Upload folder using huggingface_hub

.gitattributes

@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 checkpoints/checkpoint-4096/eval_state.json filter=lfs diff=lfs merge=lfs -text
 checkpoints-v2/checkpoint-12288/eval_state.json filter=lfs diff=lfs merge=lfs -text
 checkpoints-v2.1/checkpoint-12288/eval_state.json filter=lfs diff=lfs merge=lfs -text
+checkpoints-v2.2/checkpoint-12288/eval_state.json filter=lfs diff=lfs merge=lfs -text

checkpoints-v2.2/checkpoint-12288/eval_state.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a990443ff2a1f54dc414faf7f8caf49966af81ea98b54e4729de2796678e1b36
+size 44107577

checkpoints-v2.2/checkpoint-12288/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3831102455a0aa490d03bb1a5e22754e39d11655e69e72ca5af25165a72fb1e
+size 37402680

checkpoints-v2.2/checkpoint-12288/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:427bdc660526ed8a7598c66d9616f4a94948d0eeb9ea6e29fad3f1e3016312c0
+size 512267

checkpoints-v2.2/checkpoint-12288/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f36e6c28b77555b6da6de84681647b558ac8ebc553a1b458e45112e416a213c
+size 14645

checkpoints-v2.2/checkpoint-12288/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba52f34cff25d619c91ac6fb0309246ee28b809c6a39aff62d988a3e27f482dc
+size 1383

checkpoints-v2.2/checkpoint-12288/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:220b4c233a19883d884d3d8ad9c7134f99ecd7f2b9282b526b3c0fd015c1a12c
+size 1465

checkpoints-v2.2/checkpoint-12288/trainer_state.json

@@ -0,0 +1,790 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.5675488430095608,
+  "eval_steps": 1024,
+  "global_step": 12288,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.011823934229365849,
+      "grad_norm": 1.1521458625793457,
+      "learning_rate": 1.9615384615384617e-05,
+      "loss": 10.405,
+      "step": 256
+    },
+    {
+      "epoch": 0.023647868458731697,
+      "grad_norm": 1.1140727996826172,
+      "learning_rate": 3.930769230769231e-05,
+      "loss": 7.9214,
+      "step": 512
+    },
+    {
+      "epoch": 0.03547180268809755,
+      "grad_norm": 1.0781198740005493,
+      "learning_rate": 4.999617095521894e-05,
+      "loss": 5.638,
+      "step": 768
+    },
+    {
+      "epoch": 0.047295736917463395,
+      "grad_norm": 0.5991400480270386,
+      "learning_rate": 4.9961092368776736e-05,
+      "loss": 3.7771,
+      "step": 1024
+    },
+    {
+      "epoch": 0.047295736917463395,
+      "eval_acr_loss": 0.9959428622025878,
+      "eval_across_var": 0.0020306417274663065,
+      "eval_bleu": 0.5706335938305095,
+      "eval_ce_loss": 2.3132477236665,
+      "eval_cos_loss": 0.923946771447517,
+      "eval_cov": 0.06470938904644692,
+      "eval_cov_loss": 0.006561864716039129,
+      "eval_global_var": 0.2042425843678653,
+      "eval_loss": 2.7861235038330565,
+      "eval_mse_loss": 1.908825221399194,
+      "eval_per_var": 0.1985390023009418,
+      "eval_within_var": 0.20226115012141668,
+      "step": 1024
+    },
+    {
+      "epoch": 0.047295736917463395,
+      "eval_acr_loss": 0.9959428622025878,
+      "eval_across_var": 0.0020306417274663065,
+      "eval_bleu": 0.5706335938305095,
+      "eval_ce_loss": 2.3132477236665,
+      "eval_cos_loss": 0.923946771447517,
+      "eval_cov": 0.06470938904644692,
+      "eval_cov_loss": 0.006561864716039129,
+      "eval_global_var": 0.2042425843678653,
+      "eval_loss": 2.7861235038330565,
+      "eval_mse_loss": 1.908825221399194,
+      "eval_per_var": 0.1985390023009418,
+      "eval_runtime": 157.6333,
+      "eval_samples_per_second": 177.583,
+      "eval_steps_per_second": 2.779,
+      "eval_within_var": 0.20226115012141668,
+      "step": 1024
+    },
+    {
+      "epoch": 0.05911967114682925,
+      "grad_norm": 0.394672155380249,
+      "learning_rate": 4.988941132556799e-05,
+      "loss": 2.7053,
+      "step": 1280
+    },
+    {
+      "epoch": 0.0709436053761951,
+      "grad_norm": 0.3323874771595001,
+      "learning_rate": 4.9781232937269974e-05,
+      "loss": 2.1017,
+      "step": 1536
+    },
+    {
+      "epoch": 0.08276753960556095,
+      "grad_norm": 0.2772473096847534,
+      "learning_rate": 4.963671583455164e-05,
+      "loss": 1.7095,
+      "step": 1792
+    },
+    {
+      "epoch": 0.09459147383492679,
+      "grad_norm": 0.2364146113395691,
+      "learning_rate": 4.945607193446079e-05,
+      "loss": 1.4374,
+      "step": 2048
+    },
+    {
+      "epoch": 0.09459147383492679,
+      "eval_acr_loss": 0.9956181184069751,
+      "eval_across_var": 0.0021933579932234873,
+      "eval_bleu": 0.813955966195176,
+      "eval_ce_loss": 0.6843064660078859,
+      "eval_cos_loss": 0.7549032925470779,
+      "eval_cov": 0.06552444527682648,
+      "eval_cov_loss": 0.006829353782973469,
+      "eval_global_var": 0.21431199165239725,
+      "eval_loss": 1.0921703268131708,
+      "eval_mse_loss": 1.6229622002061643,
+      "eval_per_var": 0.208273970373145,
+      "eval_within_var": 0.21221458272302532,
+      "step": 2048
+    },
+    {
+      "epoch": 0.09459147383492679,
+      "eval_acr_loss": 0.9956181184069751,
+      "eval_across_var": 0.0021933579932234873,
+      "eval_bleu": 0.813955966195176,
+      "eval_ce_loss": 0.6843064660078859,
+      "eval_cos_loss": 0.7549032925470779,
+      "eval_cov": 0.06552444527682648,
+      "eval_cov_loss": 0.006829353782973469,
+      "eval_global_var": 0.21431199165239725,
+      "eval_loss": 1.0921703268131708,
+      "eval_mse_loss": 1.6229622002061643,
+      "eval_per_var": 0.208273970373145,
+      "eval_runtime": 153.0523,
+      "eval_samples_per_second": 182.898,
+      "eval_steps_per_second": 2.862,
+      "eval_within_var": 0.21221458272302532,
+      "step": 2048
+    },
+    {
+      "epoch": 0.10641540806429264,
+      "grad_norm": 0.1965663582086563,
+      "learning_rate": 4.923956612967301e-05,
+      "loss": 1.2396,
+      "step": 2304
+    },
+    {
+      "epoch": 0.1182393422936585,
+      "grad_norm": 0.1851101964712143,
+      "learning_rate": 4.898751590005826e-05,
+      "loss": 1.083,
+      "step": 2560
+    },
+    {
+      "epoch": 0.13006327652302435,
+      "grad_norm": 0.15827688574790955,
+      "learning_rate": 4.870029084713462e-05,
+      "loss": 0.961,
+      "step": 2816
+    },
+    {
+      "epoch": 0.1418872107523902,
+      "grad_norm": 0.139593243598938,
+      "learning_rate": 4.837831215209188e-05,
+      "loss": 0.864,
+      "step": 3072
+    },
+    {
+      "epoch": 0.1418872107523902,
+      "eval_acr_loss": 0.9949602079173746,
+      "eval_across_var": 0.002523095093956685,
+      "eval_bleu": 0.8997310988628063,
+      "eval_ce_loss": 0.32039221284324176,
+      "eval_cos_loss": 0.5973711480560913,
+      "eval_cov": 0.0646399929098887,
+      "eval_cov_loss": 0.006647485838862791,
+      "eval_global_var": 0.234542498305508,
+      "eval_loss": 0.6663434101294164,
+      "eval_mse_loss": 1.3315815294169944,
+      "eval_per_var": 0.2278395003924087,
+      "eval_within_var": 0.23215524597118978,
+      "step": 3072
+    },
+    {
+      "epoch": 0.1418872107523902,
+      "eval_acr_loss": 0.9949602079173746,
+      "eval_across_var": 0.002523095093956685,
+      "eval_bleu": 0.8997310988628063,
+      "eval_ce_loss": 0.32039221284324176,
+      "eval_cos_loss": 0.5973711480560913,
+      "eval_cov": 0.0646399929098887,
+      "eval_cov_loss": 0.006647485838862791,
+      "eval_global_var": 0.234542498305508,
+      "eval_loss": 0.6663434101294164,
+      "eval_mse_loss": 1.3315815294169944,
+      "eval_per_var": 0.2278395003924087,
+      "eval_runtime": 154.029,
+      "eval_samples_per_second": 181.739,
+      "eval_steps_per_second": 2.844,
+      "eval_within_var": 0.23215524597118978,
+      "step": 3072
+    },
+    {
+      "epoch": 0.15371114498175603,
+      "grad_norm": 0.12921613454818726,
+      "learning_rate": 4.802205195817963e-05,
+      "loss": 0.7861,
+      "step": 3328
+    },
+    {
+      "epoch": 0.1655350792111219,
+      "grad_norm": 0.13848277926445007,
+      "learning_rate": 4.763203267836576e-05,
+      "loss": 0.7193,
+      "step": 3584
+    },
+    {
+      "epoch": 0.17735901344048774,
+      "grad_norm": 0.10973386466503143,
+      "learning_rate": 4.720882622928019e-05,
+      "loss": 0.6643,
+      "step": 3840
+    },
+    {
+      "epoch": 0.18918294766985358,
+      "grad_norm": 0.1159495934844017,
+      "learning_rate": 4.675305319256765e-05,
+      "loss": 0.6185,
+      "step": 4096
+    },
+    {
+      "epoch": 0.18918294766985358,
+      "eval_acr_loss": 0.9942766772285444,
+      "eval_across_var": 0.002865787035228523,
+      "eval_bleu": 0.9364990379271309,
+      "eval_ce_loss": 0.18651799913750935,
+      "eval_cos_loss": 0.47863238975065486,
+      "eval_cov": 0.0638771231315996,
+      "eval_cov_loss": 0.006466193114020372,
+      "eval_global_var": 0.2541172254031107,
+      "eval_loss": 0.4854983908536772,
+      "eval_mse_loss": 1.1063276224484726,
+      "eval_per_var": 0.246796351045234,
+      "eval_within_var": 0.2514293484015552,
+      "step": 4096
+    },
+    {
+      "epoch": 0.18918294766985358,
+      "eval_acr_loss": 0.9942766772285444,
+      "eval_across_var": 0.002865787035228523,
+      "eval_bleu": 0.9364990379271309,
+      "eval_ce_loss": 0.18651799913750935,
+      "eval_cos_loss": 0.47863238975065486,
+      "eval_cov": 0.0638771231315996,
+      "eval_cov_loss": 0.006466193114020372,
+      "eval_global_var": 0.2541172254031107,
+      "eval_loss": 0.4854983908536772,
+      "eval_mse_loss": 1.1063276224484726,
+      "eval_per_var": 0.246796351045234,
+      "eval_runtime": 154.3189,
+      "eval_samples_per_second": 181.397,
+      "eval_steps_per_second": 2.838,
+      "eval_within_var": 0.2514293484015552,
+      "step": 4096
+    },
+    {
+      "epoch": 0.20100688189921945,
+      "grad_norm": 0.1160832941532135,
+      "learning_rate": 4.6265381904878854e-05,
+      "loss": 0.577,
+      "step": 4352
+    },
+    {
+      "epoch": 0.2128308161285853,
+      "grad_norm": 0.10667941719293594,
+      "learning_rate": 4.57465274778347e-05,
+      "loss": 0.5457,
+      "step": 4608
+    },
+    {
+      "epoch": 0.22465475035795113,
+      "grad_norm": 0.10070338845252991,
+      "learning_rate": 4.519725074940068e-05,
+      "loss": 0.5125,
+      "step": 4864
+    },
+    {
+      "epoch": 0.236478684587317,
+      "grad_norm": 0.09314938634634018,
+      "learning_rate": 4.461835716820895e-05,
+      "loss": 0.4861,
+      "step": 5120
+    },
+    {
+      "epoch": 0.236478684587317,
+      "eval_acr_loss": 0.9935455889734504,
+      "eval_across_var": 0.0032324549948294823,
+      "eval_bleu": 0.9564306066317199,
+      "eval_ce_loss": 0.12215962886946387,
+      "eval_cos_loss": 0.39296473127249715,
+      "eval_cov": 0.06288307224778825,
+      "eval_cov_loss": 0.006272464265545073,
+      "eval_global_var": 0.2724754298658676,
+      "eval_loss": 0.3871788881957259,
+      "eval_mse_loss": 0.9429605110595215,
+      "eval_per_var": 0.264601616010274,
+      "eval_within_var": 0.26944677286768615,
+      "step": 5120
+    },
+    {
+      "epoch": 0.236478684587317,
+      "eval_acr_loss": 0.9935455889734504,
+      "eval_across_var": 0.0032324549948294823,
+      "eval_bleu": 0.9564306066317199,
+      "eval_ce_loss": 0.12215962886946387,
+      "eval_cos_loss": 0.39296473127249715,
+      "eval_cov": 0.06288307224778825,
+      "eval_cov_loss": 0.006272464265545073,
+      "eval_global_var": 0.2724754298658676,
+      "eval_loss": 0.3871788881957259,
+      "eval_mse_loss": 0.9429605110595215,
+      "eval_per_var": 0.264601616010274,
+      "eval_runtime": 153.1543,
+      "eval_samples_per_second": 182.776,
+      "eval_steps_per_second": 2.86,
+      "eval_within_var": 0.26944677286768615,
+      "step": 5120
+    },
+    {
+      "epoch": 0.24830261881668284,
+      "grad_norm": 0.09335759282112122,
+      "learning_rate": 4.401069561246422e-05,
+      "loss": 0.4627,
+      "step": 5376
+    },
+    {
+      "epoch": 0.2601265530460487,
+      "grad_norm": 0.0996316596865654,
+      "learning_rate": 4.337515714516545e-05,
+      "loss": 0.4415,
+      "step": 5632
+    },
+    {
+      "epoch": 0.27195048727541454,
+      "grad_norm": 0.09572074562311172,
+      "learning_rate": 4.2712673707468434e-05,
+      "loss": 0.4263,
+      "step": 5888
+    },
+    {
+      "epoch": 0.2837744215047804,
+      "grad_norm": 0.08873972296714783,
+      "learning_rate": 4.202421675210565e-05,
+      "loss": 0.4101,
+      "step": 6144
+    },
+    {
+      "epoch": 0.2837744215047804,
+      "eval_acr_loss": 0.9926441069607321,
+      "eval_across_var": 0.0036847662828122713,
+      "eval_bleu": 0.9683519037504735,
+      "eval_ce_loss": 0.08646239441517553,
+      "eval_cos_loss": 0.33377824519595056,
+      "eval_cov": 0.061909993489583336,
+      "eval_cov_loss": 0.006098649053107181,
+      "eval_global_var": 0.2904153066138699,
+      "eval_loss": 0.3280817231495086,
+      "eval_mse_loss": 0.8322314463521792,
+      "eval_per_var": 0.2820036030251142,
+      "eval_within_var": 0.28695739764873296,
+      "step": 6144
+    },
+    {
+      "epoch": 0.2837744215047804,
+      "eval_acr_loss": 0.9926441069607321,
+      "eval_across_var": 0.0036847662828122713,
+      "eval_bleu": 0.9683519037504735,
+      "eval_ce_loss": 0.08646239441517553,
+      "eval_cos_loss": 0.33377824519595056,
+      "eval_cov": 0.061909993489583336,
+      "eval_cov_loss": 0.006098649053107181,
+      "eval_global_var": 0.2904153066138699,
+      "eval_loss": 0.3280817231495086,
+      "eval_mse_loss": 0.8322314463521792,
+      "eval_per_var": 0.2820036030251142,
+      "eval_runtime": 154.0222,
+      "eval_samples_per_second": 181.747,
+      "eval_steps_per_second": 2.844,
+      "eval_within_var": 0.28695739764873296,
+      "step": 6144
+    },
+    {
+      "epoch": 0.2955983557341462,
+      "grad_norm": 0.08218205720186234,
+      "learning_rate": 4.131079581886694e-05,
+      "loss": 0.3928,
+      "step": 6400
+    },
+    {
+      "epoch": 0.30742228996351206,
+      "grad_norm": 0.08782156556844711,
+      "learning_rate": 4.057345705423016e-05,
+      "loss": 0.3806,
+      "step": 6656
+    },
+    {
+      "epoch": 0.3192462241928779,
+      "grad_norm": 0.10694174468517303,
+      "learning_rate": 3.981328167731251e-05,
+      "loss": 0.3701,
+      "step": 6912
+    },
+    {
+      "epoch": 0.3310701584222438,
+      "grad_norm": 0.08735407143831253,
+      "learning_rate": 3.9031384394391954e-05,
+      "loss": 0.3563,
+      "step": 7168
+    },
+    {
+      "epoch": 0.3310701584222438,
+      "eval_acr_loss": 0.9915075998872382,
+      "eval_across_var": 0.004255300866690842,
+      "eval_bleu": 0.9761821772549975,
+      "eval_ce_loss": 0.06435613293353826,
+      "eval_cos_loss": 0.2929684796698017,
+      "eval_cov": 0.06080683181274971,
+      "eval_cov_loss": 0.005920343130792929,
+      "eval_global_var": 0.3081673400042808,
+      "eval_loss": 0.28994010664420583,
+      "eval_mse_loss": 0.7590125883550949,
+      "eval_per_var": 0.2992506332049087,
+      "eval_within_var": 0.30415138697515337,
+      "step": 7168
+    },
+    {
+      "epoch": 0.3310701584222438,
+      "eval_acr_loss": 0.9915075998872382,
+      "eval_across_var": 0.004255300866690842,
+      "eval_bleu": 0.9761821772549975,
+      "eval_ce_loss": 0.06435613293353826,
+      "eval_cos_loss": 0.2929684796698017,
+      "eval_cov": 0.06080683181274971,
+      "eval_cov_loss": 0.005920343130792929,
+      "eval_global_var": 0.3081673400042808,
+      "eval_loss": 0.28994010664420583,
+      "eval_mse_loss": 0.7590125883550949,
+      "eval_per_var": 0.2992506332049087,
+      "eval_runtime": 153.3148,
+      "eval_samples_per_second": 182.585,
+      "eval_steps_per_second": 2.857,
+      "eval_within_var": 0.30415138697515337,
+      "step": 7168
+    },
+    {
+      "epoch": 0.34289409265160964,
+      "grad_norm": 0.08888935297727585,
+      "learning_rate": 3.822891176432382e-05,
+      "loss": 0.3488,
+      "step": 7424
+    },
+    {
+      "epoch": 0.3547180268809755,
+      "grad_norm": 0.08499140292406082,
+      "learning_rate": 3.7407040517249335e-05,
+      "loss": 0.3398,
+      "step": 7680
+    },
+    {
+      "epoch": 0.3665419611103413,
+      "grad_norm": 0.1081831082701683,
+      "learning_rate": 3.6566975829061614e-05,
+      "loss": 0.3306,
+      "step": 7936
+    },
+    {
+      "epoch": 0.37836589533970716,
+      "grad_norm": 0.06914108246564865,
+      "learning_rate": 3.5709949554159355e-05,
+      "loss": 0.322,
+      "step": 8192
+    },
+    {
+      "epoch": 0.37836589533970716,
+      "eval_acr_loss": 0.9897522895031323,
+      "eval_across_var": 0.005137139498617619,
+      "eval_bleu": 0.9811255688400959,
+      "eval_ce_loss": 0.05007254747784437,
+      "eval_cos_loss": 0.26478874057395274,
+      "eval_cov": 0.0596730828829552,
+      "eval_cov_loss": 0.005715872895729528,
+      "eval_global_var": 0.3259957370148402,
+      "eval_loss": 0.26465164740744246,
+      "eval_mse_loss": 0.7119130991637435,
+      "eval_per_var": 0.3166442592394406,
+      "eval_within_var": 0.32112443848559846,
+      "step": 8192
+    },
+    {
+      "epoch": 0.37836589533970716,
+      "eval_acr_loss": 0.9897522895031323,
+      "eval_across_var": 0.005137139498617619,
+      "eval_bleu": 0.9811255688400959,
+      "eval_ce_loss": 0.05007254747784437,
+      "eval_cos_loss": 0.26478874057395274,
+      "eval_cov": 0.0596730828829552,
+      "eval_cov_loss": 0.005715872895729528,
+      "eval_global_var": 0.3259957370148402,
+      "eval_loss": 0.26465164740744246,
+      "eval_mse_loss": 0.7119130991637435,
+      "eval_per_var": 0.3166442592394406,
+      "eval_runtime": 150.8387,
+      "eval_samples_per_second": 185.582,
+      "eval_steps_per_second": 2.904,
+      "eval_within_var": 0.32112443848559846,
+      "step": 8192
+    },
+    {
+      "epoch": 0.390189829569073,
+      "grad_norm": 0.111445851624012,
+      "learning_rate": 3.483721841907964e-05,
+      "loss": 0.3166,
+      "step": 8448
+    },
+    {
+      "epoch": 0.4020137637984389,
+      "grad_norm": 0.08934515714645386,
+      "learning_rate": 3.395006217965885e-05,
+      "loss": 0.3105,
+      "step": 8704
+    },
+    {
+      "epoch": 0.41383769802780473,
+      "grad_norm": 0.08079824596643448,
+      "learning_rate": 3.3049781744423665e-05,
+      "loss": 0.3033,
+      "step": 8960
+    },
+    {
+      "epoch": 0.4256616322571706,
+      "grad_norm": 0.11221586167812347,
+      "learning_rate": 3.213769726696439e-05,
+      "loss": 0.2987,
+      "step": 9216
+    },
+    {
+      "epoch": 0.4256616322571706,
+      "eval_acr_loss": 0.984604048810593,
+      "eval_across_var": 0.007728356739387052,
+      "eval_bleu": 0.9847880526042291,
+      "eval_ce_loss": 0.04037455940984836,
+      "eval_cos_loss": 0.24513382110829768,
+      "eval_cov": 0.05903792707887415,
+      "eval_cov_loss": 0.005642149438996497,
+      "eval_global_var": 0.3496879682148973,
+      "eval_loss": 0.2470546718941976,
+      "eval_mse_loss": 0.6823068254342363,
+      "eval_per_var": 0.3401547784674658,
+      "eval_within_var": 0.3422601510128474,
+      "step": 9216
+    },
+    {
+      "epoch": 0.4256616322571706,
+      "eval_acr_loss": 0.984604048810593,
+      "eval_across_var": 0.007728356739387052,
+      "eval_bleu": 0.9847880526042291,
+      "eval_ce_loss": 0.04037455940984836,
+      "eval_cos_loss": 0.24513382110829768,
+      "eval_cov": 0.05903792707887415,
+      "eval_cov_loss": 0.005642149438996497,
+      "eval_global_var": 0.3496879682148973,
+      "eval_loss": 0.2470546718941976,
+      "eval_mse_loss": 0.6823068254342363,
+      "eval_per_var": 0.3401547784674658,
+      "eval_runtime": 150.8886,
+      "eval_samples_per_second": 185.521,
+      "eval_steps_per_second": 2.903,
+      "eval_within_var": 0.3422601510128474,
+      "step": 9216
+    },
+    {
+      "epoch": 0.4374855664865364,
+      "grad_norm": 0.07710844278335571,
+      "learning_rate": 3.121514621008757e-05,
+      "loss": 0.2939,
+      "step": 9472
+    },
+    {
+      "epoch": 0.44930950071590225,
+      "grad_norm": 0.08533202856779099,
+      "learning_rate": 3.0283481384586697e-05,
+      "loss": 0.2906,
+      "step": 9728
+    },
+    {
+      "epoch": 0.4611334349452681,
+      "grad_norm": 0.08804863691329956,
+      "learning_rate": 2.9344068965507027e-05,
+      "loss": 0.2852,
+      "step": 9984
+    },
+    {
+      "epoch": 0.472957369174634,
+      "grad_norm": 0.123167484998703,
+      "learning_rate": 2.839828648881323e-05,
+      "loss": 0.2808,
+      "step": 10240
+    },
+    {
+      "epoch": 0.472957369174634,
+      "eval_acr_loss": 0.859320777328047,
+      "eval_across_var": 0.07312867154269458,
+      "eval_bleu": 0.9870718825944174,
+      "eval_ce_loss": 0.033869750563854766,
+      "eval_cos_loss": 0.2345634405607502,
+      "eval_cov": 0.06209365635702055,
+      "eval_cov_loss": 0.0064547402575850215,
+      "eval_global_var": 0.4857372823915525,
+      "eval_loss": 0.22445420723527534,
+      "eval_mse_loss": 0.672757336538132,
+      "eval_per_var": 0.4763423275185502,
+      "eval_within_var": 0.4130568303610092,
+      "step": 10240
+    },
+    {
+      "epoch": 0.472957369174634,
+      "eval_acr_loss": 0.859320777328047,
+      "eval_across_var": 0.07312867154269458,
+      "eval_bleu": 0.9870718825944174,
+      "eval_ce_loss": 0.033869750563854766,
+      "eval_cos_loss": 0.2345634405607502,
+      "eval_cov": 0.06209365635702055,
+      "eval_cov_loss": 0.0064547402575850215,
+      "eval_global_var": 0.4857372823915525,
+      "eval_loss": 0.22445420723527534,
+      "eval_mse_loss": 0.672757336538132,
+      "eval_per_var": 0.4763423275185502,
+      "eval_runtime": 150.0476,
+      "eval_samples_per_second": 186.561,
+      "eval_steps_per_second": 2.919,
+      "eval_within_var": 0.4130568303610092,
+      "step": 10240
+    },
+    {
+      "epoch": 0.48478130340399983,
+      "grad_norm": 0.1995747834444046,
+      "learning_rate": 2.745124265175868e-05,
+      "loss": 0.2507,
+      "step": 10496
+    },
+    {
+      "epoch": 0.49660523763336567,
+      "grad_norm": 0.1398428976535797,
+      "learning_rate": 2.6496899297412598e-05,
+      "loss": 0.2036,
+      "step": 10752
+    },
+    {
+      "epoch": 0.5084291718627315,
+      "grad_norm": 0.12668104469776154,
+      "learning_rate": 2.554036091926675e-05,
+      "loss": 0.1934,
+      "step": 11008
+    },
+    {
+      "epoch": 0.5202531060920974,
+      "grad_norm": 0.1443246304988861,
+      "learning_rate": 2.4583030166456618e-05,
+      "loss": 0.1865,
+      "step": 11264
+    },
+    {
+      "epoch": 0.5202531060920974,
+      "eval_acr_loss": 0.016154762003698978,
+      "eval_across_var": 0.9554836936465138,
+      "eval_bleu": 0.987837564898818,
+      "eval_ce_loss": 0.030638722400925203,
+      "eval_cos_loss": 0.2311469105552865,
+      "eval_cov": 0.05655882674265125,
+      "eval_cov_loss": 0.005161343601672617,
+      "eval_global_var": 1.6112946810787672,
+      "eval_loss": 0.13616000148191298,
+      "eval_mse_loss": 0.6809118339732357,
+      "eval_per_var": 1.6024877818207763,
+      "eval_within_var": 0.6615459767922963,
+      "step": 11264
+    },
+    {
+      "epoch": 0.5202531060920974,
+      "eval_acr_loss": 0.016154762003698978,
+      "eval_across_var": 0.9554836936465138,
+      "eval_bleu": 0.987837564898818,
+      "eval_ce_loss": 0.030638722400925203,
+      "eval_cos_loss": 0.2311469105552865,
+      "eval_cov": 0.05655882674265125,
+      "eval_cov_loss": 0.005161343601672617,
+      "eval_global_var": 1.6112946810787672,
+      "eval_loss": 0.13616000148191298,
+      "eval_mse_loss": 0.6809118339732357,
+      "eval_per_var": 1.6024877818207763,
+      "eval_runtime": 150.5401,
+      "eval_samples_per_second": 185.951,
+      "eval_steps_per_second": 2.91,
+      "eval_within_var": 0.6615459767922963,
+      "step": 11264
+    },
+    {
+      "epoch": 0.5320770403214632,
+      "grad_norm": 0.12779445946216583,
+      "learning_rate": 2.3626310850040373e-05,
+      "loss": 0.1819,
+      "step": 11520
+    },
+    {
+      "epoch": 0.5439009745508291,
+      "grad_norm": 0.10633435100317001,
+      "learning_rate": 2.2671605884477816e-05,
+      "loss": 0.179,
+      "step": 11776
+    },
+    {
+      "epoch": 0.5557249087801949,
+      "grad_norm": 0.1239749938249588,
+      "learning_rate": 2.1720315230424133e-05,
+      "loss": 0.176,
+      "step": 12032
+    },
+    {
+      "epoch": 0.5675488430095608,
+      "grad_norm": 0.10707388818264008,
+      "learning_rate": 2.0773833841855016e-05,
+      "loss": 0.1743,
+      "step": 12288
+    },
+    {
+      "epoch": 0.5675488430095608,
+      "eval_acr_loss": 0.014688911844754918,
+      "eval_across_var": 0.978259073953106,
+      "eval_bleu": 0.9892687211321183,
+      "eval_ce_loss": 0.026860382261188472,
+      "eval_cos_loss": 0.2202416457337876,
+      "eval_cov": 0.05523096372003425,
+      "eval_cov_loss": 0.0049222552655562414,
+      "eval_global_var": 1.72990243293379,
+      "eval_loss": 0.1281860415487801,
+      "eval_mse_loss": 0.6658408976580998,
+      "eval_per_var": 1.7439577982305936,
+      "eval_within_var": 0.7586165843488963,
+      "step": 12288
+    },
+    {
+      "epoch": 0.5675488430095608,
+      "eval_acr_loss": 0.014688911844754918,
+      "eval_across_var": 0.978259073953106,
+      "eval_bleu": 0.9892687211321183,
+      "eval_ce_loss": 0.026860382261188472,
+      "eval_cos_loss": 0.2202416457337876,
+      "eval_cov": 0.05523096372003425,
+      "eval_cov_loss": 0.0049222552655562414,
+      "eval_global_var": 1.72990243293379,
+      "eval_loss": 0.1281860415487801,
+      "eval_mse_loss": 0.6658408976580998,
+      "eval_per_var": 1.7439577982305936,
+      "eval_runtime": 150.3829,
+      "eval_samples_per_second": 186.145,
+      "eval_steps_per_second": 2.913,
+      "eval_within_var": 0.7586165843488963,
+      "step": 12288
+    }
+  ],
+  "logging_steps": 256,
+  "max_steps": 21651,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 1024,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints-v2.2/checkpoint-12288/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0d0d75419ee5260692019faa2af9fbcf8d1b0a175b6254de59fe8e48427c029
+size 5777