Upload folder using huggingface_hub
Browse files- config.json +64 -0
- experiment_cfg/metadata.json +195 -0
- model-00001-of-00002.safetensors +3 -0
- model-00002-of-00002.safetensors +3 -0
- model.safetensors.index.json +0 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1434 -0
config.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"action_dim": 32,
|
| 3 |
+
"action_head_cfg": {
|
| 4 |
+
"action_dim": 32,
|
| 5 |
+
"action_horizon": 16,
|
| 6 |
+
"add_pos_embed": true,
|
| 7 |
+
"backbone_embedding_dim": 2048,
|
| 8 |
+
"diffusion_model_cfg": {
|
| 9 |
+
"attention_head_dim": 48,
|
| 10 |
+
"cross_attention_dim": 2048,
|
| 11 |
+
"dropout": 0.2,
|
| 12 |
+
"final_dropout": true,
|
| 13 |
+
"interleave_self_attention": true,
|
| 14 |
+
"norm_type": "ada_norm",
|
| 15 |
+
"num_attention_heads": 32,
|
| 16 |
+
"num_layers": 16,
|
| 17 |
+
"output_dim": 1024,
|
| 18 |
+
"positional_embeddings": null
|
| 19 |
+
},
|
| 20 |
+
"hidden_size": 1024,
|
| 21 |
+
"input_embedding_dim": 1536,
|
| 22 |
+
"max_action_dim": 32,
|
| 23 |
+
"max_state_dim": 64,
|
| 24 |
+
"model_dtype": "float32",
|
| 25 |
+
"noise_beta_alpha": 1.5,
|
| 26 |
+
"noise_beta_beta": 1.0,
|
| 27 |
+
"noise_s": 0.999,
|
| 28 |
+
"num_inference_timesteps": 4,
|
| 29 |
+
"num_target_vision_tokens": 32,
|
| 30 |
+
"num_timestep_buckets": 1000,
|
| 31 |
+
"tune_diffusion_model": true,
|
| 32 |
+
"tune_projector": true,
|
| 33 |
+
"use_vlln": true,
|
| 34 |
+
"vl_self_attention_cfg": {
|
| 35 |
+
"attention_head_dim": 64,
|
| 36 |
+
"dropout": 0.2,
|
| 37 |
+
"final_dropout": true,
|
| 38 |
+
"num_attention_heads": 32,
|
| 39 |
+
"num_layers": 4,
|
| 40 |
+
"positional_embeddings": null
|
| 41 |
+
}
|
| 42 |
+
},
|
| 43 |
+
"action_horizon": 16,
|
| 44 |
+
"architectures": [
|
| 45 |
+
"GR00T_N1_5"
|
| 46 |
+
],
|
| 47 |
+
"attn_implementation": null,
|
| 48 |
+
"backbone_cfg": {
|
| 49 |
+
"eagle_path": "NVEagle/eagle_er-qwen3_1_7B-Siglip2_400M_stage1_5_128gpu_er_v7_1mlp_nops",
|
| 50 |
+
"load_bf16": false,
|
| 51 |
+
"project_to_dim": null,
|
| 52 |
+
"reproject_vision": false,
|
| 53 |
+
"select_layer": 12,
|
| 54 |
+
"tune_llm": false,
|
| 55 |
+
"tune_visual": true,
|
| 56 |
+
"use_flash_attention": true
|
| 57 |
+
},
|
| 58 |
+
"compute_dtype": "bfloat16",
|
| 59 |
+
"hidden_size": 2048,
|
| 60 |
+
"model_dtype": "float32",
|
| 61 |
+
"model_type": "gr00t_n1_5",
|
| 62 |
+
"torch_dtype": "bfloat16",
|
| 63 |
+
"transformers_version": "4.51.3"
|
| 64 |
+
}
|
experiment_cfg/metadata.json
ADDED
|
@@ -0,0 +1,195 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"new_embodiment": {
|
| 3 |
+
"statistics": {
|
| 4 |
+
"state": {
|
| 5 |
+
"single_arm": {
|
| 6 |
+
"max": [
|
| 7 |
+
39.650146484375,
|
| 8 |
+
37.19563293457031,
|
| 9 |
+
99.0076675415039,
|
| 10 |
+
99.15540313720703,
|
| 11 |
+
65.35760498046875
|
| 12 |
+
],
|
| 13 |
+
"min": [
|
| 14 |
+
-43.804664611816406,
|
| 15 |
+
-99.41226196289062,
|
| 16 |
+
-44.880470275878906,
|
| 17 |
+
-1.7736486196517944,
|
| 18 |
+
-44.6033821105957
|
| 19 |
+
],
|
| 20 |
+
"mean": [
|
| 21 |
+
1.255613923072815,
|
| 22 |
+
-25.14191246032715,
|
| 23 |
+
38.33786392211914,
|
| 24 |
+
57.2556037902832,
|
| 25 |
+
2.070636749267578
|
| 26 |
+
],
|
| 27 |
+
"std": [
|
| 28 |
+
12.124439239501953,
|
| 29 |
+
41.19919204711914,
|
| 30 |
+
36.9512825012207,
|
| 31 |
+
13.600706100463867,
|
| 32 |
+
16.455419540405273
|
| 33 |
+
],
|
| 34 |
+
"q01": [
|
| 35 |
+
-30.399416198730467,
|
| 36 |
+
-99.16036987304688,
|
| 37 |
+
-31.078033447265625,
|
| 38 |
+
19.08783721923828,
|
| 39 |
+
-35.55266571044922
|
| 40 |
+
],
|
| 41 |
+
"q99": [
|
| 42 |
+
32.43440246582031,
|
| 43 |
+
26.448362350463867,
|
| 44 |
+
98.8272476196289,
|
| 45 |
+
92.40540771484359,
|
| 46 |
+
48.14460327148428
|
| 47 |
+
]
|
| 48 |
+
},
|
| 49 |
+
"gripper": {
|
| 50 |
+
"max": [
|
| 51 |
+
29.503915786743164
|
| 52 |
+
],
|
| 53 |
+
"min": [
|
| 54 |
+
1.5013054609298706
|
| 55 |
+
],
|
| 56 |
+
"mean": [
|
| 57 |
+
9.473246574401855
|
| 58 |
+
],
|
| 59 |
+
"std": [
|
| 60 |
+
6.394623756408691
|
| 61 |
+
],
|
| 62 |
+
"q01": [
|
| 63 |
+
1.6318538188934326
|
| 64 |
+
],
|
| 65 |
+
"q99": [
|
| 66 |
+
25.326370239257812
|
| 67 |
+
]
|
| 68 |
+
}
|
| 69 |
+
},
|
| 70 |
+
"action": {
|
| 71 |
+
"single_arm": {
|
| 72 |
+
"max": [
|
| 73 |
+
40.4473762512207,
|
| 74 |
+
36.02638244628906,
|
| 75 |
+
99.64189910888672,
|
| 76 |
+
100.0,
|
| 77 |
+
66.16702270507812
|
| 78 |
+
],
|
| 79 |
+
"min": [
|
| 80 |
+
-44.6277961730957,
|
| 81 |
+
-100.0,
|
| 82 |
+
-50.31333923339844,
|
| 83 |
+
-3.347639560699463,
|
| 84 |
+
-44.86081314086914
|
| 85 |
+
],
|
| 86 |
+
"mean": [
|
| 87 |
+
1.3726696968078613,
|
| 88 |
+
-27.217138290405273,
|
| 89 |
+
35.27532196044922,
|
| 90 |
+
56.85714340209961,
|
| 91 |
+
2.086531400680542
|
| 92 |
+
],
|
| 93 |
+
"std": [
|
| 94 |
+
12.177175521850586,
|
| 95 |
+
40.36943054199219,
|
| 96 |
+
37.93010711669922,
|
| 97 |
+
14.12856674194336,
|
| 98 |
+
16.518014907836914
|
| 99 |
+
],
|
| 100 |
+
"q01": [
|
| 101 |
+
-30.25302505493164,
|
| 102 |
+
-100.0,
|
| 103 |
+
-35.720680236816406,
|
| 104 |
+
17.167381286621094,
|
| 105 |
+
-35.385440826416016
|
| 106 |
+
],
|
| 107 |
+
"q99": [
|
| 108 |
+
32.60579330444322,
|
| 109 |
+
24.319868087768555,
|
| 110 |
+
99.28379821777344,
|
| 111 |
+
93.39055633544922,
|
| 112 |
+
48.17987060546875
|
| 113 |
+
]
|
| 114 |
+
},
|
| 115 |
+
"gripper": {
|
| 116 |
+
"max": [
|
| 117 |
+
29.933774948120117
|
| 118 |
+
],
|
| 119 |
+
"min": [
|
| 120 |
+
0.0
|
| 121 |
+
],
|
| 122 |
+
"mean": [
|
| 123 |
+
7.92801570892334
|
| 124 |
+
],
|
| 125 |
+
"std": [
|
| 126 |
+
7.949193954467773
|
| 127 |
+
],
|
| 128 |
+
"q01": [
|
| 129 |
+
0.0
|
| 130 |
+
],
|
| 131 |
+
"q99": [
|
| 132 |
+
25.29801368713379
|
| 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 |
+
"tip": {
|
| 148 |
+
"resolution": [
|
| 149 |
+
640,
|
| 150 |
+
480
|
| 151 |
+
],
|
| 152 |
+
"channels": 3,
|
| 153 |
+
"fps": 30.0
|
| 154 |
+
}
|
| 155 |
+
},
|
| 156 |
+
"state": {
|
| 157 |
+
"single_arm": {
|
| 158 |
+
"absolute": true,
|
| 159 |
+
"rotation_type": null,
|
| 160 |
+
"shape": [
|
| 161 |
+
5
|
| 162 |
+
],
|
| 163 |
+
"continuous": true
|
| 164 |
+
},
|
| 165 |
+
"gripper": {
|
| 166 |
+
"absolute": true,
|
| 167 |
+
"rotation_type": null,
|
| 168 |
+
"shape": [
|
| 169 |
+
1
|
| 170 |
+
],
|
| 171 |
+
"continuous": true
|
| 172 |
+
}
|
| 173 |
+
},
|
| 174 |
+
"action": {
|
| 175 |
+
"single_arm": {
|
| 176 |
+
"absolute": true,
|
| 177 |
+
"rotation_type": null,
|
| 178 |
+
"shape": [
|
| 179 |
+
5
|
| 180 |
+
],
|
| 181 |
+
"continuous": true
|
| 182 |
+
},
|
| 183 |
+
"gripper": {
|
| 184 |
+
"absolute": true,
|
| 185 |
+
"rotation_type": null,
|
| 186 |
+
"shape": [
|
| 187 |
+
1
|
| 188 |
+
],
|
| 189 |
+
"continuous": true
|
| 190 |
+
}
|
| 191 |
+
}
|
| 192 |
+
},
|
| 193 |
+
"embodiment_tag": "new_embodiment"
|
| 194 |
+
}
|
| 195 |
+
}
|
model-00001-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bc08fad81095d3d8d49d0924b86106e874103a30fc1a39d4b887ec8bb5f4b76
|
| 3 |
+
size 4999367032
|
model-00002-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:78715a09b9f9364a08dd3a12e7289775ecb669d25ac610f0a4185d7a276980f2
|
| 3 |
+
size 2586705312
|
model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd4b274129191178877b5ad3c4f7d207287fa5d1618f1a1c6baaa122028c6106
|
| 3 |
+
size 8550720062
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cbac9370d046909e602505c362b8ee7db2e725a13796693a6162bfce5fc4bea9
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6a1e3232c9dd0558d3b582345b4a537ca2d63d9ae2df7d7189cbdc92c4e384f
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1434 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.1621621621621623,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.010810810810810811,
|
| 14 |
+
"grad_norm": 3.425095319747925,
|
| 15 |
+
"learning_rate": 9e-06,
|
| 16 |
+
"loss": 0.8544,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.021621621621621623,
|
| 21 |
+
"grad_norm": 2.206361770629883,
|
| 22 |
+
"learning_rate": 1.9e-05,
|
| 23 |
+
"loss": 0.3086,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.032432432432432434,
|
| 28 |
+
"grad_norm": 2.599766492843628,
|
| 29 |
+
"learning_rate": 2.9e-05,
|
| 30 |
+
"loss": 0.2353,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.043243243243243246,
|
| 35 |
+
"grad_norm": 1.283718466758728,
|
| 36 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 37 |
+
"loss": 0.2036,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.05405405405405406,
|
| 42 |
+
"grad_norm": 1.3836016654968262,
|
| 43 |
+
"learning_rate": 4.9e-05,
|
| 44 |
+
"loss": 0.2039,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.06486486486486487,
|
| 49 |
+
"grad_norm": 0.8957734704017639,
|
| 50 |
+
"learning_rate": 5.9e-05,
|
| 51 |
+
"loss": 0.1644,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.07567567567567568,
|
| 56 |
+
"grad_norm": 1.345668077468872,
|
| 57 |
+
"learning_rate": 6.9e-05,
|
| 58 |
+
"loss": 0.1547,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.08648648648648649,
|
| 63 |
+
"grad_norm": 1.247141718864441,
|
| 64 |
+
"learning_rate": 7.900000000000001e-05,
|
| 65 |
+
"loss": 0.1351,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0972972972972973,
|
| 70 |
+
"grad_norm": 1.0575212240219116,
|
| 71 |
+
"learning_rate": 8.900000000000001e-05,
|
| 72 |
+
"loss": 0.126,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.10810810810810811,
|
| 77 |
+
"grad_norm": 1.591833233833313,
|
| 78 |
+
"learning_rate": 9.900000000000001e-05,
|
| 79 |
+
"loss": 0.1369,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.11891891891891893,
|
| 84 |
+
"grad_norm": 1.2018483877182007,
|
| 85 |
+
"learning_rate": 9.999446382823013e-05,
|
| 86 |
+
"loss": 0.141,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.12972972972972974,
|
| 91 |
+
"grad_norm": 1.506770133972168,
|
| 92 |
+
"learning_rate": 9.997532801828658e-05,
|
| 93 |
+
"loss": 0.1242,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.14054054054054055,
|
| 98 |
+
"grad_norm": 1.3931018114089966,
|
| 99 |
+
"learning_rate": 9.99425294526634e-05,
|
| 100 |
+
"loss": 0.1332,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.15135135135135136,
|
| 105 |
+
"grad_norm": 0.5888644456863403,
|
| 106 |
+
"learning_rate": 9.989607709816091e-05,
|
| 107 |
+
"loss": 0.1147,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.16216216216216217,
|
| 112 |
+
"grad_norm": 1.298045039176941,
|
| 113 |
+
"learning_rate": 9.983598365438902e-05,
|
| 114 |
+
"loss": 0.1097,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.17297297297297298,
|
| 119 |
+
"grad_norm": 1.3051056861877441,
|
| 120 |
+
"learning_rate": 9.976226555029522e-05,
|
| 121 |
+
"loss": 0.109,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.1837837837837838,
|
| 126 |
+
"grad_norm": 0.8771129250526428,
|
| 127 |
+
"learning_rate": 9.967494293967312e-05,
|
| 128 |
+
"loss": 0.1121,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1945945945945946,
|
| 133 |
+
"grad_norm": 1.1193466186523438,
|
| 134 |
+
"learning_rate": 9.95740396956525e-05,
|
| 135 |
+
"loss": 0.1104,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.20540540540540542,
|
| 140 |
+
"grad_norm": 1.024540662765503,
|
| 141 |
+
"learning_rate": 9.945958340417283e-05,
|
| 142 |
+
"loss": 0.1036,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.21621621621621623,
|
| 147 |
+
"grad_norm": 0.6443993449211121,
|
| 148 |
+
"learning_rate": 9.93316053564413e-05,
|
| 149 |
+
"loss": 0.101,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.22702702702702704,
|
| 154 |
+
"grad_norm": 1.2472195625305176,
|
| 155 |
+
"learning_rate": 9.919014054037836e-05,
|
| 156 |
+
"loss": 0.1003,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.23783783783783785,
|
| 161 |
+
"grad_norm": 0.9055164456367493,
|
| 162 |
+
"learning_rate": 9.903522763105218e-05,
|
| 163 |
+
"loss": 0.1003,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.24864864864864866,
|
| 168 |
+
"grad_norm": 0.7446444630622864,
|
| 169 |
+
"learning_rate": 9.886690898010535e-05,
|
| 170 |
+
"loss": 0.0911,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.2594594594594595,
|
| 175 |
+
"grad_norm": 1.105281949043274,
|
| 176 |
+
"learning_rate": 9.868523060417646e-05,
|
| 177 |
+
"loss": 0.0918,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.2702702702702703,
|
| 182 |
+
"grad_norm": 0.46589380502700806,
|
| 183 |
+
"learning_rate": 9.849024217231935e-05,
|
| 184 |
+
"loss": 0.0809,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2810810810810811,
|
| 189 |
+
"grad_norm": 1.0003178119659424,
|
| 190 |
+
"learning_rate": 9.82819969924244e-05,
|
| 191 |
+
"loss": 0.0748,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.2918918918918919,
|
| 196 |
+
"grad_norm": 1.0241647958755493,
|
| 197 |
+
"learning_rate": 9.806055199664446e-05,
|
| 198 |
+
"loss": 0.0967,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.3027027027027027,
|
| 203 |
+
"grad_norm": 0.8553164601325989,
|
| 204 |
+
"learning_rate": 9.782596772583026e-05,
|
| 205 |
+
"loss": 0.0857,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.31351351351351353,
|
| 210 |
+
"grad_norm": 0.7845066785812378,
|
| 211 |
+
"learning_rate": 9.757830831297914e-05,
|
| 212 |
+
"loss": 0.0975,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.32432432432432434,
|
| 217 |
+
"grad_norm": 0.8363810181617737,
|
| 218 |
+
"learning_rate": 9.731764146570173e-05,
|
| 219 |
+
"loss": 0.0779,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.33513513513513515,
|
| 224 |
+
"grad_norm": 0.8904397487640381,
|
| 225 |
+
"learning_rate": 9.704403844771128e-05,
|
| 226 |
+
"loss": 0.0925,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.34594594594594597,
|
| 231 |
+
"grad_norm": 0.3732679784297943,
|
| 232 |
+
"learning_rate": 9.675757405934103e-05,
|
| 233 |
+
"loss": 0.0778,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.3567567567567568,
|
| 238 |
+
"grad_norm": 0.49690908193588257,
|
| 239 |
+
"learning_rate": 9.645832661709444e-05,
|
| 240 |
+
"loss": 0.0781,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.3675675675675676,
|
| 245 |
+
"grad_norm": 0.6947701573371887,
|
| 246 |
+
"learning_rate": 9.614637793223425e-05,
|
| 247 |
+
"loss": 0.0632,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.3783783783783784,
|
| 252 |
+
"grad_norm": 0.4170636832714081,
|
| 253 |
+
"learning_rate": 9.582181328841611e-05,
|
| 254 |
+
"loss": 0.0589,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.3891891891891892,
|
| 259 |
+
"grad_norm": 0.5024237036705017,
|
| 260 |
+
"learning_rate": 9.548472141837286e-05,
|
| 261 |
+
"loss": 0.0673,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.4,
|
| 266 |
+
"grad_norm": 0.5261219143867493,
|
| 267 |
+
"learning_rate": 9.513519447965595e-05,
|
| 268 |
+
"loss": 0.0723,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.41081081081081083,
|
| 273 |
+
"grad_norm": 0.2542177140712738,
|
| 274 |
+
"learning_rate": 9.477332802944044e-05,
|
| 275 |
+
"loss": 0.0645,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.42162162162162165,
|
| 280 |
+
"grad_norm": 0.5510165095329285,
|
| 281 |
+
"learning_rate": 9.439922099840054e-05,
|
| 282 |
+
"loss": 0.0769,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.43243243243243246,
|
| 287 |
+
"grad_norm": 0.5587688684463501,
|
| 288 |
+
"learning_rate": 9.401297566366318e-05,
|
| 289 |
+
"loss": 0.0684,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.44324324324324327,
|
| 294 |
+
"grad_norm": 0.5018128156661987,
|
| 295 |
+
"learning_rate": 9.36146976208462e-05,
|
| 296 |
+
"loss": 0.07,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.4540540540540541,
|
| 301 |
+
"grad_norm": 0.5472277402877808,
|
| 302 |
+
"learning_rate": 9.320449575518972e-05,
|
| 303 |
+
"loss": 0.071,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.4648648648648649,
|
| 308 |
+
"grad_norm": 0.8282442092895508,
|
| 309 |
+
"learning_rate": 9.278248221178798e-05,
|
| 310 |
+
"loss": 0.0703,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.4756756756756757,
|
| 315 |
+
"grad_norm": 0.48802706599235535,
|
| 316 |
+
"learning_rate": 9.234877236492997e-05,
|
| 317 |
+
"loss": 0.0652,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.4864864864864865,
|
| 322 |
+
"grad_norm": 0.612173855304718,
|
| 323 |
+
"learning_rate": 9.190348478655724e-05,
|
| 324 |
+
"loss": 0.0702,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.4972972972972973,
|
| 329 |
+
"grad_norm": 0.6725268363952637,
|
| 330 |
+
"learning_rate": 9.144674121384757e-05,
|
| 331 |
+
"loss": 0.0672,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.5081081081081081,
|
| 336 |
+
"grad_norm": 0.6014521718025208,
|
| 337 |
+
"learning_rate": 9.097866651593317e-05,
|
| 338 |
+
"loss": 0.0679,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.518918918918919,
|
| 343 |
+
"grad_norm": 0.4493806064128876,
|
| 344 |
+
"learning_rate": 9.049938865976275e-05,
|
| 345 |
+
"loss": 0.0608,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.5297297297297298,
|
| 350 |
+
"grad_norm": 0.44421371817588806,
|
| 351 |
+
"learning_rate": 9.000903867511666e-05,
|
| 352 |
+
"loss": 0.0577,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.5405405405405406,
|
| 357 |
+
"grad_norm": 0.7918093204498291,
|
| 358 |
+
"learning_rate": 8.950775061878453e-05,
|
| 359 |
+
"loss": 0.0615,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.5513513513513514,
|
| 364 |
+
"grad_norm": 0.47384804487228394,
|
| 365 |
+
"learning_rate": 8.899566153791566e-05,
|
| 366 |
+
"loss": 0.0651,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.5621621621621622,
|
| 371 |
+
"grad_norm": 0.3251579701900482,
|
| 372 |
+
"learning_rate": 8.84729114325516e-05,
|
| 373 |
+
"loss": 0.0572,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.572972972972973,
|
| 378 |
+
"grad_norm": 0.543025553226471,
|
| 379 |
+
"learning_rate": 8.79396432173515e-05,
|
| 380 |
+
"loss": 0.0585,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.5837837837837838,
|
| 385 |
+
"grad_norm": 0.36279764771461487,
|
| 386 |
+
"learning_rate": 8.739600268252078e-05,
|
| 387 |
+
"loss": 0.0598,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.5945945945945946,
|
| 392 |
+
"grad_norm": 0.6321111917495728,
|
| 393 |
+
"learning_rate": 8.684213845395339e-05,
|
| 394 |
+
"loss": 0.0625,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.6054054054054054,
|
| 399 |
+
"grad_norm": 0.4571816027164459,
|
| 400 |
+
"learning_rate": 8.627820195259918e-05,
|
| 401 |
+
"loss": 0.0522,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.6162162162162163,
|
| 406 |
+
"grad_norm": 0.42127525806427,
|
| 407 |
+
"learning_rate": 8.570434735306671e-05,
|
| 408 |
+
"loss": 0.0567,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.6270270270270271,
|
| 413 |
+
"grad_norm": 0.5116689205169678,
|
| 414 |
+
"learning_rate": 8.512073154147362e-05,
|
| 415 |
+
"loss": 0.0524,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.6378378378378379,
|
| 420 |
+
"grad_norm": 0.5537698864936829,
|
| 421 |
+
"learning_rate": 8.452751407255541e-05,
|
| 422 |
+
"loss": 0.0633,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.6486486486486487,
|
| 427 |
+
"grad_norm": 0.546095609664917,
|
| 428 |
+
"learning_rate": 8.392485712604483e-05,
|
| 429 |
+
"loss": 0.0569,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.6594594594594595,
|
| 434 |
+
"grad_norm": 0.7676917910575867,
|
| 435 |
+
"learning_rate": 8.331292546233362e-05,
|
| 436 |
+
"loss": 0.0625,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.6702702702702703,
|
| 441 |
+
"grad_norm": 0.48038536310195923,
|
| 442 |
+
"learning_rate": 8.269188637742846e-05,
|
| 443 |
+
"loss": 0.0556,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.6810810810810811,
|
| 448 |
+
"grad_norm": 0.48643946647644043,
|
| 449 |
+
"learning_rate": 8.206190965721419e-05,
|
| 450 |
+
"loss": 0.0503,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6918918918918919,
|
| 455 |
+
"grad_norm": 0.703184187412262,
|
| 456 |
+
"learning_rate": 8.14231675310358e-05,
|
| 457 |
+
"loss": 0.0516,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.7027027027027027,
|
| 462 |
+
"grad_norm": 0.3781812787055969,
|
| 463 |
+
"learning_rate": 8.077583462461283e-05,
|
| 464 |
+
"loss": 0.0517,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.7135135135135136,
|
| 469 |
+
"grad_norm": 0.528671145439148,
|
| 470 |
+
"learning_rate": 8.012008791229826e-05,
|
| 471 |
+
"loss": 0.05,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.7243243243243244,
|
| 476 |
+
"grad_norm": 0.7048684358596802,
|
| 477 |
+
"learning_rate": 7.945610666869568e-05,
|
| 478 |
+
"loss": 0.0587,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.7351351351351352,
|
| 483 |
+
"grad_norm": 0.36129575967788696,
|
| 484 |
+
"learning_rate": 7.878407241964729e-05,
|
| 485 |
+
"loss": 0.0579,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.745945945945946,
|
| 490 |
+
"grad_norm": 0.7521095275878906,
|
| 491 |
+
"learning_rate": 7.810416889260653e-05,
|
| 492 |
+
"loss": 0.0462,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.7567567567567568,
|
| 497 |
+
"grad_norm": 0.7604436874389648,
|
| 498 |
+
"learning_rate": 7.741658196640892e-05,
|
| 499 |
+
"loss": 0.0576,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.7675675675675676,
|
| 504 |
+
"grad_norm": 0.5860543251037598,
|
| 505 |
+
"learning_rate": 7.672149962045457e-05,
|
| 506 |
+
"loss": 0.0589,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.7783783783783784,
|
| 511 |
+
"grad_norm": 0.5799636244773865,
|
| 512 |
+
"learning_rate": 7.60191118833165e-05,
|
| 513 |
+
"loss": 0.0525,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.7891891891891892,
|
| 518 |
+
"grad_norm": 0.5772054195404053,
|
| 519 |
+
"learning_rate": 7.530961078078873e-05,
|
| 520 |
+
"loss": 0.0515,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.8,
|
| 525 |
+
"grad_norm": 0.4005531966686249,
|
| 526 |
+
"learning_rate": 7.45931902833884e-05,
|
| 527 |
+
"loss": 0.0419,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.8108108108108109,
|
| 532 |
+
"grad_norm": 0.4787760078907013,
|
| 533 |
+
"learning_rate": 7.387004625332608e-05,
|
| 534 |
+
"loss": 0.0458,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.8216216216216217,
|
| 539 |
+
"grad_norm": 0.3847050964832306,
|
| 540 |
+
"learning_rate": 7.3140376390959e-05,
|
| 541 |
+
"loss": 0.0389,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.8324324324324325,
|
| 546 |
+
"grad_norm": 0.37030264735221863,
|
| 547 |
+
"learning_rate": 7.240438018074189e-05,
|
| 548 |
+
"loss": 0.044,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.8432432432432433,
|
| 553 |
+
"grad_norm": 0.36840924620628357,
|
| 554 |
+
"learning_rate": 7.166225883668969e-05,
|
| 555 |
+
"loss": 0.0456,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.8540540540540541,
|
| 560 |
+
"grad_norm": 0.5073906183242798,
|
| 561 |
+
"learning_rate": 7.091421524736784e-05,
|
| 562 |
+
"loss": 0.0487,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.8648648648648649,
|
| 567 |
+
"grad_norm": 0.3341912627220154,
|
| 568 |
+
"learning_rate": 7.016045392042452e-05,
|
| 569 |
+
"loss": 0.0452,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.8756756756756757,
|
| 574 |
+
"grad_norm": 0.2209257185459137,
|
| 575 |
+
"learning_rate": 6.940118092668022e-05,
|
| 576 |
+
"loss": 0.0519,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.8864864864864865,
|
| 581 |
+
"grad_norm": 0.4557606279850006,
|
| 582 |
+
"learning_rate": 6.863660384379017e-05,
|
| 583 |
+
"loss": 0.0469,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.8972972972972973,
|
| 588 |
+
"grad_norm": 0.5467332601547241,
|
| 589 |
+
"learning_rate": 6.786693169949455e-05,
|
| 590 |
+
"loss": 0.047,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.9081081081081082,
|
| 595 |
+
"grad_norm": 0.4541635513305664,
|
| 596 |
+
"learning_rate": 6.709237491447249e-05,
|
| 597 |
+
"loss": 0.0472,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.918918918918919,
|
| 602 |
+
"grad_norm": 0.41976454854011536,
|
| 603 |
+
"learning_rate": 6.631314524481513e-05,
|
| 604 |
+
"loss": 0.0513,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.9297297297297298,
|
| 609 |
+
"grad_norm": 0.5546081066131592,
|
| 610 |
+
"learning_rate": 6.552945572413358e-05,
|
| 611 |
+
"loss": 0.0474,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.9405405405405406,
|
| 616 |
+
"grad_norm": 0.4505249559879303,
|
| 617 |
+
"learning_rate": 6.474152060531768e-05,
|
| 618 |
+
"loss": 0.0463,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.9513513513513514,
|
| 623 |
+
"grad_norm": 0.3605041801929474,
|
| 624 |
+
"learning_rate": 6.394955530196147e-05,
|
| 625 |
+
"loss": 0.044,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.9621621621621622,
|
| 630 |
+
"grad_norm": 0.43274059891700745,
|
| 631 |
+
"learning_rate": 6.315377632947115e-05,
|
| 632 |
+
"loss": 0.0441,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.972972972972973,
|
| 637 |
+
"grad_norm": 0.6044359803199768,
|
| 638 |
+
"learning_rate": 6.235440124587198e-05,
|
| 639 |
+
"loss": 0.0517,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.9837837837837838,
|
| 644 |
+
"grad_norm": 0.31390947103500366,
|
| 645 |
+
"learning_rate": 6.155164859233012e-05,
|
| 646 |
+
"loss": 0.0373,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.9945945945945946,
|
| 651 |
+
"grad_norm": 0.3354393541812897,
|
| 652 |
+
"learning_rate": 6.074573783340562e-05,
|
| 653 |
+
"loss": 0.0471,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.0054054054054054,
|
| 658 |
+
"grad_norm": 0.403349369764328,
|
| 659 |
+
"learning_rate": 5.9936889297052986e-05,
|
| 660 |
+
"loss": 0.0388,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.0162162162162163,
|
| 665 |
+
"grad_norm": 0.32696428894996643,
|
| 666 |
+
"learning_rate": 5.912532411438576e-05,
|
| 667 |
+
"loss": 0.0378,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.027027027027027,
|
| 672 |
+
"grad_norm": 0.40215539932250977,
|
| 673 |
+
"learning_rate": 5.831126415922148e-05,
|
| 674 |
+
"loss": 0.0422,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.037837837837838,
|
| 679 |
+
"grad_norm": 0.28796258568763733,
|
| 680 |
+
"learning_rate": 5.74949319874235e-05,
|
| 681 |
+
"loss": 0.0348,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.0486486486486486,
|
| 686 |
+
"grad_norm": 0.30033326148986816,
|
| 687 |
+
"learning_rate": 5.667655077605659e-05,
|
| 688 |
+
"loss": 0.0336,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.0594594594594595,
|
| 693 |
+
"grad_norm": 0.5106898546218872,
|
| 694 |
+
"learning_rate": 5.585634426237246e-05,
|
| 695 |
+
"loss": 0.0373,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.0702702702702702,
|
| 700 |
+
"grad_norm": 0.19478657841682434,
|
| 701 |
+
"learning_rate": 5.5034536682642224e-05,
|
| 702 |
+
"loss": 0.0399,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.0810810810810811,
|
| 707 |
+
"grad_norm": 0.27128490805625916,
|
| 708 |
+
"learning_rate": 5.4211352710852495e-05,
|
| 709 |
+
"loss": 0.0389,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.0918918918918918,
|
| 714 |
+
"grad_norm": 0.44196370244026184,
|
| 715 |
+
"learning_rate": 5.3387017397281704e-05,
|
| 716 |
+
"loss": 0.0347,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.1027027027027028,
|
| 721 |
+
"grad_norm": 0.5116639137268066,
|
| 722 |
+
"learning_rate": 5.2561756106973656e-05,
|
| 723 |
+
"loss": 0.0355,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.1135135135135135,
|
| 728 |
+
"grad_norm": 0.4929458796977997,
|
| 729 |
+
"learning_rate": 5.1735794458124956e-05,
|
| 730 |
+
"loss": 0.0388,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.1243243243243244,
|
| 735 |
+
"grad_norm": 0.41133981943130493,
|
| 736 |
+
"learning_rate": 5.0909358260403186e-05,
|
| 737 |
+
"loss": 0.041,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.135135135135135,
|
| 742 |
+
"grad_norm": 0.5629432797431946,
|
| 743 |
+
"learning_rate": 5.0082673453212914e-05,
|
| 744 |
+
"loss": 0.0351,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.145945945945946,
|
| 749 |
+
"grad_norm": 0.451533704996109,
|
| 750 |
+
"learning_rate": 4.925596604392603e-05,
|
| 751 |
+
"loss": 0.0437,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.1567567567567567,
|
| 756 |
+
"grad_norm": 0.42234864830970764,
|
| 757 |
+
"learning_rate": 4.8429462046093585e-05,
|
| 758 |
+
"loss": 0.0372,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.1675675675675676,
|
| 763 |
+
"grad_norm": 0.24306081235408783,
|
| 764 |
+
"learning_rate": 4.7603387417656026e-05,
|
| 765 |
+
"loss": 0.0374,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.1783783783783783,
|
| 770 |
+
"grad_norm": 0.4273246228694916,
|
| 771 |
+
"learning_rate": 4.677796799916845e-05,
|
| 772 |
+
"loss": 0.0294,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.1891891891891893,
|
| 777 |
+
"grad_norm": 0.43465515971183777,
|
| 778 |
+
"learning_rate": 4.5953429452058135e-05,
|
| 779 |
+
"loss": 0.0386,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.2,
|
| 784 |
+
"grad_norm": 0.4051171541213989,
|
| 785 |
+
"learning_rate": 4.5129997196930845e-05,
|
| 786 |
+
"loss": 0.0339,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.2108108108108109,
|
| 791 |
+
"grad_norm": 0.31059300899505615,
|
| 792 |
+
"learning_rate": 4.430789635194324e-05,
|
| 793 |
+
"loss": 0.0342,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.2216216216216216,
|
| 798 |
+
"grad_norm": 0.4784763753414154,
|
| 799 |
+
"learning_rate": 4.348735167125771e-05,
|
| 800 |
+
"loss": 0.0399,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.2324324324324325,
|
| 805 |
+
"grad_norm": 0.3323381245136261,
|
| 806 |
+
"learning_rate": 4.2668587483596864e-05,
|
| 807 |
+
"loss": 0.0372,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.2432432432432432,
|
| 812 |
+
"grad_norm": 0.4073697328567505,
|
| 813 |
+
"learning_rate": 4.1851827630914305e-05,
|
| 814 |
+
"loss": 0.0351,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.2540540540540541,
|
| 819 |
+
"grad_norm": 0.24014854431152344,
|
| 820 |
+
"learning_rate": 4.103729540719847e-05,
|
| 821 |
+
"loss": 0.0316,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.2648648648648648,
|
| 826 |
+
"grad_norm": 0.34608376026153564,
|
| 827 |
+
"learning_rate": 4.0225213497426276e-05,
|
| 828 |
+
"loss": 0.0335,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.2756756756756757,
|
| 833 |
+
"grad_norm": 0.3887583613395691,
|
| 834 |
+
"learning_rate": 3.9415803916683224e-05,
|
| 835 |
+
"loss": 0.0351,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.2864864864864864,
|
| 840 |
+
"grad_norm": 0.4077354967594147,
|
| 841 |
+
"learning_rate": 3.860928794946682e-05,
|
| 842 |
+
"loss": 0.0348,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.2972972972972974,
|
| 847 |
+
"grad_norm": 0.28752875328063965,
|
| 848 |
+
"learning_rate": 3.780588608918947e-05,
|
| 849 |
+
"loss": 0.0325,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.308108108108108,
|
| 854 |
+
"grad_norm": 0.3700025677680969,
|
| 855 |
+
"learning_rate": 3.700581797789786e-05,
|
| 856 |
+
"loss": 0.032,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.318918918918919,
|
| 861 |
+
"grad_norm": 0.3128696084022522,
|
| 862 |
+
"learning_rate": 3.6209302346225006e-05,
|
| 863 |
+
"loss": 0.0313,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.3297297297297297,
|
| 868 |
+
"grad_norm": 0.2689746916294098,
|
| 869 |
+
"learning_rate": 3.541655695359142e-05,
|
| 870 |
+
"loss": 0.0351,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.3405405405405406,
|
| 875 |
+
"grad_norm": 0.2882322072982788,
|
| 876 |
+
"learning_rate": 3.462779852867197e-05,
|
| 877 |
+
"loss": 0.0289,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.3513513513513513,
|
| 882 |
+
"grad_norm": 0.3429780900478363,
|
| 883 |
+
"learning_rate": 3.384324271014429e-05,
|
| 884 |
+
"loss": 0.0343,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.3621621621621622,
|
| 889 |
+
"grad_norm": 0.3094107508659363,
|
| 890 |
+
"learning_rate": 3.3063103987735433e-05,
|
| 891 |
+
"loss": 0.0313,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.372972972972973,
|
| 896 |
+
"grad_norm": 0.2886780798435211,
|
| 897 |
+
"learning_rate": 3.228759564358248e-05,
|
| 898 |
+
"loss": 0.0321,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.3837837837837839,
|
| 903 |
+
"grad_norm": 0.286609411239624,
|
| 904 |
+
"learning_rate": 3.1516929693923315e-05,
|
| 905 |
+
"loss": 0.0318,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.3945945945945946,
|
| 910 |
+
"grad_norm": 0.3236129581928253,
|
| 911 |
+
"learning_rate": 3.075131683113352e-05,
|
| 912 |
+
"loss": 0.0357,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.4054054054054055,
|
| 917 |
+
"grad_norm": 0.3914252519607544,
|
| 918 |
+
"learning_rate": 2.999096636612518e-05,
|
| 919 |
+
"loss": 0.0308,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.4162162162162162,
|
| 924 |
+
"grad_norm": 0.2484455108642578,
|
| 925 |
+
"learning_rate": 2.9236086171123404e-05,
|
| 926 |
+
"loss": 0.0286,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.427027027027027,
|
| 931 |
+
"grad_norm": 0.3120715320110321,
|
| 932 |
+
"learning_rate": 2.8486882622836026e-05,
|
| 933 |
+
"loss": 0.0323,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.4378378378378378,
|
| 938 |
+
"grad_norm": 0.31994354724884033,
|
| 939 |
+
"learning_rate": 2.774356054603243e-05,
|
| 940 |
+
"loss": 0.0289,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.4486486486486487,
|
| 945 |
+
"grad_norm": 0.29684558510780334,
|
| 946 |
+
"learning_rate": 2.7006323157546386e-05,
|
| 947 |
+
"loss": 0.0293,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.4594594594594594,
|
| 952 |
+
"grad_norm": 0.31518182158470154,
|
| 953 |
+
"learning_rate": 2.6275372010718635e-05,
|
| 954 |
+
"loss": 0.0281,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.4702702702702704,
|
| 959 |
+
"grad_norm": 0.33981916308403015,
|
| 960 |
+
"learning_rate": 2.555090694029421e-05,
|
| 961 |
+
"loss": 0.0282,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.481081081081081,
|
| 966 |
+
"grad_norm": 0.3087005019187927,
|
| 967 |
+
"learning_rate": 2.4833126007789653e-05,
|
| 968 |
+
"loss": 0.0225,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.491891891891892,
|
| 973 |
+
"grad_norm": 0.3447257876396179,
|
| 974 |
+
"learning_rate": 2.4122225447344875e-05,
|
| 975 |
+
"loss": 0.0269,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.5027027027027027,
|
| 980 |
+
"grad_norm": 0.1885969340801239,
|
| 981 |
+
"learning_rate": 2.341839961207482e-05,
|
| 982 |
+
"loss": 0.0219,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.5135135135135136,
|
| 987 |
+
"grad_norm": 0.2396612912416458,
|
| 988 |
+
"learning_rate": 2.2721840920935196e-05,
|
| 989 |
+
"loss": 0.0316,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.5243243243243243,
|
| 994 |
+
"grad_norm": 0.3243270814418793,
|
| 995 |
+
"learning_rate": 2.2032739806117058e-05,
|
| 996 |
+
"loss": 0.026,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.535135135135135,
|
| 1001 |
+
"grad_norm": 0.29046112298965454,
|
| 1002 |
+
"learning_rate": 2.1351284660984572e-05,
|
| 1003 |
+
"loss": 0.0299,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.545945945945946,
|
| 1008 |
+
"grad_norm": 0.3897278308868408,
|
| 1009 |
+
"learning_rate": 2.067766178857013e-05,
|
| 1010 |
+
"loss": 0.0295,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.5567567567567568,
|
| 1015 |
+
"grad_norm": 0.23025821149349213,
|
| 1016 |
+
"learning_rate": 2.0012055350640986e-05,
|
| 1017 |
+
"loss": 0.0277,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.5675675675675675,
|
| 1022 |
+
"grad_norm": 0.20442277193069458,
|
| 1023 |
+
"learning_rate": 1.9354647317351188e-05,
|
| 1024 |
+
"loss": 0.0278,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.5783783783783782,
|
| 1029 |
+
"grad_norm": 0.17502537369728088,
|
| 1030 |
+
"learning_rate": 1.8705617417492883e-05,
|
| 1031 |
+
"loss": 0.0285,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.5891891891891892,
|
| 1036 |
+
"grad_norm": 0.39069363474845886,
|
| 1037 |
+
"learning_rate": 1.8065143089360172e-05,
|
| 1038 |
+
"loss": 0.0295,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.6,
|
| 1043 |
+
"grad_norm": 0.3228617012500763,
|
| 1044 |
+
"learning_rate": 1.743339943223926e-05,
|
| 1045 |
+
"loss": 0.0302,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.6108108108108108,
|
| 1050 |
+
"grad_norm": 0.31476905941963196,
|
| 1051 |
+
"learning_rate": 1.6810559158538092e-05,
|
| 1052 |
+
"loss": 0.0237,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.6216216216216215,
|
| 1057 |
+
"grad_norm": 0.3141558766365051,
|
| 1058 |
+
"learning_rate": 1.6196792546568472e-05,
|
| 1059 |
+
"loss": 0.0319,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.6324324324324324,
|
| 1064 |
+
"grad_norm": 0.25229302048683167,
|
| 1065 |
+
"learning_rate": 1.5592267393993716e-05,
|
| 1066 |
+
"loss": 0.0255,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.6432432432432433,
|
| 1071 |
+
"grad_norm": 0.255024790763855,
|
| 1072 |
+
"learning_rate": 1.4997148971954344e-05,
|
| 1073 |
+
"loss": 0.0274,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.654054054054054,
|
| 1078 |
+
"grad_norm": 0.275570809841156,
|
| 1079 |
+
"learning_rate": 1.4411599979884744e-05,
|
| 1080 |
+
"loss": 0.0266,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.6648648648648647,
|
| 1085 |
+
"grad_norm": 0.28709614276885986,
|
| 1086 |
+
"learning_rate": 1.383578050103268e-05,
|
| 1087 |
+
"loss": 0.0277,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.6756756756756757,
|
| 1092 |
+
"grad_norm": 0.3334985673427582,
|
| 1093 |
+
"learning_rate": 1.3269847958694148e-05,
|
| 1094 |
+
"loss": 0.0264,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.6864864864864866,
|
| 1099 |
+
"grad_norm": 0.20972363650798798,
|
| 1100 |
+
"learning_rate": 1.2713957073175425e-05,
|
| 1101 |
+
"loss": 0.0247,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.6972972972972973,
|
| 1106 |
+
"grad_norm": 0.32650667428970337,
|
| 1107 |
+
"learning_rate": 1.2168259819494066e-05,
|
| 1108 |
+
"loss": 0.0257,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.708108108108108,
|
| 1113 |
+
"grad_norm": 0.22780145704746246,
|
| 1114 |
+
"learning_rate": 1.1632905385830484e-05,
|
| 1115 |
+
"loss": 0.025,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.718918918918919,
|
| 1120 |
+
"grad_norm": 0.2709677815437317,
|
| 1121 |
+
"learning_rate": 1.1108040132741354e-05,
|
| 1122 |
+
"loss": 0.0266,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.7297297297297298,
|
| 1127 |
+
"grad_norm": 0.278166264295578,
|
| 1128 |
+
"learning_rate": 1.059380755314613e-05,
|
| 1129 |
+
"loss": 0.0258,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.7405405405405405,
|
| 1134 |
+
"grad_norm": 0.27671316266059875,
|
| 1135 |
+
"learning_rate": 1.009034823309749e-05,
|
| 1136 |
+
"loss": 0.0226,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.7513513513513512,
|
| 1141 |
+
"grad_norm": 0.1992180347442627,
|
| 1142 |
+
"learning_rate": 9.597799813346525e-06,
|
| 1143 |
+
"loss": 0.0209,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.7621621621621621,
|
| 1148 |
+
"grad_norm": 0.32951998710632324,
|
| 1149 |
+
"learning_rate": 9.116296951713133e-06,
|
| 1150 |
+
"loss": 0.0237,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.772972972972973,
|
| 1155 |
+
"grad_norm": 0.2336224764585495,
|
| 1156 |
+
"learning_rate": 8.645971286271904e-06,
|
| 1157 |
+
"loss": 0.0251,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.7837837837837838,
|
| 1162 |
+
"grad_norm": 0.31754037737846375,
|
| 1163 |
+
"learning_rate": 8.186951399363613e-06,
|
| 1164 |
+
"loss": 0.029,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.7945945945945945,
|
| 1169 |
+
"grad_norm": 0.2530142664909363,
|
| 1170 |
+
"learning_rate": 7.739362782442021e-06,
|
| 1171 |
+
"loss": 0.0207,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.8054054054054054,
|
| 1176 |
+
"grad_norm": 0.27058207988739014,
|
| 1177 |
+
"learning_rate": 7.30332780176588e-06,
|
| 1178 |
+
"loss": 0.0237,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.8162162162162163,
|
| 1183 |
+
"grad_norm": 0.26931828260421753,
|
| 1184 |
+
"learning_rate": 6.878965664945108e-06,
|
| 1185 |
+
"loss": 0.0268,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.827027027027027,
|
| 1190 |
+
"grad_norm": 0.23155780136585236,
|
| 1191 |
+
"learning_rate": 6.466392388350695e-06,
|
| 1192 |
+
"loss": 0.0262,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.8378378378378377,
|
| 1197 |
+
"grad_norm": 0.3071526885032654,
|
| 1198 |
+
"learning_rate": 6.0657207653969315e-06,
|
| 1199 |
+
"loss": 0.0242,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.8486486486486486,
|
| 1204 |
+
"grad_norm": 0.3347031772136688,
|
| 1205 |
+
"learning_rate": 5.67706033570487e-06,
|
| 1206 |
+
"loss": 0.0229,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.8594594594594596,
|
| 1211 |
+
"grad_norm": 0.28099486231803894,
|
| 1212 |
+
"learning_rate": 5.300517355155215e-06,
|
| 1213 |
+
"loss": 0.0238,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.8702702702702703,
|
| 1218 |
+
"grad_norm": 0.24978019297122955,
|
| 1219 |
+
"learning_rate": 4.936194766839103e-06,
|
| 1220 |
+
"loss": 0.0273,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.881081081081081,
|
| 1225 |
+
"grad_norm": 0.23257969319820404,
|
| 1226 |
+
"learning_rate": 4.5841921729144424e-06,
|
| 1227 |
+
"loss": 0.0228,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.8918918918918919,
|
| 1232 |
+
"grad_norm": 0.31350693106651306,
|
| 1233 |
+
"learning_rate": 4.244605807375679e-06,
|
| 1234 |
+
"loss": 0.0279,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.9027027027027028,
|
| 1239 |
+
"grad_norm": 0.2184530347585678,
|
| 1240 |
+
"learning_rate": 3.917528509744412e-06,
|
| 1241 |
+
"loss": 0.0266,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.9135135135135135,
|
| 1246 |
+
"grad_norm": 0.295645534992218,
|
| 1247 |
+
"learning_rate": 3.60304969968796e-06,
|
| 1248 |
+
"loss": 0.0242,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.9243243243243242,
|
| 1253 |
+
"grad_norm": 0.21473683416843414,
|
| 1254 |
+
"learning_rate": 3.301255352572946e-06,
|
| 1255 |
+
"loss": 0.0208,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.9351351351351351,
|
| 1260 |
+
"grad_norm": 0.29790809750556946,
|
| 1261 |
+
"learning_rate": 3.0122279759604745e-06,
|
| 1262 |
+
"loss": 0.0268,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.945945945945946,
|
| 1267 |
+
"grad_norm": 0.2922970950603485,
|
| 1268 |
+
"learning_rate": 2.73604658704939e-06,
|
| 1269 |
+
"loss": 0.0237,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.9567567567567568,
|
| 1274 |
+
"grad_norm": 0.15432117879390717,
|
| 1275 |
+
"learning_rate": 2.4727866910737583e-06,
|
| 1276 |
+
"loss": 0.0217,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.9675675675675675,
|
| 1281 |
+
"grad_norm": 0.3692725896835327,
|
| 1282 |
+
"learning_rate": 2.222520260660521e-06,
|
| 1283 |
+
"loss": 0.0207,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.9783783783783784,
|
| 1288 |
+
"grad_norm": 0.30365312099456787,
|
| 1289 |
+
"learning_rate": 1.985315716152847e-06,
|
| 1290 |
+
"loss": 0.0231,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.9891891891891893,
|
| 1295 |
+
"grad_norm": 0.18066595494747162,
|
| 1296 |
+
"learning_rate": 1.7612379069047335e-06,
|
| 1297 |
+
"loss": 0.0204,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 2.0,
|
| 1302 |
+
"grad_norm": 0.21528354287147522,
|
| 1303 |
+
"learning_rate": 1.550348093551829e-06,
|
| 1304 |
+
"loss": 0.0199,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 2.0108108108108107,
|
| 1309 |
+
"grad_norm": 0.22604824602603912,
|
| 1310 |
+
"learning_rate": 1.3527039312633827e-06,
|
| 1311 |
+
"loss": 0.0268,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 2.0216216216216214,
|
| 1316 |
+
"grad_norm": 0.1855325549840927,
|
| 1317 |
+
"learning_rate": 1.1683594539798893e-06,
|
| 1318 |
+
"loss": 0.0209,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 2.0324324324324325,
|
| 1323 |
+
"grad_norm": 0.21577879786491394,
|
| 1324 |
+
"learning_rate": 9.97365059640787e-07,
|
| 1325 |
+
"loss": 0.0218,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 2.0432432432432432,
|
| 1330 |
+
"grad_norm": 0.1882552057504654,
|
| 1331 |
+
"learning_rate": 8.397674964061075e-07,
|
| 1332 |
+
"loss": 0.0185,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 2.054054054054054,
|
| 1337 |
+
"grad_norm": 0.21449606120586395,
|
| 1338 |
+
"learning_rate": 6.956098498760389e-07,
|
| 1339 |
+
"loss": 0.0243,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 2.064864864864865,
|
| 1344 |
+
"grad_norm": 0.33573615550994873,
|
| 1345 |
+
"learning_rate": 5.64931531311741e-07,
|
| 1346 |
+
"loss": 0.0232,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 2.075675675675676,
|
| 1351 |
+
"grad_norm": 0.2893902361392975,
|
| 1352 |
+
"learning_rate": 4.4776826686069305e-07,
|
| 1353 |
+
"loss": 0.0237,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 2.0864864864864865,
|
| 1358 |
+
"grad_norm": 0.24677212536334991,
|
| 1359 |
+
"learning_rate": 3.441520877895288e-07,
|
| 1360 |
+
"loss": 0.0236,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 2.097297297297297,
|
| 1365 |
+
"grad_norm": 0.3380148112773895,
|
| 1366 |
+
"learning_rate": 2.5411132172700194e-07,
|
| 1367 |
+
"loss": 0.0229,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 2.108108108108108,
|
| 1372 |
+
"grad_norm": 0.1759500801563263,
|
| 1373 |
+
"learning_rate": 1.776705849195037e-07,
|
| 1374 |
+
"loss": 0.0222,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 2.118918918918919,
|
| 1379 |
+
"grad_norm": 0.20293094217777252,
|
| 1380 |
+
"learning_rate": 1.1485077550122402e-07,
|
| 1381 |
+
"loss": 0.0179,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.1297297297297297,
|
| 1386 |
+
"grad_norm": 0.1787509024143219,
|
| 1387 |
+
"learning_rate": 6.566906778079917e-08,
|
| 1388 |
+
"loss": 0.0249,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.1405405405405404,
|
| 1393 |
+
"grad_norm": 0.343007355928421,
|
| 1394 |
+
"learning_rate": 3.01389075460512e-08,
|
| 1395 |
+
"loss": 0.0279,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 2.1513513513513516,
|
| 1400 |
+
"grad_norm": 0.179966539144516,
|
| 1401 |
+
"learning_rate": 8.270008388022721e-09,
|
| 1402 |
+
"loss": 0.0221,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 2.1621621621621623,
|
| 1407 |
+
"grad_norm": 0.21802319586277008,
|
| 1408 |
+
"learning_rate": 6.834904537900144e-11,
|
| 1409 |
+
"loss": 0.019,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
}
|
| 1412 |
+
],
|
| 1413 |
+
"logging_steps": 10,
|
| 1414 |
+
"max_steps": 2000,
|
| 1415 |
+
"num_input_tokens_seen": 0,
|
| 1416 |
+
"num_train_epochs": 3,
|
| 1417 |
+
"save_steps": 1000,
|
| 1418 |
+
"stateful_callbacks": {
|
| 1419 |
+
"TrainerControl": {
|
| 1420 |
+
"args": {
|
| 1421 |
+
"should_epoch_stop": false,
|
| 1422 |
+
"should_evaluate": false,
|
| 1423 |
+
"should_log": false,
|
| 1424 |
+
"should_save": true,
|
| 1425 |
+
"should_training_stop": true
|
| 1426 |
+
},
|
| 1427 |
+
"attributes": {}
|
| 1428 |
+
}
|
| 1429 |
+
},
|
| 1430 |
+
"total_flos": 0.0,
|
| 1431 |
+
"train_batch_size": 32,
|
| 1432 |
+
"trial_name": null,
|
| 1433 |
+
"trial_params": null
|
| 1434 |
+
}
|