Upload folder using huggingface_hub

.gitattributes

@@ -35,3 +35,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *tfevents* 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-v2.1/checkpoint-12288/eval_state.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05b3bca630c7d08caf62d59aa41ee15d1aad0e5b971befe95d42644b1526bea9
+size 44108941

checkpoints-v2.1/checkpoint-12288/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f679c8e0b39bf014779cac799fd1c51930d771c35519b8824a50844c0bdcb78
+size 37402680

checkpoints-v2.1/checkpoint-12288/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6adc558ecff7a93c46301e5c792542680f0ae1cf6478199b736271df1782ce20
+size 512267

checkpoints-v2.1/checkpoint-12288/rng_state.pth

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

checkpoints-v2.1/checkpoint-12288/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:904fc891c4da67fa7d4b7db7eba41d622dec830fc89cf51c20e856ab78c9fc76
+size 1383

checkpoints-v2.1/checkpoint-12288/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87b2b9d7efb9af569cd6999393703c76b185862a0221b931604de9e5bf2b79a6
+size 1465

checkpoints-v2.1/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": 1.1338618993759155,
+      "learning_rate": 1.9615384615384617e-05,
+      "loss": 10.4459,
+      "step": 256
+    },
+    {
+      "epoch": 0.023647868458731697,
+      "grad_norm": 1.0851895809173584,
+      "learning_rate": 3.930769230769231e-05,
+      "loss": 7.9458,
+      "step": 512
+    },
+    {
+      "epoch": 0.03547180268809755,
+      "grad_norm": 0.9216171503067017,
+      "learning_rate": 4.999617095521894e-05,
+      "loss": 5.6401,
+      "step": 768
+    },
+    {
+      "epoch": 0.047295736917463395,
+      "grad_norm": 0.5787180066108704,
+      "learning_rate": 4.9961092368776736e-05,
+      "loss": 3.8256,
+      "step": 1024
+    },
+    {
+      "epoch": 0.047295736917463395,
+      "eval_acr_loss": 0.9939580324305791,
+      "eval_across_var": 0.003025606172010473,
+      "eval_bleu": 0.5515564027779047,
+      "eval_ce_loss": 2.3591716888288383,
+      "eval_cos_loss": 0.9272929947125857,
+      "eval_cov": 0.0706253400131992,
+      "eval_cov_loss": 0.00802828647306861,
+      "eval_global_var": 0.2767718589469178,
+      "eval_loss": 2.832596059803549,
+      "eval_mse_loss": 1.9087850182023767,
+      "eval_per_var": 0.2680246013484589,
+      "eval_within_var": 0.27379363795665845,
+      "step": 1024
+    },
+    {
+      "epoch": 0.047295736917463395,
+      "eval_acr_loss": 0.9939580324305791,
+      "eval_across_var": 0.003025606172010473,
+      "eval_bleu": 0.5515564027779047,
+      "eval_ce_loss": 2.3591716888288383,
+      "eval_cos_loss": 0.9272929947125857,
+      "eval_cov": 0.0706253400131992,
+      "eval_cov_loss": 0.00802828647306861,
+      "eval_global_var": 0.2767718589469178,
+      "eval_loss": 2.832596059803549,
+      "eval_mse_loss": 1.9087850182023767,
+      "eval_per_var": 0.2680246013484589,
+      "eval_runtime": 159.2542,
+      "eval_samples_per_second": 175.776,
+      "eval_steps_per_second": 2.75,
+      "eval_within_var": 0.27379363795665845,
+      "step": 1024
+    },
+    {
+      "epoch": 0.05911967114682925,
+      "grad_norm": 0.38924261927604675,
+      "learning_rate": 4.988941132556799e-05,
+      "loss": 2.7681,
+      "step": 1280
+    },
+    {
+      "epoch": 0.0709436053761951,
+      "grad_norm": 0.3134535551071167,
+      "learning_rate": 4.9781232937269974e-05,
+      "loss": 2.1522,
+      "step": 1536
+    },
+    {
+      "epoch": 0.08276753960556095,
+      "grad_norm": 0.2510242760181427,
+      "learning_rate": 4.963671583455164e-05,
+      "loss": 1.7487,
+      "step": 1792
+    },
+    {
+      "epoch": 0.09459147383492679,
+      "grad_norm": 0.22504042088985443,
+      "learning_rate": 4.945607193446079e-05,
+      "loss": 1.4694,
+      "step": 2048
+    },
+    {
+      "epoch": 0.09459147383492679,
+      "eval_acr_loss": 0.9901781947645423,
+      "eval_across_var": 0.004923156680556261,
+      "eval_bleu": 0.8030880203505176,
+      "eval_ce_loss": 0.700762519825539,
+      "eval_cos_loss": 0.7673689819634233,
+      "eval_cov": 0.07080983897866723,
+      "eval_cov_loss": 0.008125514746749917,
+      "eval_global_var": 0.4061385202625571,
+      "eval_loss": 1.1126885499856243,
+      "eval_mse_loss": 1.6458245439616512,
+      "eval_per_var": 0.3933824513056507,
+      "eval_within_var": 0.4013502570592105,
+      "step": 2048
+    },
+    {
+      "epoch": 0.09459147383492679,
+      "eval_acr_loss": 0.9901781947645423,
+      "eval_across_var": 0.004923156680556261,
+      "eval_bleu": 0.8030880203505176,
+      "eval_ce_loss": 0.700762519825539,
+      "eval_cos_loss": 0.7673689819634233,
+      "eval_cov": 0.07080983897866723,
+      "eval_cov_loss": 0.008125514746749917,
+      "eval_global_var": 0.4061385202625571,
+      "eval_loss": 1.1126885499856243,
+      "eval_mse_loss": 1.6458245439616512,
+      "eval_per_var": 0.3933824513056507,
+      "eval_runtime": 155.1424,
+      "eval_samples_per_second": 180.434,
+      "eval_steps_per_second": 2.823,
+      "eval_within_var": 0.4013502570592105,
+      "step": 2048
+    },
+    {
+      "epoch": 0.10641540806429264,
+      "grad_norm": 0.18675386905670166,
+      "learning_rate": 4.923956612967301e-05,
+      "loss": 1.2664,
+      "step": 2304
+    },
+    {
+      "epoch": 0.1182393422936585,
+      "grad_norm": 0.18214967846870422,
+      "learning_rate": 4.898751590005826e-05,
+      "loss": 1.1058,
+      "step": 2560
+    },
+    {
+      "epoch": 0.13006327652302435,
+      "grad_norm": 0.15246237814426422,
+      "learning_rate": 4.870029084713462e-05,
+      "loss": 0.981,
+      "step": 2816
+    },
+    {
+      "epoch": 0.1418872107523902,
+      "grad_norm": 0.1368647962808609,
+      "learning_rate": 4.837831215209188e-05,
+      "loss": 0.8816,
+      "step": 3072
+    },
+    {
+      "epoch": 0.1418872107523902,
+      "eval_acr_loss": 0.9858010461613468,
+      "eval_across_var": 0.007125071276481089,
+      "eval_bleu": 0.8956566730262217,
+      "eval_ce_loss": 0.32717657133460587,
+      "eval_cos_loss": 0.6143105866974348,
+      "eval_cov": 0.06988288496182934,
+      "eval_cov_loss": 0.007913092302252032,
+      "eval_global_var": 0.5146751926369864,
+      "eval_loss": 0.6788079935938256,
+      "eval_mse_loss": 1.3672495941593223,
+      "eval_per_var": 0.4983278039383562,
+      "eval_within_var": 0.5077701699516001,
+      "step": 3072
+    },
+    {
+      "epoch": 0.1418872107523902,
+      "eval_acr_loss": 0.9858010461613468,
+      "eval_across_var": 0.007125071276481089,
+      "eval_bleu": 0.8956566730262217,
+      "eval_ce_loss": 0.32717657133460587,
+      "eval_cos_loss": 0.6143105866974348,
+      "eval_cov": 0.06988288496182934,
+      "eval_cov_loss": 0.007913092302252032,
+      "eval_global_var": 0.5146751926369864,
+      "eval_loss": 0.6788079935938256,
+      "eval_mse_loss": 1.3672495941593223,
+      "eval_per_var": 0.4983278039383562,
+      "eval_runtime": 156.011,
+      "eval_samples_per_second": 179.43,
+      "eval_steps_per_second": 2.807,
+      "eval_within_var": 0.5077701699516001,
+      "step": 3072
+    },
+    {
+      "epoch": 0.15371114498175603,
+      "grad_norm": 0.13020840287208557,
+      "learning_rate": 4.802205195817963e-05,
+      "loss": 0.8019,
+      "step": 3328
+    },
+    {
+      "epoch": 0.1655350792111219,
+      "grad_norm": 0.12300444394350052,
+      "learning_rate": 4.763203267836576e-05,
+      "loss": 0.7339,
+      "step": 3584
+    },
+    {
+      "epoch": 0.17735901344048774,
+      "grad_norm": 0.10956571996212006,
+      "learning_rate": 4.720882622928019e-05,
+      "loss": 0.6774,
+      "step": 3840
+    },
+    {
+      "epoch": 0.18918294766985358,
+      "grad_norm": 0.11182258278131485,
+      "learning_rate": 4.675305319256765e-05,
+      "loss": 0.6307,
+      "step": 4096
+    },
+    {
+      "epoch": 0.18918294766985358,
+      "eval_acr_loss": 0.9774447257660296,
+      "eval_across_var": 0.011342588644134536,
+      "eval_bleu": 0.9352497637682005,
+      "eval_ce_loss": 0.18979091641225226,
+      "eval_cos_loss": 0.49574217472446563,
+      "eval_cov": 0.06955429843571632,
+      "eval_cov_loss": 0.007831381105490403,
+      "eval_global_var": 0.6115789544092466,
+      "eval_loss": 0.49388179734145127,
+      "eval_mse_loss": 1.144643609110079,
+      "eval_per_var": 0.592599529109589,
+      "eval_within_var": 0.600530758568141,
+      "step": 4096
+    },
+    {
+      "epoch": 0.18918294766985358,
+      "eval_acr_loss": 0.9774447257660296,
+      "eval_across_var": 0.011342588644134536,
+      "eval_bleu": 0.9352497637682005,
+      "eval_ce_loss": 0.18979091641225226,
+      "eval_cos_loss": 0.49574217472446563,
+      "eval_cov": 0.06955429843571632,
+      "eval_cov_loss": 0.007831381105490403,
+      "eval_global_var": 0.6115789544092466,
+      "eval_loss": 0.49388179734145127,
+      "eval_mse_loss": 1.144643609110079,
+      "eval_per_var": 0.592599529109589,
+      "eval_runtime": 155.2844,
+      "eval_samples_per_second": 180.269,
+      "eval_steps_per_second": 2.821,
+      "eval_within_var": 0.600530758568141,
+      "step": 4096
+    },
+    {
+      "epoch": 0.20100688189921945,
+      "grad_norm": 0.1080719456076622,
+      "learning_rate": 4.6265381904878854e-05,
+      "loss": 0.588,
+      "step": 4352
+    },
+    {
+      "epoch": 0.2128308161285853,
+      "grad_norm": 0.10819243639707565,
+      "learning_rate": 4.57465274778347e-05,
+      "loss": 0.5554,
+      "step": 4608
+    },
+    {
+      "epoch": 0.22465475035795113,
+      "grad_norm": 0.1115206629037857,
+      "learning_rate": 4.519725074940068e-05,
+      "loss": 0.5198,
+      "step": 4864
+    },
+    {
+      "epoch": 0.236478684587317,
+      "grad_norm": 0.1552964597940445,
+      "learning_rate": 4.461835716820895e-05,
+      "loss": 0.473,
+      "step": 5120
+    },
+    {
+      "epoch": 0.236478684587317,
+      "eval_acr_loss": 0.11126784305088222,
+      "eval_across_var": 0.680014660641483,
+      "eval_bleu": 0.9523306149700821,
+      "eval_ce_loss": 0.12964293614165967,
+      "eval_cos_loss": 0.4240787292588247,
+      "eval_cov": 0.07667060747538527,
+      "eval_cov_loss": 0.010102802944969232,
+      "eval_global_var": 1.6775805329623288,
+      "eval_loss": 0.31928212995126365,
+      "eval_mse_loss": 1.0156728980475909,
+      "eval_per_var": 1.6460250784817352,
+      "eval_within_var": 1.0011268127454471,
+      "step": 5120
+    },
+    {
+      "epoch": 0.236478684587317,
+      "eval_acr_loss": 0.11126784305088222,
+      "eval_across_var": 0.680014660641483,
+      "eval_bleu": 0.9523306149700821,
+      "eval_ce_loss": 0.12964293614165967,
+      "eval_cos_loss": 0.4240787292588247,
+      "eval_cov": 0.07667060747538527,
+      "eval_cov_loss": 0.010102802944969232,
+      "eval_global_var": 1.6775805329623288,
+      "eval_loss": 0.31928212995126365,
+      "eval_mse_loss": 1.0156728980475909,
+      "eval_per_var": 1.6460250784817352,
+      "eval_runtime": 154.2182,
+      "eval_samples_per_second": 181.516,
+      "eval_steps_per_second": 2.84,
+      "eval_within_var": 1.0011268127454471,
+      "step": 5120
+    },
+    {
+      "epoch": 0.24830261881668284,
+      "grad_norm": 0.11613737791776657,
+      "learning_rate": 4.401069561246422e-05,
+      "loss": 0.3958,
+      "step": 5376
+    },
+    {
+      "epoch": 0.2601265530460487,
+      "grad_norm": 0.11101594567298889,
+      "learning_rate": 4.337515714516545e-05,
+      "loss": 0.3648,
+      "step": 5632
+    },
+    {
+      "epoch": 0.27195048727541454,
+      "grad_norm": 0.14844343066215515,
+      "learning_rate": 4.2712673707468434e-05,
+      "loss": 0.3464,
+      "step": 5888
+    },
+    {
+      "epoch": 0.2837744215047804,
+      "grad_norm": 0.10389428585767746,
+      "learning_rate": 4.202421675210565e-05,
+      "loss": 0.3281,
+      "step": 6144
+    },
+    {
+      "epoch": 0.2837744215047804,
+      "eval_acr_loss": 0.015462962337612025,
+      "eval_across_var": 0.9652468334866441,
+      "eval_bleu": 0.9665450842704094,
+      "eval_ce_loss": 0.08923274265882904,
+      "eval_cos_loss": 0.35504293285276245,
+      "eval_cov": 0.06710940844392123,
+      "eval_cov_loss": 0.007327720047650884,
+      "eval_global_var": 2.2822310216894977,
+      "eval_loss": 0.24187510450408883,
+      "eval_mse_loss": 0.8843358904803724,
+      "eval_per_var": 2.3488914454908674,
+      "eval_within_var": 1.3261088448572376,
+      "step": 6144
+    },
+    {
+      "epoch": 0.2837744215047804,
+      "eval_acr_loss": 0.015462962337612025,
+      "eval_across_var": 0.9652468334866441,
+      "eval_bleu": 0.9665450842704094,
+      "eval_ce_loss": 0.08923274265882904,
+      "eval_cos_loss": 0.35504293285276245,
+      "eval_cov": 0.06710940844392123,
+      "eval_cov_loss": 0.007327720047650884,
+      "eval_global_var": 2.2822310216894977,
+      "eval_loss": 0.24187510450408883,
+      "eval_mse_loss": 0.8843358904803724,
+      "eval_per_var": 2.3488914454908674,
+      "eval_runtime": 155.0579,
+      "eval_samples_per_second": 180.533,
+      "eval_steps_per_second": 2.825,
+      "eval_within_var": 1.3261088448572376,
+      "step": 6144
+    },
+    {
+      "epoch": 0.2955983557341462,
+      "grad_norm": 0.129238098859787,
+      "learning_rate": 4.131079581886694e-05,
+      "loss": 0.3099,
+      "step": 6400
+    },
+    {
+      "epoch": 0.30742228996351206,
+      "grad_norm": 0.1061507984995842,
+      "learning_rate": 4.057345705423016e-05,
+      "loss": 0.2963,
+      "step": 6656
+    },
+    {
+      "epoch": 0.3192462241928779,
+      "grad_norm": 0.10803277790546417,
+      "learning_rate": 3.981328167731251e-05,
+      "loss": 0.2854,
+      "step": 6912
+    },
+    {
+      "epoch": 0.3310701584222438,
+      "grad_norm": 0.10297808796167374,
+      "learning_rate": 3.9031384394391954e-05,
+      "loss": 0.2709,
+      "step": 7168
+    },
+    {
+      "epoch": 0.3310701584222438,
+      "eval_acr_loss": 0.014716924630300589,
+      "eval_across_var": 0.9652062489834006,
+      "eval_bleu": 0.9750850142716162,
+      "eval_ce_loss": 0.0659905160653945,
+      "eval_cos_loss": 0.31133458848413265,
+      "eval_cov": 0.06627595805686358,
+      "eval_cov_loss": 0.007171058822125537,
+      "eval_global_var": 2.3926940639269407,
+      "eval_loss": 0.20149783922793113,
+      "eval_mse_loss": 0.8055339337211765,
+      "eval_per_var": 2.513992936643836,
+      "eval_within_var": 1.4371973874906427,
+      "step": 7168
+    },
+    {
+      "epoch": 0.3310701584222438,
+      "eval_acr_loss": 0.014716924630300589,
+      "eval_across_var": 0.9652062489834006,
+      "eval_bleu": 0.9750850142716162,
+      "eval_ce_loss": 0.0659905160653945,
+      "eval_cos_loss": 0.31133458848413265,
+      "eval_cov": 0.06627595805686358,
+      "eval_cov_loss": 0.007171058822125537,
+      "eval_global_var": 2.3926940639269407,
+      "eval_loss": 0.20149783922793113,
+      "eval_mse_loss": 0.8055339337211765,
+      "eval_per_var": 2.513992936643836,
+      "eval_runtime": 152.9262,
+      "eval_samples_per_second": 183.049,
+      "eval_steps_per_second": 2.864,
+      "eval_within_var": 1.4371973874906427,
+      "step": 7168
+    },
+    {
+      "epoch": 0.34289409265160964,
+      "grad_norm": 0.08976765722036362,
+      "learning_rate": 3.822891176432382e-05,
+      "loss": 0.2629,
+      "step": 7424
+    },
+    {
+      "epoch": 0.3547180268809755,
+      "grad_norm": 0.25151142477989197,
+      "learning_rate": 3.7407040517249335e-05,
+      "loss": 0.2533,
+      "step": 7680
+    },
+    {
+      "epoch": 0.3665419611103413,
+      "grad_norm": 0.09347163140773773,
+      "learning_rate": 3.6566975829061614e-05,
+      "loss": 0.2437,
+      "step": 7936
+    },
+    {
+      "epoch": 0.37836589533970716,
+      "grad_norm": 0.15004394948482513,
+      "learning_rate": 3.5709949554159355e-05,
+      "loss": 0.2348,
+      "step": 8192
+    },
+    {
+      "epoch": 0.37836589533970716,
+      "eval_acr_loss": 0.014583685287022457,
+      "eval_across_var": 0.9538344581649728,
+      "eval_bleu": 0.980561100967318,
+      "eval_ce_loss": 0.05129089445454073,
+      "eval_cos_loss": 0.28137742255104187,
+      "eval_cov": 0.06562282614511987,
+      "eval_cov_loss": 0.007066378377369482,
+      "eval_global_var": 2.479759738869863,
+      "eval_loss": 0.17527394500225102,
+      "eval_mse_loss": 0.7551626324925793,
+      "eval_per_var": 2.6612933433219177,
+      "eval_within_var": 1.53611661745533,
+      "step": 8192
+    },
+    {
+      "epoch": 0.37836589533970716,
+      "eval_acr_loss": 0.014583685287022457,
+      "eval_across_var": 0.9538344581649728,
+      "eval_bleu": 0.980561100967318,
+      "eval_ce_loss": 0.05129089445454073,
+      "eval_cos_loss": 0.28137742255104187,
+      "eval_cov": 0.06562282614511987,
+      "eval_cov_loss": 0.007066378377369482,
+      "eval_global_var": 2.479759738869863,
+      "eval_loss": 0.17527394500225102,
+      "eval_mse_loss": 0.7551626324925793,
+      "eval_per_var": 2.6612933433219177,
+      "eval_runtime": 151.3473,
+      "eval_samples_per_second": 184.959,
+      "eval_steps_per_second": 2.894,
+      "eval_within_var": 1.53611661745533,
+      "step": 8192
+    },
+    {
+      "epoch": 0.390189829569073,
+      "grad_norm": 0.09314695745706558,
+      "learning_rate": 3.483721841907964e-05,
+      "loss": 0.2288,
+      "step": 8448
+    },
+    {
+      "epoch": 0.4020137637984389,
+      "grad_norm": 0.09459128230810165,
+      "learning_rate": 3.395006217965885e-05,
+      "loss": 0.2225,
+      "step": 8704
+    },
+    {
+      "epoch": 0.41383769802780473,
+      "grad_norm": 0.1038607731461525,
+      "learning_rate": 3.3049781744423665e-05,
+      "loss": 0.215,
+      "step": 8960
+    },
+    {
+      "epoch": 0.4256616322571706,
+      "grad_norm": 0.06964848935604095,
+      "learning_rate": 3.213769726696439e-05,
+      "loss": 0.2103,
+      "step": 9216
+    },
+    {
+      "epoch": 0.4256616322571706,
+      "eval_acr_loss": 0.01336662589730249,
+      "eval_across_var": 0.987674045780478,
+      "eval_bleu": 0.9840502134478613,
+      "eval_ce_loss": 0.0413797390200708,
+      "eval_cos_loss": 0.2597663049080056,
+      "eval_cov": 0.0652730584688927,
+      "eval_cov_loss": 0.007007736591494655,
+      "eval_global_var": 2.6066415168378994,
+      "eval_loss": 0.15707423451216254,
+      "eval_mse_loss": 0.7215510081482804,
+      "eval_per_var": 2.776688249143836,
+      "eval_within_var": 1.6285416399507218,
+      "step": 9216
+    },
+    {
+      "epoch": 0.4256616322571706,
+      "eval_acr_loss": 0.01336662589730249,
+      "eval_across_var": 0.987674045780478,
+      "eval_bleu": 0.9840502134478613,
+      "eval_ce_loss": 0.0413797390200708,
+      "eval_cos_loss": 0.2597663049080056,
+      "eval_cov": 0.0652730584688927,
+      "eval_cov_loss": 0.007007736591494655,
+      "eval_global_var": 2.6066415168378994,
+      "eval_loss": 0.15707423451216254,
+      "eval_mse_loss": 0.7215510081482804,
+      "eval_per_var": 2.776688249143836,
+      "eval_runtime": 151.5839,
+      "eval_samples_per_second": 184.67,
+      "eval_steps_per_second": 2.889,
+      "eval_within_var": 1.6285416399507218,
+      "step": 9216
+    },
+    {
+      "epoch": 0.4374855664865364,
+      "grad_norm": 0.07443105429410934,
+      "learning_rate": 3.121514621008757e-05,
+      "loss": 0.2053,
+      "step": 9472
+    },
+    {
+      "epoch": 0.44930950071590225,
+      "grad_norm": 0.08320944011211395,
+      "learning_rate": 3.0283481384586697e-05,
+      "loss": 0.2017,
+      "step": 9728
+    },
+    {
+      "epoch": 0.4611334349452681,
+      "grad_norm": 0.1169746071100235,
+      "learning_rate": 2.9344068965507027e-05,
+      "loss": 0.1966,
+      "step": 9984
+    },
+    {
+      "epoch": 0.472957369174634,
+      "grad_norm": 0.08953411877155304,
+      "learning_rate": 2.840199155190943e-05,
+      "loss": 0.1938,
+      "step": 10240
+    },
+    {
+      "epoch": 0.472957369174634,
+      "eval_acr_loss": 0.012974254141045301,
+      "eval_across_var": 0.9815338524781405,
+      "eval_bleu": 0.9866441773938769,
+      "eval_ce_loss": 0.034374653984745755,
+      "eval_cos_loss": 0.24375769958648508,
+      "eval_cov": 0.06485466107930223,
+      "eval_cov_loss": 0.006919686747374668,
+      "eval_global_var": 2.6645574700342465,
+      "eval_loss": 0.1440824061359989,
+      "eval_mse_loss": 0.6987405730981261,
+      "eval_per_var": 2.866460652111872,
+      "eval_within_var": 1.6929740287941886,
+      "step": 10240
+    },
+    {
+      "epoch": 0.472957369174634,
+      "eval_acr_loss": 0.012974254141045301,
+      "eval_across_var": 0.9815338524781405,
+      "eval_bleu": 0.9866441773938769,
+      "eval_ce_loss": 0.034374653984745755,
+      "eval_cos_loss": 0.24375769958648508,
+      "eval_cov": 0.06485466107930223,
+      "eval_cov_loss": 0.006919686747374668,
+      "eval_global_var": 2.6645574700342465,
+      "eval_loss": 0.1440824061359989,
+      "eval_mse_loss": 0.6987405730981261,
+      "eval_per_var": 2.866460652111872,
+      "eval_runtime": 151.0941,
+      "eval_samples_per_second": 185.269,
+      "eval_steps_per_second": 2.899,
+      "eval_within_var": 1.6929740287941886,
+      "step": 10240
+    },
+    {
+      "epoch": 0.48478130340399983,
+      "grad_norm": 0.08886408805847168,
+      "learning_rate": 2.745124265175868e-05,
+      "loss": 0.1892,
+      "step": 10496
+    },
+    {
+      "epoch": 0.49660523763336567,
+      "grad_norm": 0.11508477479219437,
+      "learning_rate": 2.6496899297412598e-05,
+      "loss": 0.1853,
+      "step": 10752
+    },
+    {
+      "epoch": 0.5084291718627315,
+      "grad_norm": 0.0753609761595726,
+      "learning_rate": 2.554036091926675e-05,
+      "loss": 0.1839,
+      "step": 11008
+    },
+    {
+      "epoch": 0.5202531060920974,
+      "grad_norm": 0.09823817759752274,
+      "learning_rate": 2.4583030166456618e-05,
+      "loss": 0.18,
+      "step": 11264
+    },
+    {
+      "epoch": 0.5202531060920974,
+      "eval_acr_loss": 0.013261525430046982,
+      "eval_across_var": 1.0205278622505327,
+      "eval_bleu": 0.9883516965111658,
+      "eval_ce_loss": 0.029720396500880316,
+      "eval_cos_loss": 0.23187202939840212,
+      "eval_cov": 0.06469566310377425,
+      "eval_cov_loss": 0.006903172251619569,
+      "eval_global_var": 2.7817003781392695,
+      "eval_loss": 0.13512653335248498,
+      "eval_mse_loss": 0.6834642570040542,
+      "eval_per_var": 3.0010434503424657,
+      "eval_within_var": 1.7715753001165173,
+      "step": 11264
+    },
+    {
+      "epoch": 0.5202531060920974,
+      "eval_acr_loss": 0.013261525430046982,
+      "eval_across_var": 1.0205278622505327,
+      "eval_bleu": 0.9883516965111658,
+      "eval_ce_loss": 0.029720396500880316,
+      "eval_cos_loss": 0.23187202939840212,
+      "eval_cov": 0.06469566310377425,
+      "eval_cov_loss": 0.006903172251619569,
+      "eval_global_var": 2.7817003781392695,
+      "eval_loss": 0.13512653335248498,
+      "eval_mse_loss": 0.6834642570040542,
+      "eval_per_var": 3.0010434503424657,
+      "eval_runtime": 151.0663,
+      "eval_samples_per_second": 185.303,
+      "eval_steps_per_second": 2.899,
+      "eval_within_var": 1.7715753001165173,
+      "step": 11264
+    },
+    {
+      "epoch": 0.5320770403214632,
+      "grad_norm": 0.09726043790578842,
+      "learning_rate": 2.3626310850040373e-05,
+      "loss": 0.1772,
+      "step": 11520
+    },
+    {
+      "epoch": 0.5439009745508291,
+      "grad_norm": 0.0948578342795372,
+      "learning_rate": 2.2671605884477816e-05,
+      "loss": 0.1754,
+      "step": 11776
+    },
+    {
+      "epoch": 0.5557249087801949,
+      "grad_norm": 0.08570394665002823,
+      "learning_rate": 2.1720315230424133e-05,
+      "loss": 0.1733,
+      "step": 12032
+    },
+    {
+      "epoch": 0.5675488430095608,
+      "grad_norm": 0.10034994781017303,
+      "learning_rate": 2.0777519879097458e-05,
+      "loss": 0.1722,
+      "step": 12288
+    },
+    {
+      "epoch": 0.5675488430095608,
+      "eval_acr_loss": 0.01237716493560104,
+      "eval_across_var": 0.997331017772901,
+      "eval_bleu": 0.9896437649984566,
+      "eval_ce_loss": 0.026328031107630222,
+      "eval_cos_loss": 0.22285830515296493,
+      "eval_cov": 0.064281376529502,
+      "eval_cov_loss": 0.006827230591828761,
+      "eval_global_var": 2.807202482876712,
+      "eval_loss": 0.12841411385702217,
+      "eval_mse_loss": 0.6729901853489549,
+      "eval_per_var": 3.002407962328767,
+      "eval_within_var": 1.8194298801356799,
+      "step": 12288
+    },
+    {
+      "epoch": 0.5675488430095608,
+      "eval_acr_loss": 0.01237716493560104,
+      "eval_across_var": 0.997331017772901,
+      "eval_bleu": 0.9896437649984566,
+      "eval_ce_loss": 0.026328031107630222,
+      "eval_cos_loss": 0.22285830515296493,
+      "eval_cov": 0.064281376529502,
+      "eval_cov_loss": 0.006827230591828761,
+      "eval_global_var": 2.807202482876712,
+      "eval_loss": 0.12841411385702217,
+      "eval_mse_loss": 0.6729901853489549,
+      "eval_per_var": 3.002407962328767,
+      "eval_runtime": 149.5452,
+      "eval_samples_per_second": 187.188,
+      "eval_steps_per_second": 2.929,
+      "eval_within_var": 1.8194298801356799,
+      "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-v2.1/checkpoint-12288/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c810769b954d6649ec9ac0f62a330cbf05d281180d4c04842e778e387b6a864
+size 5777