm": 0.3968784511089325,
|
| 547 |
+
"learning_rate": 9.456907747213748e-05,
|
| 548 |
+
"loss": 0.0164,
|
| 549 |
"step": 770
|
| 550 |
},
|
| 551 |
{
|
| 552 |
+
"epoch": 0.654911838790932,
|
| 553 |
+
"grad_norm": 0.5368077754974365,
|
| 554 |
+
"learning_rate": 9.438019615201336e-05,
|
| 555 |
+
"loss": 0.0183,
|
| 556 |
"step": 780
|
| 557 |
},
|
| 558 |
{
|
| 559 |
+
"epoch": 0.6633081444164568,
|
| 560 |
+
"grad_norm": 0.32421955466270447,
|
| 561 |
+
"learning_rate": 9.418828150443469e-05,
|
| 562 |
+
"loss": 0.0154,
|
| 563 |
"step": 790
|
| 564 |
},
|
| 565 |
{
|
| 566 |
+
"epoch": 0.6717044500419815,
|
| 567 |
+
"grad_norm": 0.36530694365501404,
|
| 568 |
+
"learning_rate": 9.399334664651262e-05,
|
| 569 |
"loss": 0.0173,
|
| 570 |
"step": 800
|
| 571 |
},
|
| 572 |
{
|
| 573 |
+
"epoch": 0.6801007556675063,
|
| 574 |
+
"grad_norm": 0.24784396588802338,
|
| 575 |
+
"learning_rate": 9.379540490178581e-05,
|
| 576 |
+
"loss": 0.0199,
|
| 577 |
"step": 810
|
| 578 |
},
|
| 579 |
{
|
| 580 |
+
"epoch": 0.6884970612930311,
|
| 581 |
+
"grad_norm": 0.4272478520870209,
|
| 582 |
+
"learning_rate": 9.359446979930955e-05,
|
| 583 |
+
"loss": 0.0178,
|
| 584 |
"step": 820
|
| 585 |
},
|
| 586 |
{
|
| 587 |
+
"epoch": 0.6968933669185559,
|
| 588 |
+
"grad_norm": 0.39507076144218445,
|
| 589 |
+
"learning_rate": 9.33905550727312e-05,
|
| 590 |
+
"loss": 0.0183,
|
| 591 |
"step": 830
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"epoch": 0.7052896725440806,
|
| 595 |
+
"grad_norm": 0.20049844682216644,
|
| 596 |
+
"learning_rate": 9.318367465935142e-05,
|
| 597 |
+
"loss": 0.0156,
|
| 598 |
"step": 840
|
| 599 |
},
|
| 600 |
{
|
| 601 |
+
"epoch": 0.7136859781696053,
|
| 602 |
+
"grad_norm": 0.8157637119293213,
|
| 603 |
+
"learning_rate": 9.29738426991717e-05,
|
| 604 |
+
"loss": 0.0207,
|
| 605 |
"step": 850
|
| 606 |
},
|
| 607 |
{
|
| 608 |
+
"epoch": 0.7220822837951302,
|
| 609 |
+
"grad_norm": 0.5720648765563965,
|
| 610 |
+
"learning_rate": 9.276107353392774e-05,
|
| 611 |
+
"loss": 0.02,
|
| 612 |
"step": 860
|
| 613 |
},
|
| 614 |
{
|
| 615 |
+
"epoch": 0.7304785894206549,
|
| 616 |
+
"grad_norm": 0.7491711974143982,
|
| 617 |
+
"learning_rate": 9.254538170610938e-05,
|
| 618 |
+
"loss": 0.0215,
|
| 619 |
"step": 870
|
| 620 |
},
|
| 621 |
{
|
| 622 |
+
"epoch": 0.7388748950461796,
|
| 623 |
+
"grad_norm": 0.6429204344749451,
|
| 624 |
+
"learning_rate": 9.232678195796654e-05,
|
| 625 |
+
"loss": 0.0161,
|
| 626 |
"step": 880
|
| 627 |
},
|
| 628 |
{
|
| 629 |
+
"epoch": 0.7472712006717045,
|
| 630 |
+
"grad_norm": 0.5460074543952942,
|
| 631 |
+
"learning_rate": 9.210528923050164e-05,
|
| 632 |
+
"loss": 0.014,
|
| 633 |
"step": 890
|
| 634 |
},
|
| 635 |
{
|
| 636 |
+
"epoch": 0.7556675062972292,
|
| 637 |
+
"grad_norm": 0.5716153383255005,
|
| 638 |
+
"learning_rate": 9.188091866244834e-05,
|
| 639 |
+
"loss": 0.0209,
|
| 640 |
"step": 900
|
| 641 |
},
|
| 642 |
{
|
| 643 |
+
"epoch": 0.764063811922754,
|
| 644 |
+
"grad_norm": 0.5055357813835144,
|
| 645 |
+
"learning_rate": 9.165368558923695e-05,
|
| 646 |
+
"loss": 0.0143,
|
| 647 |
"step": 910
|
| 648 |
},
|
| 649 |
{
|
| 650 |
+
"epoch": 0.7724601175482787,
|
| 651 |
+
"grad_norm": 0.5202294588088989,
|
| 652 |
+
"learning_rate": 9.142360554194618e-05,
|
| 653 |
+
"loss": 0.0173,
|
| 654 |
"step": 920
|
| 655 |
},
|
| 656 |
{
|
| 657 |
+
"epoch": 0.7808564231738035,
|
| 658 |
+
"grad_norm": 0.2911999821662903,
|
| 659 |
+
"learning_rate": 9.119069424624163e-05,
|
| 660 |
+
"loss": 0.0152,
|
| 661 |
"step": 930
|
| 662 |
},
|
| 663 |
{
|
| 664 |
+
"epoch": 0.7892527287993283,
|
| 665 |
+
"grad_norm": 0.28226837515830994,
|
| 666 |
+
"learning_rate": 9.0954967621301e-05,
|
| 667 |
+
"loss": 0.0125,
|
| 668 |
"step": 940
|
| 669 |
},
|
| 670 |
{
|
| 671 |
+
"epoch": 0.797649034424853,
|
| 672 |
+
"grad_norm": 0.24133360385894775,
|
| 673 |
+
"learning_rate": 9.071644177872594e-05,
|
| 674 |
+
"loss": 0.018,
|
| 675 |
"step": 950
|
| 676 |
},
|
| 677 |
{
|
| 678 |
+
"epoch": 0.8060453400503779,
|
| 679 |
+
"grad_norm": 0.28321564197540283,
|
| 680 |
+
"learning_rate": 9.047513302144095e-05,
|
| 681 |
+
"loss": 0.0126,
|
| 682 |
"step": 960
|
| 683 |
},
|
| 684 |
{
|
| 685 |
+
"epoch": 0.8144416456759026,
|
| 686 |
+
"grad_norm": 0.3964134752750397,
|
| 687 |
+
"learning_rate": 9.023105784257906e-05,
|
| 688 |
+
"loss": 0.0131,
|
| 689 |
"step": 970
|
| 690 |
},
|
| 691 |
{
|
| 692 |
+
"epoch": 0.8228379513014273,
|
| 693 |
+
"grad_norm": 0.4938385486602783,
|
| 694 |
+
"learning_rate": 8.998423292435454e-05,
|
| 695 |
+
"loss": 0.0166,
|
| 696 |
"step": 980
|
| 697 |
},
|
| 698 |
{
|
| 699 |
+
"epoch": 0.8312342569269522,
|
| 700 |
+
"grad_norm": 0.38572144508361816,
|
| 701 |
+
"learning_rate": 8.973467513692265e-05,
|
| 702 |
+
"loss": 0.0194,
|
| 703 |
"step": 990
|
| 704 |
},
|
| 705 |
{
|
| 706 |
+
"epoch": 0.8396305625524769,
|
| 707 |
+
"grad_norm": 0.5354552268981934,
|
| 708 |
+
"learning_rate": 8.94824015372267e-05,
|
| 709 |
+
"loss": 0.0159,
|
| 710 |
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.8480268681780017,
|
| 714 |
+
"grad_norm": 0.5144203305244446,
|
| 715 |
+
"learning_rate": 8.922742936783207e-05,
|
| 716 |
+
"loss": 0.0142,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.8564231738035264,
|
| 721 |
+
"grad_norm": 0.49846023321151733,
|
| 722 |
+
"learning_rate": 8.896977605574788e-05,
|
| 723 |
+
"loss": 0.015,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.8648194794290512,
|
| 728 |
+
"grad_norm": 0.48377203941345215,
|
| 729 |
+
"learning_rate": 8.870945921123576e-05,
|
| 730 |
+
"loss": 0.0151,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.873215785054576,
|
| 735 |
+
"grad_norm": 0.45066890120506287,
|
| 736 |
+
"learning_rate": 8.844649662660624e-05,
|
| 737 |
+
"loss": 0.0145,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.8816120906801007,
|
| 742 |
+
"grad_norm": 0.4206666946411133,
|
| 743 |
+
"learning_rate": 8.818090627500266e-05,
|
| 744 |
+
"loss": 0.0129,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.8900083963056256,
|
| 749 |
+
"grad_norm": 0.41736483573913574,
|
| 750 |
+
"learning_rate": 8.791270630917275e-05,
|
| 751 |
+
"loss": 0.0192,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.8984047019311503,
|
| 756 |
+
"grad_norm": 0.3356010615825653,
|
| 757 |
+
"learning_rate": 8.764191506022795e-05,
|
| 758 |
+
"loss": 0.0148,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.906801007556675,
|
| 763 |
+
"grad_norm": 0.3665725886821747,
|
| 764 |
+
"learning_rate": 8.736855103639037e-05,
|
| 765 |
+
"loss": 0.0155,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.9151973131821999,
|
| 770 |
+
"grad_norm": NaN,
|
| 771 |
+
"learning_rate": 8.709263292172794e-05,
|
| 772 |
+
"loss": 0.0082,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.9235936188077246,
|
| 777 |
+
"grad_norm": NaN,
|
| 778 |
+
"learning_rate": 8.681417957487729e-05,
|
| 779 |
+
"loss": 0.0,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.9319899244332494,
|
| 784 |
+
"grad_norm": NaN,
|
| 785 |
+
"learning_rate": 8.653321002775478e-05,
|
| 786 |
+
"loss": 0.0,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.9403862300587741,
|
| 791 |
+
"grad_norm": NaN,
|
| 792 |
+
"learning_rate": 8.624974348425574e-05,
|
| 793 |
+
"loss": 0.0,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.9487825356842989,
|
| 798 |
+
"grad_norm": NaN,
|
| 799 |
+
"learning_rate": 8.596379931894188e-05,
|
| 800 |
+
"loss": 0.0,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.9571788413098237,
|
| 805 |
+
"grad_norm": NaN,
|
| 806 |
+
"learning_rate": 8.567539707571703e-05,
|
| 807 |
+
"loss": 0.0,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.9655751469353484,
|
| 812 |
+
"grad_norm": NaN,
|
| 813 |
+
"learning_rate": 8.538455646649146e-05,
|
| 814 |
+
"loss": 0.0,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.9739714525608733,
|
| 819 |
+
"grad_norm": NaN,
|
| 820 |
+
"learning_rate": 8.509129736983446e-05,
|
| 821 |
+
"loss": 0.0,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.982367758186398,
|
| 826 |
+
"grad_norm": NaN,
|
| 827 |
+
"learning_rate": 8.479563982961571e-05,
|
| 828 |
+
"loss": 0.0,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.9907640638119227,
|
| 833 |
+
"grad_norm": NaN,
|
| 834 |
+
"learning_rate": 8.449760405363539e-05,
|
| 835 |
+
"loss": 0.0,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.9991603694374476,
|
| 840 |
+
"grad_norm": NaN,
|
| 841 |
+
"learning_rate": 8.419721041224287e-05,
|
| 842 |
+
"loss": 0.0,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.0075566750629723,
|
| 847 |
+
"grad_norm": NaN,
|
| 848 |
+
"learning_rate": 8.389447943694451e-05,
|
| 849 |
+
"loss": 0.0,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.015952980688497,
|
| 854 |
+
"grad_norm": NaN,
|
| 855 |
+
"learning_rate": 8.358943181900032e-05,
|
| 856 |
+
"loss": 0.0,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.0243492863140218,
|
| 861 |
+
"grad_norm": NaN,
|
| 862 |
+
"learning_rate": 8.328208840800981e-05,
|
| 863 |
+
"loss": 0.0,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.0327455919395465,
|
| 868 |
+
"grad_norm": NaN,
|
| 869 |
+
"learning_rate": 8.297247021048686e-05,
|
| 870 |
+
"loss": 0.0,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.0411418975650715,
|
| 875 |
+
"grad_norm": NaN,
|
| 876 |
+
"learning_rate": 8.266059838842396e-05,
|
| 877 |
+
"loss": 0.0,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.0495382031905962,
|
| 882 |
+
"grad_norm": NaN,
|
| 883 |
+
"learning_rate": 8.23464942578459e-05,
|
| 884 |
+
"loss": 0.0,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.057934508816121,
|
| 889 |
+
"grad_norm": NaN,
|
| 890 |
+
"learning_rate": 8.203017928735277e-05,
|
| 891 |
+
"loss": 0.0,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.0663308144416457,
|
| 896 |
+
"grad_norm": NaN,
|
| 897 |
+
"learning_rate": 8.17116750966526e-05,
|
| 898 |
+
"loss": 0.0,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.0747271200671704,
|
| 903 |
+
"grad_norm": NaN,
|
| 904 |
+
"learning_rate": 8.139100345508377e-05,
|
| 905 |
+
"loss": 0.0,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.0831234256926952,
|
| 910 |
+
"grad_norm": NaN,
|
| 911 |
+
"learning_rate": 8.106818628012697e-05,
|
| 912 |
+
"loss": 0.0,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.0915197313182199,
|
| 917 |
+
"grad_norm": NaN,
|
| 918 |
+
"learning_rate": 8.074324563590736e-05,
|
| 919 |
+
"loss": 0.0,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.0999160369437448,
|
| 924 |
+
"grad_norm": NaN,
|
| 925 |
+
"learning_rate": 8.041620373168628e-05,
|
| 926 |
+
"loss": 0.0,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.1083123425692696,
|
| 931 |
+
"grad_norm": NaN,
|
| 932 |
+
"learning_rate": 8.008708292034349e-05,
|
| 933 |
+
"loss": 0.0,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.1167086481947943,
|
| 938 |
+
"grad_norm": NaN,
|
| 939 |
+
"learning_rate": 7.975590569684925e-05,
|
| 940 |
+
"loss": 0.0,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.125104953820319,
|
| 945 |
+
"grad_norm": NaN,
|
| 946 |
+
"learning_rate": 7.942269469672687e-05,
|
| 947 |
+
"loss": 0.0,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.1335012594458438,
|
| 952 |
+
"grad_norm": NaN,
|
| 953 |
+
"learning_rate": 7.908747269450558e-05,
|
| 954 |
+
"loss": 0.0,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.1418975650713685,
|
| 959 |
+
"grad_norm": NaN,
|
| 960 |
+
"learning_rate": 7.875026260216393e-05,
|
| 961 |
+
"loss": 0.0,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.1502938706968933,
|
| 966 |
+
"grad_norm": NaN,
|
| 967 |
+
"learning_rate": 7.841108746756382e-05,
|
| 968 |
+
"loss": 0.0,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.1586901763224182,
|
| 973 |
+
"grad_norm": NaN,
|
| 974 |
+
"learning_rate": 7.806997047287516e-05,
|
| 975 |
+
"loss": 0.0,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.167086481947943,
|
| 980 |
+
"grad_norm": NaN,
|
| 981 |
+
"learning_rate": 7.772693493299138e-05,
|
| 982 |
+
"loss": 0.0,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.1754827875734677,
|
| 987 |
+
"grad_norm": NaN,
|
| 988 |
+
"learning_rate": 7.7382004293936e-05,
|
| 989 |
+
"loss": 0.0,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.1838790931989924,
|
| 994 |
+
"grad_norm": NaN,
|
| 995 |
+
"learning_rate": 7.703520213126e-05,
|
| 996 |
+
"loss": 0.0,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.1922753988245172,
|
| 1001 |
+
"grad_norm": NaN,
|
| 1002 |
+
"learning_rate": 7.66865521484305e-05,
|
| 1003 |
+
"loss": 0.0,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.200671704450042,
|
| 1008 |
+
"grad_norm": NaN,
|
| 1009 |
+
"learning_rate": 7.633607817521074e-05,
|
| 1010 |
+
"loss": 0.0,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.2090680100755669,
|
| 1015 |
+
"grad_norm": NaN,
|
| 1016 |
+
"learning_rate": 7.598380416603119e-05,
|
| 1017 |
+
"loss": 0.0,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.2174643157010916,
|
| 1022 |
+
"grad_norm": NaN,
|
| 1023 |
+
"learning_rate": 7.562975419835247e-05,
|
| 1024 |
+
"loss": 0.0,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.2258606213266163,
|
| 1029 |
+
"grad_norm": NaN,
|
| 1030 |
+
"learning_rate": 7.527395247101956e-05,
|
| 1031 |
+
"loss": 0.0,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.234256926952141,
|
| 1036 |
+
"grad_norm": NaN,
|
| 1037 |
+
"learning_rate": 7.491642330260789e-05,
|
| 1038 |
+
"loss": 0.0,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.2426532325776658,
|
| 1043 |
+
"grad_norm": NaN,
|
| 1044 |
+
"learning_rate": 7.45571911297612e-05,
|
| 1045 |
+
"loss": 0.0,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.2510495382031905,
|
| 1050 |
+
"grad_norm": NaN,
|
| 1051 |
+
"learning_rate": 7.419628050552131e-05,
|
| 1052 |
+
"loss": 0.0,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.2594458438287153,
|
| 1057 |
+
"grad_norm": NaN,
|
| 1058 |
+
"learning_rate": 7.383371609764999e-05,
|
| 1059 |
+
"loss": 0.0,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.26784214945424,
|
| 1064 |
+
"grad_norm": NaN,
|
| 1065 |
+
"learning_rate": 7.346952268694288e-05,
|
| 1066 |
+
"loss": 0.0,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.276238455079765,
|
| 1071 |
+
"grad_norm": NaN,
|
| 1072 |
+
"learning_rate": 7.310372516553585e-05,
|
| 1073 |
+
"loss": 0.0,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.2846347607052897,
|
| 1078 |
+
"grad_norm": NaN,
|
| 1079 |
+
"learning_rate": 7.273634853520356e-05,
|
| 1080 |
+
"loss": 0.0,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.2930310663308144,
|
| 1085 |
+
"grad_norm": NaN,
|
| 1086 |
+
"learning_rate": 7.236741790565072e-05,
|
| 1087 |
+
"loss": 0.0,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.3014273719563392,
|
| 1092 |
+
"grad_norm": NaN,
|
| 1093 |
+
"learning_rate": 7.199695849279576e-05,
|
| 1094 |
+
"loss": 0.0,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.309823677581864,
|
| 1099 |
+
"grad_norm": NaN,
|
| 1100 |
+
"learning_rate": 7.162499561704747e-05,
|
| 1101 |
+
"loss": 0.0,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.3182199832073889,
|
| 1106 |
+
"grad_norm": NaN,
|
| 1107 |
+
"learning_rate": 7.125155470157429e-05,
|
| 1108 |
+
"loss": 0.0,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.3266162888329136,
|
| 1113 |
+
"grad_norm": NaN,
|
| 1114 |
+
"learning_rate": 7.087666127056675e-05,
|
| 1115 |
+
"loss": 0.0,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.3350125944584383,
|
| 1120 |
+
"grad_norm": NaN,
|
| 1121 |
+
"learning_rate": 7.050034094749286e-05,
|
| 1122 |
+
"loss": 0.0,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.343408900083963,
|
| 1127 |
+
"grad_norm": NaN,
|
| 1128 |
+
"learning_rate": 7.012261945334683e-05,
|
| 1129 |
+
"loss": 0.0,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.3518052057094878,
|
| 1134 |
+
"grad_norm": NaN,
|
| 1135 |
+
"learning_rate": 6.974352260489103e-05,
|
| 1136 |
+
"loss": 0.0,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.3602015113350125,
|
| 1141 |
+
"grad_norm": NaN,
|
| 1142 |
+
"learning_rate": 6.936307631289148e-05,
|
| 1143 |
+
"loss": 0.0,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.3685978169605373,
|
| 1148 |
+
"grad_norm": NaN,
|
| 1149 |
+
"learning_rate": 6.898130658034685e-05,
|
| 1150 |
+
"loss": 0.0,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.376994122586062,
|
| 1155 |
+
"grad_norm": NaN,
|
| 1156 |
+
"learning_rate": 6.859823950071127e-05,
|
| 1157 |
+
"loss": 0.0,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.385390428211587,
|
| 1162 |
+
"grad_norm": NaN,
|
| 1163 |
+
"learning_rate": 6.821390125611078e-05,
|
| 1164 |
+
"loss": 0.0,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.3937867338371117,
|
| 1169 |
+
"grad_norm": NaN,
|
| 1170 |
+
"learning_rate": 6.782831811555385e-05,
|
| 1171 |
+
"loss": 0.0,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.4021830394626364,
|
| 1176 |
+
"grad_norm": NaN,
|
| 1177 |
+
"learning_rate": 6.744151643313597e-05,
|
| 1178 |
+
"loss": 0.0,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.4105793450881612,
|
| 1183 |
+
"grad_norm": NaN,
|
| 1184 |
+
"learning_rate": 6.705352264623828e-05,
|
| 1185 |
+
"loss": 0.0,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.418975650713686,
|
| 1190 |
+
"grad_norm": NaN,
|
| 1191 |
+
"learning_rate": 6.666436327372078e-05,
|
| 1192 |
+
"loss": 0.0,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.4273719563392109,
|
| 1197 |
+
"grad_norm": NaN,
|
| 1198 |
+
"learning_rate": 6.62740649141096e-05,
|
| 1199 |
+
"loss": 0.0,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.4357682619647356,
|
| 1204 |
+
"grad_norm": NaN,
|
| 1205 |
+
"learning_rate": 6.588265424377919e-05,
|
| 1206 |
+
"loss": 0.0,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.4441645675902604,
|
| 1211 |
+
"grad_norm": NaN,
|
| 1212 |
+
"learning_rate": 6.549015801512895e-05,
|
| 1213 |
+
"loss": 0.0,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.452560873215785,
|
| 1218 |
+
"grad_norm": NaN,
|
| 1219 |
+
"learning_rate": 6.509660305475468e-05,
|
| 1220 |
+
"loss": 0.0,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.4609571788413098,
|
| 1225 |
+
"grad_norm": NaN,
|
| 1226 |
+
"learning_rate": 6.47020162616152e-05,
|
| 1227 |
+
"loss": 0.0,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.4693534844668346,
|
| 1232 |
+
"grad_norm": NaN,
|
| 1233 |
+
"learning_rate": 6.430642460519365e-05,
|
| 1234 |
+
"loss": 0.0,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.4777497900923593,
|
| 1239 |
+
"grad_norm": NaN,
|
| 1240 |
+
"learning_rate": 6.390985512365426e-05,
|
| 1241 |
+
"loss": 0.0,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.486146095717884,
|
| 1246 |
+
"grad_norm": NaN,
|
| 1247 |
+
"learning_rate": 6.351233492199431e-05,
|
| 1248 |
+
"loss": 0.0,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.4945424013434088,
|
| 1253 |
+
"grad_norm": NaN,
|
| 1254 |
+
"learning_rate": 6.311389117019155e-05,
|
| 1255 |
+
"loss": 0.0,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.5029387069689337,
|
| 1260 |
+
"grad_norm": NaN,
|
| 1261 |
+
"learning_rate": 6.271455110134713e-05,
|
| 1262 |
+
"loss": 0.0,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.5113350125944585,
|
| 1267 |
+
"grad_norm": NaN,
|
| 1268 |
+
"learning_rate": 6.231434200982428e-05,
|
| 1269 |
+
"loss": 0.0,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.5197313182199832,
|
| 1274 |
+
"grad_norm": NaN,
|
| 1275 |
+
"learning_rate": 6.191329124938285e-05,
|
| 1276 |
+
"loss": 0.0,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.528127623845508,
|
| 1281 |
+
"grad_norm": NaN,
|
| 1282 |
+
"learning_rate": 6.15114262313095e-05,
|
| 1283 |
+
"loss": 0.0,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.536523929471033,
|
| 1288 |
+
"grad_norm": NaN,
|
| 1289 |
+
"learning_rate": 6.110877442254444e-05,
|
| 1290 |
+
"loss": 0.0,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.5449202350965576,
|
| 1295 |
+
"grad_norm": NaN,
|
| 1296 |
+
"learning_rate": 6.0705363343803946e-05,
|
| 1297 |
+
"loss": 0.0,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.5533165407220824,
|
| 1302 |
+
"grad_norm": NaN,
|
| 1303 |
+
"learning_rate": 6.030122056769934e-05,
|
| 1304 |
+
"loss": 0.0,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.561712846347607,
|
| 1309 |
+
"grad_norm": NaN,
|
| 1310 |
+
"learning_rate": 5.989637371685257e-05,
|
| 1311 |
+
"loss": 0.0,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.5701091519731318,
|
| 1316 |
+
"grad_norm": NaN,
|
| 1317 |
+
"learning_rate": 5.949085046200808e-05,
|
| 1318 |
+
"loss": 0.0,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.5785054575986566,
|
| 1323 |
+
"grad_norm": NaN,
|
| 1324 |
+
"learning_rate": 5.908467852014169e-05,
|
| 1325 |
+
"loss": 0.0,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.5869017632241813,
|
| 1330 |
+
"grad_norm": NaN,
|
| 1331 |
+
"learning_rate": 5.867788565256607e-05,
|
| 1332 |
+
"loss": 0.0,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.595298068849706,
|
| 1337 |
+
"grad_norm": NaN,
|
| 1338 |
+
"learning_rate": 5.827049966303335e-05,
|
| 1339 |
+
"loss": 0.0,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.6036943744752308,
|
| 1344 |
+
"grad_norm": NaN,
|
| 1345 |
+
"learning_rate": 5.786254839583478e-05,
|
| 1346 |
+
"loss": 0.0,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.6120906801007555,
|
| 1351 |
+
"grad_norm": NaN,
|
| 1352 |
+
"learning_rate": 5.745405973389757e-05,
|
| 1353 |
+
"loss": 0.0,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.6204869857262805,
|
| 1358 |
+
"grad_norm": NaN,
|
| 1359 |
+
"learning_rate": 5.7045061596879134e-05,
|
| 1360 |
+
"loss": 0.0,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.6288832913518052,
|
| 1365 |
+
"grad_norm": NaN,
|
| 1366 |
+
"learning_rate": 5.6635581939258855e-05,
|
| 1367 |
+
"loss": 0.0,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.63727959697733,
|
| 1372 |
+
"grad_norm": NaN,
|
| 1373 |
+
"learning_rate": 5.622564874842742e-05,
|
| 1374 |
+
"loss": 0.0,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.645675902602855,
|
| 1379 |
+
"grad_norm": NaN,
|
| 1380 |
+
"learning_rate": 5.5815290042773836e-05,
|
| 1381 |
+
"loss": 0.0,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.6540722082283796,
|
| 1386 |
+
"grad_norm": NaN,
|
| 1387 |
+
"learning_rate": 5.540453386977058e-05,
|
| 1388 |
+
"loss": 0.0,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.6624685138539044,
|
| 1393 |
+
"grad_norm": NaN,
|
| 1394 |
+
"learning_rate": 5.4993408304056425e-05,
|
| 1395 |
+
"loss": 0.0,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.670864819479429,
|
| 1400 |
+
"grad_norm": NaN,
|
| 1401 |
+
"learning_rate": 5.458194144551768e-05,
|
| 1402 |
+
"loss": 0.0,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.6792611251049538,
|
| 1407 |
+
"grad_norm": NaN,
|
| 1408 |
+
"learning_rate": 5.417016141736756e-05,
|
| 1409 |
+
"loss": 0.0,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.6876574307304786,
|
| 1414 |
+
"grad_norm": NaN,
|
| 1415 |
+
"learning_rate": 5.375809636422399e-05,
|
| 1416 |
+
"loss": 0.0,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.6960537363560033,
|
| 1421 |
+
"grad_norm": NaN,
|
| 1422 |
+
"learning_rate": 5.334577445018599e-05,
|
| 1423 |
+
"loss": 0.0,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.704450041981528,
|
| 1428 |
+
"grad_norm": NaN,
|
| 1429 |
+
"learning_rate": 5.293322385690867e-05,
|
| 1430 |
+
"loss": 0.0,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.7128463476070528,
|
| 1435 |
+
"grad_norm": NaN,
|
| 1436 |
+
"learning_rate": 5.252047278167709e-05,
|
| 1437 |
+
"loss": 0.0,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.7212426532325775,
|
| 1442 |
+
"grad_norm": NaN,
|
| 1443 |
+
"learning_rate": 5.210754943547893e-05,
|
| 1444 |
+
"loss": 0.0,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.7296389588581025,
|
| 1449 |
+
"grad_norm": NaN,
|
| 1450 |
+
"learning_rate": 5.169448204107643e-05,
|
| 1451 |
+
"loss": 0.0,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.7380352644836272,
|
| 1456 |
+
"grad_norm": NaN,
|
| 1457 |
+
"learning_rate": 5.128129883107729e-05,
|
| 1458 |
+
"loss": 0.0,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.746431570109152,
|
| 1463 |
+
"grad_norm": NaN,
|
| 1464 |
+
"learning_rate": 5.086802804600505e-05,
|
| 1465 |
+
"loss": 0.0,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.7548278757346767,
|
| 1470 |
+
"grad_norm": NaN,
|
| 1471 |
+
"learning_rate": 5.045469793236892e-05,
|
| 1472 |
+
"loss": 0.0,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.7632241813602016,
|
| 1477 |
+
"grad_norm": NaN,
|
| 1478 |
+
"learning_rate": 5.00413367407331e-05,
|
| 1479 |
+
"loss": 0.0,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.7716204869857264,
|
| 1484 |
+
"grad_norm": NaN,
|
| 1485 |
+
"learning_rate": 4.9627972723785964e-05,
|
| 1486 |
+
"loss": 0.0,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.7800167926112511,
|
| 1491 |
+
"grad_norm": NaN,
|
| 1492 |
+
"learning_rate": 4.921463413440898e-05,
|
| 1493 |
+
"loss": 0.0,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.7884130982367759,
|
| 1498 |
+
"grad_norm": NaN,
|
| 1499 |
+
"learning_rate": 4.8801349223745654e-05,
|
| 1500 |
+
"loss": 0.0,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.7968094038623006,
|
| 1505 |
+
"grad_norm": NaN,
|
| 1506 |
+
"learning_rate": 4.838814623927067e-05,
|
| 1507 |
+
"loss": 0.0,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.8052057094878253,
|
| 1512 |
+
"grad_norm": NaN,
|
| 1513 |
+
"learning_rate": 4.797505342285912e-05,
|
| 1514 |
+
"loss": 0.0,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.81360201511335,
|
| 1519 |
+
"grad_norm": NaN,
|
| 1520 |
+
"learning_rate": 4.756209900885628e-05,
|
| 1521 |
+
"loss": 0.0,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.8219983207388748,
|
| 1526 |
+
"grad_norm": NaN,
|
| 1527 |
+
"learning_rate": 4.714931122214781e-05,
|
| 1528 |
+
"loss": 0.0,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.8303946263643995,
|
| 1533 |
+
"grad_norm": NaN,
|
| 1534 |
+
"learning_rate": 4.673671827623058e-05,
|
| 1535 |
+
"loss": 0.0,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.8387909319899243,
|
| 1540 |
+
"grad_norm": NaN,
|
| 1541 |
+
"learning_rate": 4.632434837128443e-05,
|
| 1542 |
+
"loss": 0.0,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.8471872376154492,
|
| 1547 |
+
"grad_norm": NaN,
|
| 1548 |
+
"learning_rate": 4.591222969224453e-05,
|
| 1549 |
+
"loss": 0.0,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.855583543240974,
|
| 1554 |
+
"grad_norm": NaN,
|
| 1555 |
+
"learning_rate": 4.550039040687518e-05,
|
| 1556 |
+
"loss": 0.0,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.8639798488664987,
|
| 1561 |
+
"grad_norm": NaN,
|
| 1562 |
+
"learning_rate": 4.508885866384446e-05,
|
| 1563 |
+
"loss": 0.0,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.8723761544920237,
|
| 1568 |
+
"grad_norm": NaN,
|
| 1569 |
+
"learning_rate": 4.4677662590800355e-05,
|
| 1570 |
+
"loss": 0.0,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.8807724601175484,
|
| 1575 |
+
"grad_norm": NaN,
|
| 1576 |
+
"learning_rate": 4.426683029244825e-05,
|
| 1577 |
+
"loss": 0.0,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.8891687657430731,
|
| 1582 |
+
"grad_norm": NaN,
|
| 1583 |
+
"learning_rate": 4.385638984863e-05,
|
| 1584 |
+
"loss": 0.0,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.8975650713685979,
|
| 1589 |
+
"grad_norm": NaN,
|
| 1590 |
+
"learning_rate": 4.3446369312404745e-05,
|
| 1591 |
+
"loss": 0.0,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.9059613769941226,
|
| 1596 |
+
"grad_norm": NaN,
|
| 1597 |
+
"learning_rate": 4.3036796708131474e-05,
|
| 1598 |
+
"loss": 0.0,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.9143576826196473,
|
| 1603 |
+
"grad_norm": NaN,
|
| 1604 |
+
"learning_rate": 4.262770002955363e-05,
|
| 1605 |
+
"loss": 0.0,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.922753988245172,
|
| 1610 |
+
"grad_norm": NaN,
|
| 1611 |
+
"learning_rate": 4.221910723788578e-05,
|
| 1612 |
+
"loss": 0.0,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.9311502938706968,
|
| 1617 |
+
"grad_norm": NaN,
|
| 1618 |
+
"learning_rate": 4.1811046259902474e-05,
|
| 1619 |
+
"loss": 0.0,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.9395465994962215,
|
| 1624 |
+
"grad_norm": NaN,
|
| 1625 |
+
"learning_rate": 4.140354498602952e-05,
|
| 1626 |
+
"loss": 0.0,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.9479429051217463,
|
| 1631 |
+
"grad_norm": NaN,
|
| 1632 |
+
"learning_rate": 4.099663126843769e-05,
|
| 1633 |
+
"loss": 0.0,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.9563392107472712,
|
| 1638 |
+
"grad_norm": NaN,
|
| 1639 |
+
"learning_rate": 4.059033291913902e-05,
|
| 1640 |
+
"loss": 0.0,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.964735516372796,
|
| 1645 |
+
"grad_norm": NaN,
|
| 1646 |
+
"learning_rate": 4.0184677708086014e-05,
|
| 1647 |
+
"loss": 0.0,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.9731318219983207,
|
| 1652 |
+
"grad_norm": NaN,
|
| 1653 |
+
"learning_rate": 3.977969336127348e-05,
|
| 1654 |
+
"loss": 0.0,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.9815281276238457,
|
| 1659 |
+
"grad_norm": NaN,
|
| 1660 |
+
"learning_rate": 3.937540755884357e-05,
|
| 1661 |
+
"loss": 0.0,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.9899244332493704,
|
| 1666 |
+
"grad_norm": NaN,
|
| 1667 |
+
"learning_rate": 3.897184793319384e-05,
|
| 1668 |
+
"loss": 0.0,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.9983207388748951,
|
| 1673 |
+
"grad_norm": NaN,
|
| 1674 |
+
"learning_rate": 3.856904206708863e-05,
|
| 1675 |
+
"loss": 0.0,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.00671704450042,
|
| 1680 |
+
"grad_norm": NaN,
|
| 1681 |
+
"learning_rate": 3.8167017491773847e-05,
|
| 1682 |
+
"loss": 0.0,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 2.0151133501259446,
|
| 1687 |
+
"grad_norm": NaN,
|
| 1688 |
+
"learning_rate": 3.776580168509516e-05,
|
| 1689 |
+
"loss": 0.0,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 2.0235096557514693,
|
| 1694 |
+
"grad_norm": NaN,
|
| 1695 |
+
"learning_rate": 3.736542206962e-05,
|
| 1696 |
+
"loss": 0.0,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 2.031905961376994,
|
| 1701 |
+
"grad_norm": NaN,
|
| 1702 |
+
"learning_rate": 3.696590601076326e-05,
|
| 1703 |
+
"loss": 0.0,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 2.040302267002519,
|
| 1708 |
+
"grad_norm": NaN,
|
| 1709 |
+
"learning_rate": 3.656728081491686e-05,
|
| 1710 |
+
"loss": 0.0,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.0486985726280436,
|
| 1715 |
+
"grad_norm": NaN,
|
| 1716 |
+
"learning_rate": 3.6169573727583405e-05,
|
| 1717 |
+
"loss": 0.0,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.0570948782535683,
|
| 1722 |
+
"grad_norm": NaN,
|
| 1723 |
+
"learning_rate": 3.5772811931514036e-05,
|
| 1724 |
+
"loss": 0.0,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 2.065491183879093,
|
| 1729 |
+
"grad_norm": NaN,
|
| 1730 |
+
"learning_rate": 3.5377022544850505e-05,
|
| 1731 |
+
"loss": 0.0,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 2.0738874895046178,
|
| 1736 |
+
"grad_norm": NaN,
|
| 1737 |
+
"learning_rate": 3.498223261927158e-05,
|
| 1738 |
+
"loss": 0.0,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 2.082283795130143,
|
| 1743 |
+
"grad_norm": NaN,
|
| 1744 |
+
"learning_rate": 3.4588469138144295e-05,
|
| 1745 |
+
"loss": 0.0,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 2.0906801007556677,
|
| 1750 |
+
"grad_norm": NaN,
|
| 1751 |
+
"learning_rate": 3.419575901467952e-05,
|
| 1752 |
+
"loss": 0.0,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 2.0990764063811924,
|
| 1757 |
+
"grad_norm": NaN,
|
| 1758 |
+
"learning_rate": 3.380412909009254e-05,
|
| 1759 |
+
"loss": 0.0,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 2.107472712006717,
|
| 1764 |
+
"grad_norm": NaN,
|
| 1765 |
+
"learning_rate": 3.3413606131768475e-05,
|
| 1766 |
+
"loss": 0.0,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 2.115869017632242,
|
| 1771 |
+
"grad_norm": NaN,
|
| 1772 |
+
"learning_rate": 3.302421683143279e-05,
|
| 1773 |
+
"loss": 0.0,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 2.1242653232577666,
|
| 1778 |
+
"grad_norm": NaN,
|
| 1779 |
+
"learning_rate": 3.2635987803326896e-05,
|
| 1780 |
+
"loss": 0.0,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.1326616288832914,
|
| 1785 |
+
"grad_norm": NaN,
|
| 1786 |
+
"learning_rate": 3.224894558238918e-05,
|
| 1787 |
+
"loss": 0.0,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 2.141057934508816,
|
| 1792 |
+
"grad_norm": NaN,
|
| 1793 |
+
"learning_rate": 3.18631166224413e-05,
|
| 1794 |
+
"loss": 0.0,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 2.149454240134341,
|
| 1799 |
+
"grad_norm": NaN,
|
| 1800 |
+
"learning_rate": 3.147852729438017e-05,
|
| 1801 |
+
"loss": 0.0,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.1578505457598656,
|
| 1806 |
+
"grad_norm": NaN,
|
| 1807 |
+
"learning_rate": 3.109520388437548e-05,
|
| 1808 |
+
"loss": 0.0,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.1662468513853903,
|
| 1813 |
+
"grad_norm": NaN,
|
| 1814 |
+
"learning_rate": 3.0713172592073116e-05,
|
| 1815 |
+
"loss": 0.0,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.174643157010915,
|
| 1820 |
+
"grad_norm": NaN,
|
| 1821 |
+
"learning_rate": 3.0332459528804457e-05,
|
| 1822 |
+
"loss": 0.0,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.1830394626364398,
|
| 1827 |
+
"grad_norm": NaN,
|
| 1828 |
+
"learning_rate": 2.9953090715801634e-05,
|
| 1829 |
+
"loss": 0.0,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 2.1914357682619645,
|
| 1834 |
+
"grad_norm": NaN,
|
| 1835 |
+
"learning_rate": 2.9575092082419086e-05,
|
| 1836 |
+
"loss": 0.0,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.1998320738874897,
|
| 1841 |
+
"grad_norm": NaN,
|
| 1842 |
+
"learning_rate": 2.9198489464361288e-05,
|
| 1843 |
+
"loss": 0.0,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.2082283795130144,
|
| 1848 |
+
"grad_norm": NaN,
|
| 1849 |
+
"learning_rate": 2.8823308601916948e-05,
|
| 1850 |
+
"loss": 0.0,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.216624685138539,
|
| 1855 |
+
"grad_norm": NaN,
|
| 1856 |
+
"learning_rate": 2.8449575138199613e-05,
|
| 1857 |
+
"loss": 0.0,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.225020990764064,
|
| 1862 |
+
"grad_norm": NaN,
|
| 1863 |
+
"learning_rate": 2.807731461739509e-05,
|
| 1864 |
+
"loss": 0.0,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.2334172963895886,
|
| 1869 |
+
"grad_norm": NaN,
|
| 1870 |
+
"learning_rate": 2.7706552483015485e-05,
|
| 1871 |
+
"loss": 0.0,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 2.2418136020151134,
|
| 1876 |
+
"grad_norm": NaN,
|
| 1877 |
+
"learning_rate": 2.733731407616018e-05,
|
| 1878 |
+
"loss": 0.0,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.250209907640638,
|
| 1883 |
+
"grad_norm": NaN,
|
| 1884 |
+
"learning_rate": 2.6969624633783806e-05,
|
| 1885 |
+
"loss": 0.0,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.258606213266163,
|
| 1890 |
+
"grad_norm": NaN,
|
| 1891 |
+
"learning_rate": 2.660350928697134e-05,
|
| 1892 |
+
"loss": 0.0,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 2.2670025188916876,
|
| 1897 |
+
"grad_norm": NaN,
|
| 1898 |
+
"learning_rate": 2.6238993059220395e-05,
|
| 1899 |
+
"loss": 0.0,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.2753988245172123,
|
| 1904 |
+
"grad_norm": NaN,
|
| 1905 |
+
"learning_rate": 2.5876100864730933e-05,
|
| 1906 |
+
"loss": 0.0,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.283795130142737,
|
| 1911 |
+
"grad_norm": NaN,
|
| 1912 |
+
"learning_rate": 2.5514857506702405e-05,
|
| 1913 |
+
"loss": 0.0,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.292191435768262,
|
| 1918 |
+
"grad_norm": NaN,
|
| 1919 |
+
"learning_rate": 2.5155287675638474e-05,
|
| 1920 |
+
"loss": 0.0,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.3005877413937865,
|
| 1925 |
+
"grad_norm": NaN,
|
| 1926 |
+
"learning_rate": 2.4797415947659457e-05,
|
| 1927 |
+
"loss": 0.0,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.3089840470193117,
|
| 1932 |
+
"grad_norm": NaN,
|
| 1933 |
+
"learning_rate": 2.4441266782822588e-05,
|
| 1934 |
+
"loss": 0.0,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 2.3173803526448364,
|
| 1939 |
+
"grad_norm": NaN,
|
| 1940 |
+
"learning_rate": 2.4086864523450183e-05,
|
| 1941 |
+
"loss": 0.0,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 2.325776658270361,
|
| 1946 |
+
"grad_norm": NaN,
|
| 1947 |
+
"learning_rate": 2.3734233392465903e-05,
|
| 1948 |
+
"loss": 0.0,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.334172963895886,
|
| 1953 |
+
"grad_norm": NaN,
|
| 1954 |
+
"learning_rate": 2.3383397491739145e-05,
|
| 1955 |
+
"loss": 0.0,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.3425692695214106,
|
| 1960 |
+
"grad_norm": NaN,
|
| 1961 |
+
"learning_rate": 2.3034380800437678e-05,
|
| 1962 |
+
"loss": 0.0,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 2.3509655751469354,
|
| 1967 |
+
"grad_norm": NaN,
|
| 1968 |
+
"learning_rate": 2.2687207173388743e-05,
|
| 1969 |
+
"loss": 0.0,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 2.35936188077246,
|
| 1974 |
+
"grad_norm": NaN,
|
| 1975 |
+
"learning_rate": 2.234190033944858e-05,
|
| 1976 |
+
"loss": 0.0,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 2.367758186397985,
|
| 1981 |
+
"grad_norm": NaN,
|
| 1982 |
+
"learning_rate": 2.1998483899880596e-05,
|
| 1983 |
+
"loss": 0.0,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.3761544920235096,
|
| 1988 |
+
"grad_norm": NaN,
|
| 1989 |
+
"learning_rate": 2.1656981326742266e-05,
|
| 1990 |
+
"loss": 0.0,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.3845507976490343,
|
| 1995 |
+
"grad_norm": NaN,
|
| 1996 |
+
"learning_rate": 2.1317415961280824e-05,
|
| 1997 |
+
"loss": 0.0,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.392947103274559,
|
| 2002 |
+
"grad_norm": NaN,
|
| 2003 |
+
"learning_rate": 2.097981101233794e-05,
|
| 2004 |
+
"loss": 0.0,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.401343408900084,
|
| 2009 |
+
"grad_norm": NaN,
|
| 2010 |
+
"learning_rate": 2.0644189554763417e-05,
|
| 2011 |
+
"loss": 0.0,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.4097397145256085,
|
| 2016 |
+
"grad_norm": NaN,
|
| 2017 |
+
"learning_rate": 2.0310574527838072e-05,
|
| 2018 |
+
"loss": 0.0,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.4181360201511337,
|
| 2023 |
+
"grad_norm": NaN,
|
| 2024 |
+
"learning_rate": 1.9978988733705807e-05,
|
| 2025 |
+
"loss": 0.0,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.4265323257766584,
|
| 2030 |
+
"grad_norm": NaN,
|
| 2031 |
+
"learning_rate": 1.9649454835815202e-05,
|
| 2032 |
+
"loss": 0.0,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.434928631402183,
|
| 2037 |
+
"grad_norm": NaN,
|
| 2038 |
+
"learning_rate": 1.932199535737045e-05,
|
| 2039 |
+
"loss": 0.0,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.443324937027708,
|
| 2044 |
+
"grad_norm": NaN,
|
| 2045 |
+
"learning_rate": 1.8996632679791914e-05,
|
| 2046 |
+
"loss": 0.0,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.4517212426532327,
|
| 2051 |
+
"grad_norm": NaN,
|
| 2052 |
+
"learning_rate": 1.8673389041186418e-05,
|
| 2053 |
+
"loss": 0.0,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.4601175482787574,
|
| 2058 |
+
"grad_norm": NaN,
|
| 2059 |
+
"learning_rate": 1.8352286534827274e-05,
|
| 2060 |
+
"loss": 0.0,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 2.468513853904282,
|
| 2065 |
+
"grad_norm": NaN,
|
| 2066 |
+
"learning_rate": 1.803334710764426e-05,
|
| 2067 |
+
"loss": 0.0,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 2.476910159529807,
|
| 2072 |
+
"grad_norm": NaN,
|
| 2073 |
+
"learning_rate": 1.7716592558723556e-05,
|
| 2074 |
+
"loss": 0.0,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.4853064651553316,
|
| 2079 |
+
"grad_norm": NaN,
|
| 2080 |
+
"learning_rate": 1.7402044537817824e-05,
|
| 2081 |
+
"loss": 0.0,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 2.4937027707808563,
|
| 2086 |
+
"grad_norm": NaN,
|
| 2087 |
+
"learning_rate": 1.7089724543866465e-05,
|
| 2088 |
+
"loss": 0.0,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 2.502099076406381,
|
| 2093 |
+
"grad_norm": NaN,
|
| 2094 |
+
"learning_rate": 1.6779653923526188e-05,
|
| 2095 |
+
"loss": 0.0,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.510495382031906,
|
| 2100 |
+
"grad_norm": NaN,
|
| 2101 |
+
"learning_rate": 1.6471853869712023e-05,
|
| 2102 |
+
"loss": 0.0,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 2.5188916876574305,
|
| 2107 |
+
"grad_norm": NaN,
|
| 2108 |
+
"learning_rate": 1.6166345420148787e-05,
|
| 2109 |
+
"loss": 0.0,
|
| 2110 |
+
"step": 3000
|
| 2111 |
}
|
| 2112 |
],
|
| 2113 |
"logging_steps": 10,
|
| 2114 |
+
"max_steps": 4000,
|
| 2115 |
"num_input_tokens_seen": 0,
|
| 2116 |
+
"num_train_epochs": 4,
|
| 2117 |
"save_steps": 1000,
|
| 2118 |
"stateful_callbacks": {
|
| 2119 |
"TrainerControl": {
|