Add files using upload-large-folder tool
Browse files- README.md +1 -0
- config.json +64 -0
- experiment_cfg/metadata.json +187 -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_0.pth +3 -0
- rng_state_1.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +734 -0
README.md
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
Finetuned on 333x3 episodes of red, green, & blue cubes being touched. Generalizes well to other colors and shapes. Checkpointed at 1k steps.
|
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 +1,187 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"new_embodiment": {
|
| 3 |
+
"statistics": {
|
| 4 |
+
"state": {
|
| 5 |
+
"single_arm": {
|
| 6 |
+
"max": [
|
| 7 |
+
2.0003252029418945,
|
| 8 |
+
2.8696048259735107,
|
| 9 |
+
-2.973998198285699e-05,
|
| 10 |
+
1.2001643180847168,
|
| 11 |
+
1.636069416999817
|
| 12 |
+
],
|
| 13 |
+
"min": [
|
| 14 |
+
-1.6366058588027954,
|
| 15 |
+
-0.0013483419315889478,
|
| 16 |
+
-2.867553949356079,
|
| 17 |
+
-0.15347807109355927,
|
| 18 |
+
-2.408719539642334
|
| 19 |
+
],
|
| 20 |
+
"mean": [
|
| 21 |
+
0.5613881349563599,
|
| 22 |
+
1.2009493112564087,
|
| 23 |
+
-1.1912317276000977,
|
| 24 |
+
0.7247916460037231,
|
| 25 |
+
-0.5904473662376404
|
| 26 |
+
],
|
| 27 |
+
"std": [
|
| 28 |
+
0.9785857200622559,
|
| 29 |
+
0.7331446409225464,
|
| 30 |
+
0.6282459497451782,
|
| 31 |
+
0.23302637040615082,
|
| 32 |
+
0.9657405614852905
|
| 33 |
+
],
|
| 34 |
+
"q01": [
|
| 35 |
+
-1.5449657726287842,
|
| 36 |
+
-1.0683193213480989e-05,
|
| 37 |
+
-2.7843262195587157,
|
| 38 |
+
0.2979250168800354,
|
| 39 |
+
-1.9947426080703736
|
| 40 |
+
],
|
| 41 |
+
"q99": [
|
| 42 |
+
1.9999996423721313,
|
| 43 |
+
2.7311124897003163,
|
| 44 |
+
-0.3246664524078371,
|
| 45 |
+
1.2000004053115845,
|
| 46 |
+
1.5377632951736446
|
| 47 |
+
]
|
| 48 |
+
},
|
| 49 |
+
"gripper": {
|
| 50 |
+
"max": [
|
| 51 |
+
0.703035831451416
|
| 52 |
+
],
|
| 53 |
+
"min": [
|
| 54 |
+
-0.06299631297588348
|
| 55 |
+
],
|
| 56 |
+
"mean": [
|
| 57 |
+
0.024504341185092926
|
| 58 |
+
],
|
| 59 |
+
"std": [
|
| 60 |
+
0.08595108985900879
|
| 61 |
+
],
|
| 62 |
+
"q01": [
|
| 63 |
+
-0.010759906060993672
|
| 64 |
+
],
|
| 65 |
+
"q99": [
|
| 66 |
+
0.4408635914325714
|
| 67 |
+
]
|
| 68 |
+
}
|
| 69 |
+
},
|
| 70 |
+
"action": {
|
| 71 |
+
"single_arm": {
|
| 72 |
+
"max": [
|
| 73 |
+
2.000000476837158,
|
| 74 |
+
2.8696048259735107,
|
| 75 |
+
-0.029101531952619553,
|
| 76 |
+
1.2001643180847168,
|
| 77 |
+
1.6363924741744995
|
| 78 |
+
],
|
| 79 |
+
"min": [
|
| 80 |
+
-1.6367615461349487,
|
| 81 |
+
-0.0013483419315889478,
|
| 82 |
+
-2.870177745819092,
|
| 83 |
+
-0.15347807109355927,
|
| 84 |
+
-2.378706455230713
|
| 85 |
+
],
|
| 86 |
+
"mean": [
|
| 87 |
+
0.5605649352073669,
|
| 88 |
+
1.2173237800598145,
|
| 89 |
+
-1.1972321271896362,
|
| 90 |
+
0.7227405309677124,
|
| 91 |
+
-0.5876930356025696
|
| 92 |
+
],
|
| 93 |
+
"std": [
|
| 94 |
+
0.9839946627616882,
|
| 95 |
+
0.7347026467323303,
|
| 96 |
+
0.6269937753677368,
|
| 97 |
+
0.23072293400764465,
|
| 98 |
+
0.9713127017021179
|
| 99 |
+
],
|
| 100 |
+
"q01": [
|
| 101 |
+
-1.5492996978759765,
|
| 102 |
+
-8.676056131662335e-06,
|
| 103 |
+
-2.7688085556030275,
|
| 104 |
+
0.30216108322143553,
|
| 105 |
+
-1.9937553739547729
|
| 106 |
+
],
|
| 107 |
+
"q99": [
|
| 108 |
+
1.9999995231628418,
|
| 109 |
+
2.744625654220581,
|
| 110 |
+
-0.33259372472763077,
|
| 111 |
+
1.200000286102295,
|
| 112 |
+
1.5410190296173096
|
| 113 |
+
]
|
| 114 |
+
},
|
| 115 |
+
"gripper": {
|
| 116 |
+
"max": [
|
| 117 |
+
0.703035831451416
|
| 118 |
+
],
|
| 119 |
+
"min": [
|
| 120 |
+
-0.06299631297588348
|
| 121 |
+
],
|
| 122 |
+
"mean": [
|
| 123 |
+
0.02159128151834011
|
| 124 |
+
],
|
| 125 |
+
"std": [
|
| 126 |
+
0.07854700088500977
|
| 127 |
+
],
|
| 128 |
+
"q01": [
|
| 129 |
+
-0.01098821684718132
|
| 130 |
+
],
|
| 131 |
+
"q99": [
|
| 132 |
+
0.41087187886238097
|
| 133 |
+
]
|
| 134 |
+
}
|
| 135 |
+
}
|
| 136 |
+
},
|
| 137 |
+
"modalities": {
|
| 138 |
+
"video": {
|
| 139 |
+
"webcam": {
|
| 140 |
+
"resolution": [
|
| 141 |
+
640,
|
| 142 |
+
480
|
| 143 |
+
],
|
| 144 |
+
"channels": 3,
|
| 145 |
+
"fps": 30.0
|
| 146 |
+
}
|
| 147 |
+
},
|
| 148 |
+
"state": {
|
| 149 |
+
"single_arm": {
|
| 150 |
+
"absolute": true,
|
| 151 |
+
"rotation_type": null,
|
| 152 |
+
"shape": [
|
| 153 |
+
5
|
| 154 |
+
],
|
| 155 |
+
"continuous": true
|
| 156 |
+
},
|
| 157 |
+
"gripper": {
|
| 158 |
+
"absolute": true,
|
| 159 |
+
"rotation_type": null,
|
| 160 |
+
"shape": [
|
| 161 |
+
1
|
| 162 |
+
],
|
| 163 |
+
"continuous": true
|
| 164 |
+
}
|
| 165 |
+
},
|
| 166 |
+
"action": {
|
| 167 |
+
"single_arm": {
|
| 168 |
+
"absolute": true,
|
| 169 |
+
"rotation_type": null,
|
| 170 |
+
"shape": [
|
| 171 |
+
5
|
| 172 |
+
],
|
| 173 |
+
"continuous": true
|
| 174 |
+
},
|
| 175 |
+
"gripper": {
|
| 176 |
+
"absolute": true,
|
| 177 |
+
"rotation_type": null,
|
| 178 |
+
"shape": [
|
| 179 |
+
1
|
| 180 |
+
],
|
| 181 |
+
"continuous": true
|
| 182 |
+
}
|
| 183 |
+
}
|
| 184 |
+
},
|
| 185 |
+
"embodiment_tag": "new_embodiment"
|
| 186 |
+
}
|
| 187 |
+
}
|
model-00001-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a5a2e3982d2f0d475a6bbfd4b7eb7ace8d29cb89aead3eba6a96736e3138ec5
|
| 3 |
+
size 4999367032
|
model-00002-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f8f6e3d4c9ae4e9a3ed1d025d893b99a236845708cd12495469812f41b57a3d
|
| 3 |
+
size 2586508600
|
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:f75938496ae2e6484380c63bc241c2aa134598e299c3cde3af74c90448f6a3cc
|
| 3 |
+
size 8550325978
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6942e7d0fad9ea5ca8d6849b863c5fca113d1802c0c6b4b6cb63f75db30e17a1
|
| 3 |
+
size 14512
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:675c736ed11ba57de7a85176857b7317f9f79aec1c909ebf5cc00810df70079a
|
| 3 |
+
size 14512
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:057194249d9cdc822a0752df7fedc436fc30dce92062cd380d9a3f5704199672
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,734 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.8090614886731392,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.008090614886731391,
|
| 14 |
+
"grad_norm": 7.823570251464844,
|
| 15 |
+
"learning_rate": 6e-06,
|
| 16 |
+
"loss": 0.8903,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.016181229773462782,
|
| 21 |
+
"grad_norm": 2.272317886352539,
|
| 22 |
+
"learning_rate": 1.2666666666666668e-05,
|
| 23 |
+
"loss": 0.4601,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.024271844660194174,
|
| 28 |
+
"grad_norm": 1.5555341243743896,
|
| 29 |
+
"learning_rate": 1.9333333333333333e-05,
|
| 30 |
+
"loss": 0.276,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.032362459546925564,
|
| 35 |
+
"grad_norm": 1.7244459390640259,
|
| 36 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 37 |
+
"loss": 0.1865,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.040453074433656956,
|
| 42 |
+
"grad_norm": 1.4668376445770264,
|
| 43 |
+
"learning_rate": 3.266666666666667e-05,
|
| 44 |
+
"loss": 0.1422,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.04854368932038835,
|
| 49 |
+
"grad_norm": 2.02225661277771,
|
| 50 |
+
"learning_rate": 3.933333333333333e-05,
|
| 51 |
+
"loss": 0.1238,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.05663430420711974,
|
| 56 |
+
"grad_norm": 1.1487993001937866,
|
| 57 |
+
"learning_rate": 4.600000000000001e-05,
|
| 58 |
+
"loss": 0.0791,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.06472491909385113,
|
| 63 |
+
"grad_norm": 0.5569996237754822,
|
| 64 |
+
"learning_rate": 5.266666666666666e-05,
|
| 65 |
+
"loss": 0.0819,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.07281553398058252,
|
| 70 |
+
"grad_norm": 1.194751262664795,
|
| 71 |
+
"learning_rate": 5.9333333333333343e-05,
|
| 72 |
+
"loss": 0.0698,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.08090614886731391,
|
| 77 |
+
"grad_norm": 0.8545956015586853,
|
| 78 |
+
"learning_rate": 6.6e-05,
|
| 79 |
+
"loss": 0.0732,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.0889967637540453,
|
| 84 |
+
"grad_norm": 0.6914941072463989,
|
| 85 |
+
"learning_rate": 7.266666666666667e-05,
|
| 86 |
+
"loss": 0.0641,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.0970873786407767,
|
| 91 |
+
"grad_norm": 0.4049356281757355,
|
| 92 |
+
"learning_rate": 7.933333333333334e-05,
|
| 93 |
+
"loss": 0.0565,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.10517799352750809,
|
| 98 |
+
"grad_norm": 0.7143370509147644,
|
| 99 |
+
"learning_rate": 8.6e-05,
|
| 100 |
+
"loss": 0.0555,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.11326860841423948,
|
| 105 |
+
"grad_norm": 0.9746547937393188,
|
| 106 |
+
"learning_rate": 9.266666666666666e-05,
|
| 107 |
+
"loss": 0.0523,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.12135922330097088,
|
| 112 |
+
"grad_norm": 0.6225072145462036,
|
| 113 |
+
"learning_rate": 9.933333333333334e-05,
|
| 114 |
+
"loss": 0.0492,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.12944983818770225,
|
| 119 |
+
"grad_norm": 0.6568852066993713,
|
| 120 |
+
"learning_rate": 9.999753945398704e-05,
|
| 121 |
+
"loss": 0.0478,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.13754045307443366,
|
| 126 |
+
"grad_norm": 0.345377653837204,
|
| 127 |
+
"learning_rate": 9.998903417374228e-05,
|
| 128 |
+
"loss": 0.0465,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.14563106796116504,
|
| 133 |
+
"grad_norm": 0.756610095500946,
|
| 134 |
+
"learning_rate": 9.997445481536973e-05,
|
| 135 |
+
"loss": 0.0467,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.15372168284789645,
|
| 140 |
+
"grad_norm": 0.494722843170166,
|
| 141 |
+
"learning_rate": 9.995380315038119e-05,
|
| 142 |
+
"loss": 0.0407,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.16181229773462782,
|
| 147 |
+
"grad_norm": 0.7570058107376099,
|
| 148 |
+
"learning_rate": 9.99270816881235e-05,
|
| 149 |
+
"loss": 0.042,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.16990291262135923,
|
| 154 |
+
"grad_norm": 0.3776613473892212,
|
| 155 |
+
"learning_rate": 9.989429367547377e-05,
|
| 156 |
+
"loss": 0.0429,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.1779935275080906,
|
| 161 |
+
"grad_norm": 0.5018635988235474,
|
| 162 |
+
"learning_rate": 9.985544309644475e-05,
|
| 163 |
+
"loss": 0.0367,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.18608414239482202,
|
| 168 |
+
"grad_norm": 0.3979959487915039,
|
| 169 |
+
"learning_rate": 9.98105346717008e-05,
|
| 170 |
+
"loss": 0.0341,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.1941747572815534,
|
| 175 |
+
"grad_norm": 0.5319691300392151,
|
| 176 |
+
"learning_rate": 9.97595738579843e-05,
|
| 177 |
+
"loss": 0.0347,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.2022653721682848,
|
| 182 |
+
"grad_norm": 0.5348716974258423,
|
| 183 |
+
"learning_rate": 9.970256684745258e-05,
|
| 184 |
+
"loss": 0.0308,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.21035598705501618,
|
| 189 |
+
"grad_norm": 0.6536835432052612,
|
| 190 |
+
"learning_rate": 9.963952056692549e-05,
|
| 191 |
+
"loss": 0.0301,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.21844660194174756,
|
| 196 |
+
"grad_norm": 0.5518924593925476,
|
| 197 |
+
"learning_rate": 9.957044267704384e-05,
|
| 198 |
+
"loss": 0.0364,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.22653721682847897,
|
| 203 |
+
"grad_norm": 0.40961554646492004,
|
| 204 |
+
"learning_rate": 9.949534157133844e-05,
|
| 205 |
+
"loss": 0.0293,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.23462783171521034,
|
| 210 |
+
"grad_norm": 0.7217941880226135,
|
| 211 |
+
"learning_rate": 9.941422637521035e-05,
|
| 212 |
+
"loss": 0.0363,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.24271844660194175,
|
| 217 |
+
"grad_norm": 0.5057093501091003,
|
| 218 |
+
"learning_rate": 9.932710694482191e-05,
|
| 219 |
+
"loss": 0.0295,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.25080906148867316,
|
| 224 |
+
"grad_norm": 0.75019770860672,
|
| 225 |
+
"learning_rate": 9.923399386589933e-05,
|
| 226 |
+
"loss": 0.033,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.2588996763754045,
|
| 231 |
+
"grad_norm": 0.48032259941101074,
|
| 232 |
+
"learning_rate": 9.913489845244626e-05,
|
| 233 |
+
"loss": 0.0277,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.2669902912621359,
|
| 238 |
+
"grad_norm": 0.5418844223022461,
|
| 239 |
+
"learning_rate": 9.902983274536912e-05,
|
| 240 |
+
"loss": 0.0264,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2750809061488673,
|
| 245 |
+
"grad_norm": 0.5400299429893494,
|
| 246 |
+
"learning_rate": 9.891880951101407e-05,
|
| 247 |
+
"loss": 0.028,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.28317152103559873,
|
| 252 |
+
"grad_norm": 0.7953261733055115,
|
| 253 |
+
"learning_rate": 9.880184223961573e-05,
|
| 254 |
+
"loss": 0.0245,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.2912621359223301,
|
| 259 |
+
"grad_norm": 0.47858819365501404,
|
| 260 |
+
"learning_rate": 9.867894514365802e-05,
|
| 261 |
+
"loss": 0.0272,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.2993527508090615,
|
| 266 |
+
"grad_norm": 0.3962852954864502,
|
| 267 |
+
"learning_rate": 9.855013315614725e-05,
|
| 268 |
+
"loss": 0.0262,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.3074433656957929,
|
| 273 |
+
"grad_norm": 0.840124785900116,
|
| 274 |
+
"learning_rate": 9.841542192879762e-05,
|
| 275 |
+
"loss": 0.0249,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3155339805825243,
|
| 280 |
+
"grad_norm": 0.6563257575035095,
|
| 281 |
+
"learning_rate": 9.82748278301294e-05,
|
| 282 |
+
"loss": 0.0268,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.32362459546925565,
|
| 287 |
+
"grad_norm": 0.4959515929222107,
|
| 288 |
+
"learning_rate": 9.812836794348004e-05,
|
| 289 |
+
"loss": 0.0241,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.33171521035598706,
|
| 294 |
+
"grad_norm": 0.5108354091644287,
|
| 295 |
+
"learning_rate": 9.797606006492841e-05,
|
| 296 |
+
"loss": 0.0329,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.33980582524271846,
|
| 301 |
+
"grad_norm": 0.5321593880653381,
|
| 302 |
+
"learning_rate": 9.781792270113241e-05,
|
| 303 |
+
"loss": 0.0248,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3478964401294498,
|
| 308 |
+
"grad_norm": 0.7112411856651306,
|
| 309 |
+
"learning_rate": 9.765397506708023e-05,
|
| 310 |
+
"loss": 0.0266,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3559870550161812,
|
| 315 |
+
"grad_norm": 0.4580034911632538,
|
| 316 |
+
"learning_rate": 9.748423708375563e-05,
|
| 317 |
+
"loss": 0.0228,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.3640776699029126,
|
| 322 |
+
"grad_norm": 0.43798476457595825,
|
| 323 |
+
"learning_rate": 9.730872937571739e-05,
|
| 324 |
+
"loss": 0.0239,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.37216828478964403,
|
| 329 |
+
"grad_norm": 0.5347399711608887,
|
| 330 |
+
"learning_rate": 9.712747326859315e-05,
|
| 331 |
+
"loss": 0.0265,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.3802588996763754,
|
| 336 |
+
"grad_norm": 0.5635089874267578,
|
| 337 |
+
"learning_rate": 9.69404907864883e-05,
|
| 338 |
+
"loss": 0.027,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.3883495145631068,
|
| 343 |
+
"grad_norm": 0.3755838871002197,
|
| 344 |
+
"learning_rate": 9.674780464930979e-05,
|
| 345 |
+
"loss": 0.0234,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.3964401294498382,
|
| 350 |
+
"grad_norm": 0.522113561630249,
|
| 351 |
+
"learning_rate": 9.654943827000548e-05,
|
| 352 |
+
"loss": 0.021,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.4045307443365696,
|
| 357 |
+
"grad_norm": 0.43958115577697754,
|
| 358 |
+
"learning_rate": 9.634541575171929e-05,
|
| 359 |
+
"loss": 0.0214,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.41262135922330095,
|
| 364 |
+
"grad_norm": 0.4480895698070526,
|
| 365 |
+
"learning_rate": 9.613576188486253e-05,
|
| 366 |
+
"loss": 0.0258,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.42071197411003236,
|
| 371 |
+
"grad_norm": 0.410576730966568,
|
| 372 |
+
"learning_rate": 9.59205021441015e-05,
|
| 373 |
+
"loss": 0.0226,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.42880258899676377,
|
| 378 |
+
"grad_norm": 0.45604780316352844,
|
| 379 |
+
"learning_rate": 9.569966268526232e-05,
|
| 380 |
+
"loss": 0.0256,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.4368932038834951,
|
| 385 |
+
"grad_norm": 0.30307430028915405,
|
| 386 |
+
"learning_rate": 9.54732703421526e-05,
|
| 387 |
+
"loss": 0.0204,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4449838187702265,
|
| 392 |
+
"grad_norm": 0.29722708463668823,
|
| 393 |
+
"learning_rate": 9.524135262330098e-05,
|
| 394 |
+
"loss": 0.0198,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.45307443365695793,
|
| 399 |
+
"grad_norm": 0.38580235838890076,
|
| 400 |
+
"learning_rate": 9.50039377086147e-05,
|
| 401 |
+
"loss": 0.0168,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.46116504854368934,
|
| 406 |
+
"grad_norm": 0.39507967233657837,
|
| 407 |
+
"learning_rate": 9.476105444595534e-05,
|
| 408 |
+
"loss": 0.0157,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.4692556634304207,
|
| 413 |
+
"grad_norm": 0.34573355317115784,
|
| 414 |
+
"learning_rate": 9.451273234763371e-05,
|
| 415 |
+
"loss": 0.0176,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.4773462783171521,
|
| 420 |
+
"grad_norm": 0.2983342111110687,
|
| 421 |
+
"learning_rate": 9.425900158682385e-05,
|
| 422 |
+
"loss": 0.0177,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.4854368932038835,
|
| 427 |
+
"grad_norm": 0.38746461272239685,
|
| 428 |
+
"learning_rate": 9.399989299389661e-05,
|
| 429 |
+
"loss": 0.0216,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.4935275080906149,
|
| 434 |
+
"grad_norm": 0.39340198040008545,
|
| 435 |
+
"learning_rate": 9.373543805267368e-05,
|
| 436 |
+
"loss": 0.0221,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5016181229773463,
|
| 441 |
+
"grad_norm": 0.47980770468711853,
|
| 442 |
+
"learning_rate": 9.346566889660193e-05,
|
| 443 |
+
"loss": 0.0172,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5097087378640777,
|
| 448 |
+
"grad_norm": 0.421115905046463,
|
| 449 |
+
"learning_rate": 9.319061830484898e-05,
|
| 450 |
+
"loss": 0.0156,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.517799352750809,
|
| 455 |
+
"grad_norm": 0.3385259509086609,
|
| 456 |
+
"learning_rate": 9.291031969832026e-05,
|
| 457 |
+
"loss": 0.0176,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5258899676375405,
|
| 462 |
+
"grad_norm": 0.32277145981788635,
|
| 463 |
+
"learning_rate": 9.262480713559808e-05,
|
| 464 |
+
"loss": 0.0169,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5339805825242718,
|
| 469 |
+
"grad_norm": 0.32181084156036377,
|
| 470 |
+
"learning_rate": 9.233411530880326e-05,
|
| 471 |
+
"loss": 0.0187,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5420711974110033,
|
| 476 |
+
"grad_norm": 0.5838663578033447,
|
| 477 |
+
"learning_rate": 9.20382795393797e-05,
|
| 478 |
+
"loss": 0.0228,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.5501618122977346,
|
| 483 |
+
"grad_norm": 0.30314013361930847,
|
| 484 |
+
"learning_rate": 9.173733577380258e-05,
|
| 485 |
+
"loss": 0.02,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.558252427184466,
|
| 490 |
+
"grad_norm": 0.5964832901954651,
|
| 491 |
+
"learning_rate": 9.143132057921058e-05,
|
| 492 |
+
"loss": 0.018,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.5663430420711975,
|
| 497 |
+
"grad_norm": 0.4126530587673187,
|
| 498 |
+
"learning_rate": 9.112027113896262e-05,
|
| 499 |
+
"loss": 0.0173,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5744336569579288,
|
| 504 |
+
"grad_norm": 0.4253070652484894,
|
| 505 |
+
"learning_rate": 9.080422524811982e-05,
|
| 506 |
+
"loss": 0.0257,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.5825242718446602,
|
| 511 |
+
"grad_norm": 0.5417248606681824,
|
| 512 |
+
"learning_rate": 9.048322130885305e-05,
|
| 513 |
+
"loss": 0.0177,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.5906148867313916,
|
| 518 |
+
"grad_norm": 0.3459491431713104,
|
| 519 |
+
"learning_rate": 9.015729832577681e-05,
|
| 520 |
+
"loss": 0.019,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.598705501618123,
|
| 525 |
+
"grad_norm": 0.3335317373275757,
|
| 526 |
+
"learning_rate": 8.982649590120982e-05,
|
| 527 |
+
"loss": 0.0169,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.6067961165048543,
|
| 532 |
+
"grad_norm": 0.6572862267494202,
|
| 533 |
+
"learning_rate": 8.949085423036296e-05,
|
| 534 |
+
"loss": 0.0198,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6148867313915858,
|
| 539 |
+
"grad_norm": 0.540212869644165,
|
| 540 |
+
"learning_rate": 8.91504140964553e-05,
|
| 541 |
+
"loss": 0.016,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6229773462783171,
|
| 546 |
+
"grad_norm": 0.24550016224384308,
|
| 547 |
+
"learning_rate": 8.880521686575857e-05,
|
| 548 |
+
"loss": 0.0168,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6310679611650486,
|
| 553 |
+
"grad_norm": 0.5790821313858032,
|
| 554 |
+
"learning_rate": 8.845530448257085e-05,
|
| 555 |
+
"loss": 0.0184,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.63915857605178,
|
| 560 |
+
"grad_norm": 0.6583966612815857,
|
| 561 |
+
"learning_rate": 8.810071946411989e-05,
|
| 562 |
+
"loss": 0.0167,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6472491909385113,
|
| 567 |
+
"grad_norm": 0.3252386450767517,
|
| 568 |
+
"learning_rate": 8.774150489539707e-05,
|
| 569 |
+
"loss": 0.0173,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.6553398058252428,
|
| 574 |
+
"grad_norm": 0.37483498454093933,
|
| 575 |
+
"learning_rate": 8.737770442392212e-05,
|
| 576 |
+
"loss": 0.0169,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6634304207119741,
|
| 581 |
+
"grad_norm": 0.29752904176712036,
|
| 582 |
+
"learning_rate": 8.700936225443959e-05,
|
| 583 |
+
"loss": 0.018,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6715210355987055,
|
| 588 |
+
"grad_norm": 0.37370291352272034,
|
| 589 |
+
"learning_rate": 8.663652314354765e-05,
|
| 590 |
+
"loss": 0.0178,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.6796116504854369,
|
| 595 |
+
"grad_norm": 0.24446973204612732,
|
| 596 |
+
"learning_rate": 8.625923239425978e-05,
|
| 597 |
+
"loss": 0.0132,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.6877022653721683,
|
| 602 |
+
"grad_norm": 0.45145106315612793,
|
| 603 |
+
"learning_rate": 8.587753585050004e-05,
|
| 604 |
+
"loss": 0.0196,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.6957928802588996,
|
| 609 |
+
"grad_norm": 0.24976196885108948,
|
| 610 |
+
"learning_rate": 8.549147989153276e-05,
|
| 611 |
+
"loss": 0.017,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.7038834951456311,
|
| 616 |
+
"grad_norm": 0.41941019892692566,
|
| 617 |
+
"learning_rate": 8.510111142632698e-05,
|
| 618 |
+
"loss": 0.0153,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.7119741100323624,
|
| 623 |
+
"grad_norm": 0.3423904776573181,
|
| 624 |
+
"learning_rate": 8.470647788785665e-05,
|
| 625 |
+
"loss": 0.0143,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.7200647249190939,
|
| 630 |
+
"grad_norm": 0.28540822863578796,
|
| 631 |
+
"learning_rate": 8.430762722733714e-05,
|
| 632 |
+
"loss": 0.0125,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.7281553398058253,
|
| 637 |
+
"grad_norm": 0.38557255268096924,
|
| 638 |
+
"learning_rate": 8.390460790839882e-05,
|
| 639 |
+
"loss": 0.0193,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.7362459546925566,
|
| 644 |
+
"grad_norm": 0.496142715215683,
|
| 645 |
+
"learning_rate": 8.349746890119826e-05,
|
| 646 |
+
"loss": 0.0137,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.7443365695792881,
|
| 651 |
+
"grad_norm": 0.42488303780555725,
|
| 652 |
+
"learning_rate": 8.308625967646795e-05,
|
| 653 |
+
"loss": 0.0159,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7524271844660194,
|
| 658 |
+
"grad_norm": 0.28551360964775085,
|
| 659 |
+
"learning_rate": 8.267103019950529e-05,
|
| 660 |
+
"loss": 0.0155,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.7605177993527508,
|
| 665 |
+
"grad_norm": 0.3401723802089691,
|
| 666 |
+
"learning_rate": 8.225183092410128e-05,
|
| 667 |
+
"loss": 0.0133,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.7686084142394822,
|
| 672 |
+
"grad_norm": 0.34012413024902344,
|
| 673 |
+
"learning_rate": 8.182871278641009e-05,
|
| 674 |
+
"loss": 0.0191,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.7766990291262136,
|
| 679 |
+
"grad_norm": 0.385560005903244,
|
| 680 |
+
"learning_rate": 8.140172719875979e-05,
|
| 681 |
+
"loss": 0.0108,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.7847896440129449,
|
| 686 |
+
"grad_norm": 0.3427627980709076,
|
| 687 |
+
"learning_rate": 8.097092604340542e-05,
|
| 688 |
+
"loss": 0.0121,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.7928802588996764,
|
| 693 |
+
"grad_norm": 0.45653820037841797,
|
| 694 |
+
"learning_rate": 8.053636166622476e-05,
|
| 695 |
+
"loss": 0.0154,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.8009708737864077,
|
| 700 |
+
"grad_norm": 0.3446105122566223,
|
| 701 |
+
"learning_rate": 8.009808687035798e-05,
|
| 702 |
+
"loss": 0.0171,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.8090614886731392,
|
| 707 |
+
"grad_norm": 0.25365254282951355,
|
| 708 |
+
"learning_rate": 7.965615490979163e-05,
|
| 709 |
+
"loss": 0.014,
|
| 710 |
+
"step": 1000
|
| 711 |
+
}
|
| 712 |
+
],
|
| 713 |
+
"logging_steps": 10,
|
| 714 |
+
"max_steps": 3000,
|
| 715 |
+
"num_input_tokens_seen": 0,
|
| 716 |
+
"num_train_epochs": 3,
|
| 717 |
+
"save_steps": 1000,
|
| 718 |
+
"stateful_callbacks": {
|
| 719 |
+
"TrainerControl": {
|
| 720 |
+
"args": {
|
| 721 |
+
"should_epoch_stop": false,
|
| 722 |
+
"should_evaluate": false,
|
| 723 |
+
"should_log": false,
|
| 724 |
+
"should_save": true,
|
| 725 |
+
"should_training_stop": false
|
| 726 |
+
},
|
| 727 |
+
"attributes": {}
|
| 728 |
+
}
|
| 729 |
+
},
|
| 730 |
+
"total_flos": 0.0,
|
| 731 |
+
"train_batch_size": 32,
|
| 732 |
+
"trial_name": null,
|
| 733 |
+
"trial_params": null
|
| 734 |
+
}
|