Upload folder using huggingface_hub
Browse files- checkpoints-v3.1-discrete-conditional/checkpoint-8100/model.safetensors +3 -0
- checkpoints-v3.1-discrete-conditional/checkpoint-8100/optimizer.pt +3 -0
- checkpoints-v3.1-discrete-conditional/checkpoint-8100/rng_state.pth +3 -0
- checkpoints-v3.1-discrete-conditional/checkpoint-8100/scaler.pt +3 -0
- checkpoints-v3.1-discrete-conditional/checkpoint-8100/scheduler.pt +3 -0
- checkpoints-v3.1-discrete-conditional/checkpoint-8100/trainer_state.json +608 -0
- checkpoints-v3.1-discrete-conditional/checkpoint-8100/training_args.bin +3 -0
checkpoints-v3.1-discrete-conditional/checkpoint-8100/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4f929eca48afa794bafbeaef23a320af534c8bd061a65e5a9b8896df03a3c5c
|
| 3 |
+
size 45167832
|
checkpoints-v3.1-discrete-conditional/checkpoint-8100/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:62a3a3733cea6ec737cf4650c8449c674aca4e6cc41ecfd193e560170bb25edf
|
| 3 |
+
size 42312267
|
checkpoints-v3.1-discrete-conditional/checkpoint-8100/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:758cfef58bad9a16b6c7d1ae5d4792a78dc4416e9b987eae48f1f5a05ec94134
|
| 3 |
+
size 14645
|
checkpoints-v3.1-discrete-conditional/checkpoint-8100/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a9c9b3ad31e2e671747caa2299d893bff504371cef2d248efc43d4062fae720e
|
| 3 |
+
size 1383
|
checkpoints-v3.1-discrete-conditional/checkpoint-8100/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:71760c942926f7fa3025971e1a39a433eb072d72b2a0429a47879487d0239308
|
| 3 |
+
size 1465
|
checkpoints-v3.1-discrete-conditional/checkpoint-8100/trainer_state.json
ADDED
|
@@ -0,0 +1,608 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 4.0,
|
| 6 |
+
"eval_steps": 1024,
|
| 7 |
+
"global_step": 8100,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.06320987654320988,
|
| 14 |
+
"grad_norm": 0.6517145037651062,
|
| 15 |
+
"learning_rate": 0.000248046875,
|
| 16 |
+
"loss": 0.17245662212371826,
|
| 17 |
+
"step": 128
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.12641975308641976,
|
| 21 |
+
"grad_norm": 0.2118282914161682,
|
| 22 |
+
"learning_rate": 0.000498046875,
|
| 23 |
+
"loss": 0.15388090908527374,
|
| 24 |
+
"step": 256
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.18962962962962962,
|
| 28 |
+
"grad_norm": 0.23596829175949097,
|
| 29 |
+
"learning_rate": 0.000748046875,
|
| 30 |
+
"loss": 0.1529599130153656,
|
| 31 |
+
"step": 384
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.2528395061728395,
|
| 35 |
+
"grad_norm": 0.16465483605861664,
|
| 36 |
+
"learning_rate": 0.000998046875,
|
| 37 |
+
"loss": 0.15048745274543762,
|
| 38 |
+
"step": 512
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.3160493827160494,
|
| 42 |
+
"grad_norm": 0.09756451845169067,
|
| 43 |
+
"learning_rate": 0.0009993089770195807,
|
| 44 |
+
"loss": 0.1506166309118271,
|
| 45 |
+
"step": 640
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.37925925925925924,
|
| 49 |
+
"grad_norm": 0.2256879061460495,
|
| 50 |
+
"learning_rate": 0.0009972160460972733,
|
| 51 |
+
"loss": 0.15044106543064117,
|
| 52 |
+
"step": 768
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.44246913580246916,
|
| 56 |
+
"grad_norm": 0.11761642247438431,
|
| 57 |
+
"learning_rate": 0.0009937270408736224,
|
| 58 |
+
"loss": 0.1501142978668213,
|
| 59 |
+
"step": 896
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.505679012345679,
|
| 63 |
+
"grad_norm": 0.0816347524523735,
|
| 64 |
+
"learning_rate": 0.0009888517577149526,
|
| 65 |
+
"loss": 0.15066105127334595,
|
| 66 |
+
"step": 1024
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.505679012345679,
|
| 70 |
+
"eval_bleu": 0.0,
|
| 71 |
+
"eval_cos_loss": 0.03857117618729428,
|
| 72 |
+
"eval_loss": 0.148995438000051,
|
| 73 |
+
"eval_mse_loss": 0.148995438000051,
|
| 74 |
+
"step": 1024
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.505679012345679,
|
| 78 |
+
"eval_bleu": 0.0,
|
| 79 |
+
"eval_cos_loss": 0.03857117618729428,
|
| 80 |
+
"eval_loss": 0.148995438000051,
|
| 81 |
+
"eval_mse_loss": 0.148995438000051,
|
| 82 |
+
"eval_runtime": 17.631,
|
| 83 |
+
"eval_samples_per_second": 148.488,
|
| 84 |
+
"eval_steps_per_second": 2.325,
|
| 85 |
+
"step": 1024
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.5688888888888889,
|
| 89 |
+
"grad_norm": 0.08801733702421188,
|
| 90 |
+
"learning_rate": 0.0009826038853539248,
|
| 91 |
+
"loss": 0.1511014848947525,
|
| 92 |
+
"step": 1152
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.6320987654320988,
|
| 96 |
+
"grad_norm": 0.07067783921957016,
|
| 97 |
+
"learning_rate": 0.0009750009664545572,
|
| 98 |
+
"loss": 0.1478763073682785,
|
| 99 |
+
"step": 1280
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.6953086419753086,
|
| 103 |
+
"grad_norm": 0.08224646747112274,
|
| 104 |
+
"learning_rate": 0.0009660643483562486,
|
| 105 |
+
"loss": 0.15107029676437378,
|
| 106 |
+
"step": 1408
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.7585185185185185,
|
| 110 |
+
"grad_norm": 0.03443925082683563,
|
| 111 |
+
"learning_rate": 0.0009558191231351013,
|
| 112 |
+
"loss": 0.14998815953731537,
|
| 113 |
+
"step": 1536
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.8217283950617283,
|
| 117 |
+
"grad_norm": 0.06813549995422363,
|
| 118 |
+
"learning_rate": 0.0009442940571508399,
|
| 119 |
+
"loss": 0.14969073235988617,
|
| 120 |
+
"step": 1664
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.8849382716049383,
|
| 124 |
+
"grad_norm": 0.059174299240112305,
|
| 125 |
+
"learning_rate": 0.0009315215102771411,
|
| 126 |
+
"loss": 0.14814843237400055,
|
| 127 |
+
"step": 1792
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.9481481481481482,
|
| 131 |
+
"grad_norm": 0.09348400682210922,
|
| 132 |
+
"learning_rate": 0.0009175373450421618,
|
| 133 |
+
"loss": 0.15098626911640167,
|
| 134 |
+
"step": 1920
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.011358024691358,
|
| 138 |
+
"grad_norm": 0.03807132691144943,
|
| 139 |
+
"learning_rate": 0.0009023808259343743,
|
| 140 |
+
"loss": 0.14861960709095,
|
| 141 |
+
"step": 2048
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.011358024691358,
|
| 145 |
+
"eval_bleu": 0.0,
|
| 146 |
+
"eval_cos_loss": 0.03768022817264243,
|
| 147 |
+
"eval_loss": 0.14497914565045658,
|
| 148 |
+
"eval_mse_loss": 0.14497914565045658,
|
| 149 |
+
"step": 2048
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.011358024691358,
|
| 153 |
+
"eval_bleu": 0.0,
|
| 154 |
+
"eval_cos_loss": 0.03768022817264243,
|
| 155 |
+
"eval_loss": 0.14497914565045658,
|
| 156 |
+
"eval_mse_loss": 0.14497914565045658,
|
| 157 |
+
"eval_runtime": 17.3913,
|
| 158 |
+
"eval_samples_per_second": 150.535,
|
| 159 |
+
"eval_steps_per_second": 2.357,
|
| 160 |
+
"step": 2048
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.074567901234568,
|
| 164 |
+
"grad_norm": 0.047278061509132385,
|
| 165 |
+
"learning_rate": 0.0008860945091564399,
|
| 166 |
+
"loss": 0.14657916128635406,
|
| 167 |
+
"step": 2176
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 1.1377777777777778,
|
| 171 |
+
"grad_norm": 0.04802122339606285,
|
| 172 |
+
"learning_rate": 0.0008687241231366662,
|
| 173 |
+
"loss": 0.1490355134010315,
|
| 174 |
+
"step": 2304
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.2009876543209876,
|
| 178 |
+
"grad_norm": 0.047573771327733994,
|
| 179 |
+
"learning_rate": 0.0008503184401335448,
|
| 180 |
+
"loss": 0.15079016983509064,
|
| 181 |
+
"step": 2432
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.2641975308641975,
|
| 185 |
+
"grad_norm": 0.051725756376981735,
|
| 186 |
+
"learning_rate": 0.0008309291392938795,
|
| 187 |
+
"loss": 0.15022365748882294,
|
| 188 |
+
"step": 2560
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.3274074074074074,
|
| 192 |
+
"grad_norm": 0.027799006551504135,
|
| 193 |
+
"learning_rate": 0.0008106106615490032,
|
| 194 |
+
"loss": 0.1516639143228531,
|
| 195 |
+
"step": 2688
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.3906172839506172,
|
| 199 |
+
"grad_norm": 0.07822522521018982,
|
| 200 |
+
"learning_rate": 0.0007894200567565075,
|
| 201 |
+
"loss": 0.15127751231193542,
|
| 202 |
+
"step": 2816
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 1.453827160493827,
|
| 206 |
+
"grad_norm": 0.050239723175764084,
|
| 207 |
+
"learning_rate": 0.0007674168235166747,
|
| 208 |
+
"loss": 0.14863863587379456,
|
| 209 |
+
"step": 2944
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.5170370370370372,
|
| 213 |
+
"grad_norm": 0.07048199325799942,
|
| 214 |
+
"learning_rate": 0.000744662742113375,
|
| 215 |
+
"loss": 0.14983540773391724,
|
| 216 |
+
"step": 3072
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.5170370370370372,
|
| 220 |
+
"eval_bleu": 0.0,
|
| 221 |
+
"eval_cos_loss": 0.03912174238300905,
|
| 222 |
+
"eval_loss": 0.151152161926758,
|
| 223 |
+
"eval_mse_loss": 0.151152161926758,
|
| 224 |
+
"step": 3072
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.5170370370370372,
|
| 228 |
+
"eval_bleu": 0.0,
|
| 229 |
+
"eval_cos_loss": 0.03912174238300905,
|
| 230 |
+
"eval_loss": 0.151152161926758,
|
| 231 |
+
"eval_mse_loss": 0.151152161926758,
|
| 232 |
+
"eval_runtime": 17.2135,
|
| 233 |
+
"eval_samples_per_second": 152.09,
|
| 234 |
+
"eval_steps_per_second": 2.382,
|
| 235 |
+
"step": 3072
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 1.5802469135802468,
|
| 239 |
+
"grad_norm": 0.03784182295203209,
|
| 240 |
+
"learning_rate": 0.0007212217010484917,
|
| 241 |
+
"loss": 0.1514694094657898,
|
| 242 |
+
"step": 3200
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 1.643456790123457,
|
| 246 |
+
"grad_norm": 0.02600058726966381,
|
| 247 |
+
"learning_rate": 0.0006971595176569332,
|
| 248 |
+
"loss": 0.14975027740001678,
|
| 249 |
+
"step": 3328
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.7066666666666666,
|
| 253 |
+
"grad_norm": 0.0323154516518116,
|
| 254 |
+
"learning_rate": 0.0006725437533058971,
|
| 255 |
+
"loss": 0.1495063155889511,
|
| 256 |
+
"step": 3456
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.7698765432098766,
|
| 260 |
+
"grad_norm": 0.032654162496328354,
|
| 261 |
+
"learning_rate": 0.0006474435236972767,
|
| 262 |
+
"loss": 0.15044330060482025,
|
| 263 |
+
"step": 3584
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.8330864197530863,
|
| 267 |
+
"grad_norm": 0.054954059422016144,
|
| 268 |
+
"learning_rate": 0.0006219293048058301,
|
| 269 |
+
"loss": 0.14830181002616882,
|
| 270 |
+
"step": 3712
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.8962962962962964,
|
| 274 |
+
"grad_norm": 0.03366395831108093,
|
| 275 |
+
"learning_rate": 0.0005960727349980042,
|
| 276 |
+
"loss": 0.14830124378204346,
|
| 277 |
+
"step": 3840
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 1.959506172839506,
|
| 281 |
+
"grad_norm": 0.0262282807379961,
|
| 282 |
+
"learning_rate": 0.0005699464138870106,
|
| 283 |
+
"loss": 0.14904989302158356,
|
| 284 |
+
"step": 3968
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.022716049382716,
|
| 288 |
+
"grad_norm": 0.029652466997504234,
|
| 289 |
+
"learning_rate": 0.0005436236984889356,
|
| 290 |
+
"loss": 0.1490490734577179,
|
| 291 |
+
"step": 4096
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.022716049382716,
|
| 295 |
+
"eval_bleu": 0.0,
|
| 296 |
+
"eval_cos_loss": 0.03822143793832965,
|
| 297 |
+
"eval_loss": 0.14761792813859334,
|
| 298 |
+
"eval_mse_loss": 0.14761792813859334,
|
| 299 |
+
"step": 4096
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.022716049382716,
|
| 303 |
+
"eval_bleu": 0.0,
|
| 304 |
+
"eval_cos_loss": 0.03822143793832965,
|
| 305 |
+
"eval_loss": 0.14761792813859334,
|
| 306 |
+
"eval_mse_loss": 0.14761792813859334,
|
| 307 |
+
"eval_runtime": 17.5759,
|
| 308 |
+
"eval_samples_per_second": 148.954,
|
| 309 |
+
"eval_steps_per_second": 2.333,
|
| 310 |
+
"step": 4096
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 2.0859259259259257,
|
| 314 |
+
"grad_norm": 0.02397269941866398,
|
| 315 |
+
"learning_rate": 0.0005171784972522235,
|
| 316 |
+
"loss": 0.15046730637550354,
|
| 317 |
+
"step": 4224
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 2.149135802469136,
|
| 321 |
+
"grad_norm": 0.03038075752556324,
|
| 322 |
+
"learning_rate": 0.0004906850625388614,
|
| 323 |
+
"loss": 0.1505787968635559,
|
| 324 |
+
"step": 4352
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 2.212345679012346,
|
| 328 |
+
"grad_norm": 0.030504263937473297,
|
| 329 |
+
"learning_rate": 0.0004642177821399269,
|
| 330 |
+
"loss": 0.14973199367523193,
|
| 331 |
+
"step": 4480
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 2.2755555555555556,
|
| 335 |
+
"grad_norm": 0.02406897582113743,
|
| 336 |
+
"learning_rate": 0.00043785097041088293,
|
| 337 |
+
"loss": 0.1480320692062378,
|
| 338 |
+
"step": 4608
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.338765432098765,
|
| 342 |
+
"grad_norm": 0.025209980085492134,
|
| 343 |
+
"learning_rate": 0.00041165865961306135,
|
| 344 |
+
"loss": 0.15078303217887878,
|
| 345 |
+
"step": 4736
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.4019753086419753,
|
| 349 |
+
"grad_norm": 0.021273162215948105,
|
| 350 |
+
"learning_rate": 0.000385714392047208,
|
| 351 |
+
"loss": 0.14956723153591156,
|
| 352 |
+
"step": 4864
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 2.4651851851851854,
|
| 356 |
+
"grad_norm": 0.019383525475859642,
|
| 357 |
+
"learning_rate": 0.00036009101356272814,
|
| 358 |
+
"loss": 0.1489405333995819,
|
| 359 |
+
"step": 4992
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 2.528395061728395,
|
| 363 |
+
"grad_norm": 0.020899692550301552,
|
| 364 |
+
"learning_rate": 0.00033486046902241664,
|
| 365 |
+
"loss": 0.14915253221988678,
|
| 366 |
+
"step": 5120
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 2.528395061728395,
|
| 370 |
+
"eval_bleu": 0.0,
|
| 371 |
+
"eval_cos_loss": 0.03704260857548655,
|
| 372 |
+
"eval_loss": 0.14219033936174905,
|
| 373 |
+
"eval_mse_loss": 0.14219033936174905,
|
| 374 |
+
"step": 5120
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.528395061728395,
|
| 378 |
+
"eval_bleu": 0.0,
|
| 379 |
+
"eval_cos_loss": 0.03704260857548655,
|
| 380 |
+
"eval_loss": 0.14219033936174905,
|
| 381 |
+
"eval_mse_loss": 0.14219033936174905,
|
| 382 |
+
"eval_runtime": 17.4466,
|
| 383 |
+
"eval_samples_per_second": 150.058,
|
| 384 |
+
"eval_steps_per_second": 2.35,
|
| 385 |
+
"step": 5120
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 2.591604938271605,
|
| 389 |
+
"grad_norm": 0.02908194251358509,
|
| 390 |
+
"learning_rate": 0.00031009360029696107,
|
| 391 |
+
"loss": 0.14911262691020966,
|
| 392 |
+
"step": 5248
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 2.6548148148148147,
|
| 396 |
+
"grad_norm": 0.06822603195905685,
|
| 397 |
+
"learning_rate": 0.00028585994735640823,
|
| 398 |
+
"loss": 0.1485205888748169,
|
| 399 |
+
"step": 5376
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.718024691358025,
|
| 403 |
+
"grad_norm": 0.02113441936671734,
|
| 404 |
+
"learning_rate": 0.0002622275530170825,
|
| 405 |
+
"loss": 0.14886020123958588,
|
| 406 |
+
"step": 5504
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.7812345679012345,
|
| 410 |
+
"grad_norm": 0.026887331157922745,
|
| 411 |
+
"learning_rate": 0.0002392627718921852,
|
| 412 |
+
"loss": 0.14821897447109222,
|
| 413 |
+
"step": 5632
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 2.8444444444444446,
|
| 417 |
+
"grad_norm": 0.029895195737481117,
|
| 418 |
+
"learning_rate": 0.00021703008408250187,
|
| 419 |
+
"loss": 0.14855998754501343,
|
| 420 |
+
"step": 5760
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 2.907654320987654,
|
| 424 |
+
"grad_norm": 0.017637664452195168,
|
| 425 |
+
"learning_rate": 0.00019559191413033017,
|
| 426 |
+
"loss": 0.14899146556854248,
|
| 427 |
+
"step": 5888
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 2.9708641975308643,
|
| 431 |
+
"grad_norm": 0.035336632281541824,
|
| 432 |
+
"learning_rate": 0.00017500845574497032,
|
| 433 |
+
"loss": 0.14816507697105408,
|
| 434 |
+
"step": 6016
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 3.034074074074074,
|
| 438 |
+
"grad_norm": 0.019196584820747375,
|
| 439 |
+
"learning_rate": 0.00015533750279190617,
|
| 440 |
+
"loss": 0.14883184432983398,
|
| 441 |
+
"step": 6144
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 3.034074074074074,
|
| 445 |
+
"eval_bleu": 0.0,
|
| 446 |
+
"eval_cos_loss": 0.038591444583200823,
|
| 447 |
+
"eval_loss": 0.1482949287789624,
|
| 448 |
+
"eval_mse_loss": 0.1482949287789624,
|
| 449 |
+
"step": 6144
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 3.034074074074074,
|
| 453 |
+
"eval_bleu": 0.0,
|
| 454 |
+
"eval_cos_loss": 0.038591444583200823,
|
| 455 |
+
"eval_loss": 0.1482949287789624,
|
| 456 |
+
"eval_mse_loss": 0.1482949287789624,
|
| 457 |
+
"eval_runtime": 17.2605,
|
| 458 |
+
"eval_samples_per_second": 151.676,
|
| 459 |
+
"eval_steps_per_second": 2.375,
|
| 460 |
+
"step": 6144
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 3.097283950617284,
|
| 464 |
+
"grad_norm": 0.02671050652861595,
|
| 465 |
+
"learning_rate": 0.000136634287020226,
|
| 466 |
+
"loss": 0.14740729331970215,
|
| 467 |
+
"step": 6272
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 3.1604938271604937,
|
| 471 |
+
"grad_norm": 0.026964925229549408,
|
| 472 |
+
"learning_rate": 0.00011895132298390743,
|
| 473 |
+
"loss": 0.14907075464725494,
|
| 474 |
+
"step": 6400
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 3.2237037037037037,
|
| 478 |
+
"grad_norm": 0.03113582730293274,
|
| 479 |
+
"learning_rate": 0.00010233826059239426,
|
| 480 |
+
"loss": 0.15129905939102173,
|
| 481 |
+
"step": 6528
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 3.286913580246914,
|
| 485 |
+
"grad_norm": 0.016162540763616562,
|
| 486 |
+
"learning_rate": 8.68417457044704e-05,
|
| 487 |
+
"loss": 0.1497078388929367,
|
| 488 |
+
"step": 6656
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 3.3501234567901235,
|
| 492 |
+
"grad_norm": 0.03715677186846733,
|
| 493 |
+
"learning_rate": 7.250528915685422e-05,
|
| 494 |
+
"loss": 0.15141023695468903,
|
| 495 |
+
"step": 6784
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 3.413333333333333,
|
| 499 |
+
"grad_norm": 0.023826338350772858,
|
| 500 |
+
"learning_rate": 5.93691445952505e-05,
|
| 501 |
+
"loss": 0.14596687257289886,
|
| 502 |
+
"step": 6912
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 3.476543209876543,
|
| 506 |
+
"grad_norm": 0.015810539945960045,
|
| 507 |
+
"learning_rate": 4.7470195450886376e-05,
|
| 508 |
+
"loss": 0.1492277979850769,
|
| 509 |
+
"step": 7040
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 3.5397530864197533,
|
| 513 |
+
"grad_norm": 0.01591351442039013,
|
| 514 |
+
"learning_rate": 3.684185137987378e-05,
|
| 515 |
+
"loss": 0.1490740180015564,
|
| 516 |
+
"step": 7168
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 3.5397530864197533,
|
| 520 |
+
"eval_bleu": 0.0,
|
| 521 |
+
"eval_cos_loss": 0.038417009682190126,
|
| 522 |
+
"eval_loss": 0.14850174581132283,
|
| 523 |
+
"eval_mse_loss": 0.14850174581132283,
|
| 524 |
+
"step": 7168
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 3.5397530864197533,
|
| 528 |
+
"eval_bleu": 0.0,
|
| 529 |
+
"eval_cos_loss": 0.038417009682190126,
|
| 530 |
+
"eval_loss": 0.14850174581132283,
|
| 531 |
+
"eval_mse_loss": 0.14850174581132283,
|
| 532 |
+
"eval_runtime": 17.2573,
|
| 533 |
+
"eval_samples_per_second": 151.704,
|
| 534 |
+
"eval_steps_per_second": 2.376,
|
| 535 |
+
"step": 7168
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 3.602962962962963,
|
| 539 |
+
"grad_norm": 0.01680172048509121,
|
| 540 |
+
"learning_rate": 2.751395445617594e-05,
|
| 541 |
+
"loss": 0.1489180326461792,
|
| 542 |
+
"step": 7296
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 3.6661728395061726,
|
| 546 |
+
"grad_norm": 0.018111728131771088,
|
| 547 |
+
"learning_rate": 1.9512695381567302e-05,
|
| 548 |
+
"loss": 0.1487143635749817,
|
| 549 |
+
"step": 7424
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 3.7293827160493827,
|
| 553 |
+
"grad_norm": 0.03240951895713806,
|
| 554 |
+
"learning_rate": 1.2860539947850303e-05,
|
| 555 |
+
"loss": 0.1489432007074356,
|
| 556 |
+
"step": 7552
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 3.7925925925925927,
|
| 560 |
+
"grad_norm": 0.025667285546660423,
|
| 561 |
+
"learning_rate": 7.5761659578078565e-06,
|
| 562 |
+
"loss": 0.15040472149848938,
|
| 563 |
+
"step": 7680
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 3.8558024691358024,
|
| 567 |
+
"grad_norm": 0.01822933368384838,
|
| 568 |
+
"learning_rate": 3.674410782003812e-06,
|
| 569 |
+
"loss": 0.15010671317577362,
|
| 570 |
+
"step": 7808
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 3.9190123456790125,
|
| 574 |
+
"grad_norm": 0.023645251989364624,
|
| 575 |
+
"learning_rate": 1.1662296986795217e-06,
|
| 576 |
+
"loss": 0.1507875919342041,
|
| 577 |
+
"step": 7936
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 3.982222222222222,
|
| 581 |
+
"grad_norm": 0.02925770729780197,
|
| 582 |
+
"learning_rate": 5.866513372004834e-08,
|
| 583 |
+
"loss": 0.1489885449409485,
|
| 584 |
+
"step": 8064
|
| 585 |
+
}
|
| 586 |
+
],
|
| 587 |
+
"logging_steps": 128,
|
| 588 |
+
"max_steps": 8100,
|
| 589 |
+
"num_input_tokens_seen": 0,
|
| 590 |
+
"num_train_epochs": 4,
|
| 591 |
+
"save_steps": 1024,
|
| 592 |
+
"stateful_callbacks": {
|
| 593 |
+
"TrainerControl": {
|
| 594 |
+
"args": {
|
| 595 |
+
"should_epoch_stop": false,
|
| 596 |
+
"should_evaluate": false,
|
| 597 |
+
"should_log": false,
|
| 598 |
+
"should_save": true,
|
| 599 |
+
"should_training_stop": true
|
| 600 |
+
},
|
| 601 |
+
"attributes": {}
|
| 602 |
+
}
|
| 603 |
+
},
|
| 604 |
+
"total_flos": 0.0,
|
| 605 |
+
"train_batch_size": 64,
|
| 606 |
+
"trial_name": null,
|
| 607 |
+
"trial_params": null
|
| 608 |
+
}
|
checkpoints-v3.1-discrete-conditional/checkpoint-8100/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c53f3e10d2a055b85b1814f3e029cdcd51dd4f5e4025edaf0233172e301d1199
|
| 3 |
+
size 5137
|