Upload folder using huggingface_hub

.gitattributes

@@ -115,3 +115,4 @@ checkpoints-v2.0b-discrete-condition/checkpoint-15360/eval_state.json filter=lfs
 checkpoints-v2.0b-discrete-condition/checkpoint-21651/eval_state.json filter=lfs diff=lfs merge=lfs -text
 checkpoints-v2.0c-discrete-condition/checkpoint-5120/eval_state.json filter=lfs diff=lfs merge=lfs -text
 checkpoints-v2.0c-discrete-condition/checkpoint-12288/eval_state.json filter=lfs diff=lfs merge=lfs -text
+checkpoints-v2.0c-discrete-condition/checkpoint-15360/eval_state.json filter=lfs diff=lfs merge=lfs -text

checkpoints-v2.0c-discrete-condition/checkpoint-15360/eval_state.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b24a75080c41bc32e8c457cfea4bae8467e3bd48635715af0c734fda0e911ff6
+size 44268679

checkpoints-v2.0c-discrete-condition/checkpoint-15360/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8cd6c41c450a6d1ac73ac0227f30190da97ecf03b5e2f290741c7631e66d6e0
+size 15210032

checkpoints-v2.0c-discrete-condition/checkpoint-15360/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb1fa5d79ec18bf21d09d8dc23dd87f185f52e112a30e15f2cf0e20e7da42510
+size 13675339

checkpoints-v2.0c-discrete-condition/checkpoint-15360/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc41eaba0d4516b32fb344341d12fe04439463cc61ef161e6eb071976ff3bc30
+size 14645

checkpoints-v2.0c-discrete-condition/checkpoint-15360/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc04c8dd55f7b7bea5c67b44f5d213d3a00e77277d46903e997e903f6cebead4
+size 1383

checkpoints-v2.0c-discrete-condition/checkpoint-15360/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73b7518420a541537ada4d348aaec8bae1a30b4c2513f08dc3817ea0a09a1265
+size 1465

checkpoints-v2.0c-discrete-condition/checkpoint-15360/trainer_state.json

@@ -0,0 +1,859 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.709436053761951,
+  "eval_steps": 1024,
+  "global_step": 15360,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.011823934229365849,
+      "grad_norm": 0.5773382782936096,
+      "learning_rate": 0.000498046875,
+      "loss": 3.839907646179199,
+      "step": 256
+    },
+    {
+      "epoch": 0.023647868458731697,
+      "grad_norm": 1.2586220502853394,
+      "learning_rate": 0.000998046875,
+      "loss": 2.8118715286254883,
+      "step": 512
+    },
+    {
+      "epoch": 0.03547180268809755,
+      "grad_norm": 1.7730108499526978,
+      "learning_rate": 0.000999640996023194,
+      "loss": 2.415459632873535,
+      "step": 768
+    },
+    {
+      "epoch": 0.047295736917463395,
+      "grad_norm": 2.2911810874938965,
+      "learning_rate": 0.0009985588674043958,
+      "loss": 2.244023084640503,
+      "step": 1024
+    },
+    {
+      "epoch": 0.047295736917463395,
+      "eval_bleu": 0.0009678500579046042,
+      "eval_cos0_loss": 0.3716520652787326,
+      "eval_cos_loss": 0.37628904168736443,
+      "eval_lcos_loss": 0.37847947033300794,
+      "eval_lmse_loss": 1.3972665550501924,
+      "eval_loss": 2.154246258409056,
+      "eval_mse0_loss": 1.5301919299173572,
+      "eval_mse_loss": 1.3810340264072156,
+      "step": 1024
+    },
+    {
+      "epoch": 0.047295736917463395,
+      "eval_bleu": 0.0009678500579046042,
+      "eval_cos0_loss": 0.3716520652787326,
+      "eval_cos_loss": 0.37628904168736443,
+      "eval_lcos_loss": 0.37847947033300794,
+      "eval_lmse_loss": 1.3972665550501924,
+      "eval_loss": 2.154246258409056,
+      "eval_mse0_loss": 1.5301919299173572,
+      "eval_mse_loss": 1.3810340264072156,
+      "eval_runtime": 613.4435,
+      "eval_samples_per_second": 45.633,
+      "eval_steps_per_second": 0.714,
+      "step": 1024
+    },
+    {
+      "epoch": 0.05911967114682925,
+      "grad_norm": 1.7562695741653442,
+      "learning_rate": 0.0009967551747861387,
+      "loss": 2.1243996620178223,
+      "step": 1280
+    },
+    {
+      "epoch": 0.0709436053761951,
+      "grad_norm": 1.6881698369979858,
+      "learning_rate": 0.000994232528651847,
+      "loss": 2.0403342247009277,
+      "step": 1536
+    },
+    {
+      "epoch": 0.08276753960556095,
+      "grad_norm": 2.1058273315429688,
+      "learning_rate": 0.0009909945800260092,
+      "loss": 1.9448120594024658,
+      "step": 1792
+    },
+    {
+      "epoch": 0.09459147383492679,
+      "grad_norm": 2.6205785274505615,
+      "learning_rate": 0.0009870460151900522,
+      "loss": 1.8664555549621582,
+      "step": 2048
+    },
+    {
+      "epoch": 0.09459147383492679,
+      "eval_bleu": 0.002253202053838634,
+      "eval_cos0_loss": 0.30266132699028,
+      "eval_cos_loss": 0.3371365039043775,
+      "eval_lcos_loss": 0.3387556804777825,
+      "eval_lmse_loss": 1.2267555978200206,
+      "eval_loss": 1.827787241434942,
+      "eval_mse0_loss": 1.211973919171721,
+      "eval_mse_loss": 1.2168449740431624,
+      "step": 2048
+    },
+    {
+      "epoch": 0.09459147383492679,
+      "eval_bleu": 0.002253202053838634,
+      "eval_cos0_loss": 0.30266132699028,
+      "eval_cos_loss": 0.3371365039043775,
+      "eval_lcos_loss": 0.3387556804777825,
+      "eval_lmse_loss": 1.2267555978200206,
+      "eval_loss": 1.827787241434942,
+      "eval_mse0_loss": 1.211973919171721,
+      "eval_mse_loss": 1.2168449740431624,
+      "eval_runtime": 612.9503,
+      "eval_samples_per_second": 45.669,
+      "eval_steps_per_second": 0.715,
+      "step": 2048
+    },
+    {
+      "epoch": 0.10641540806429264,
+      "grad_norm": 2.103069305419922,
+      "learning_rate": 0.0009823925488998885,
+      "loss": 1.7817965745925903,
+      "step": 2304
+    },
+    {
+      "epoch": 0.1182393422936585,
+      "grad_norm": 2.6513352394104004,
+      "learning_rate": 0.0009770409161149525,
+      "loss": 1.700106143951416,
+      "step": 2560
+    },
+    {
+      "epoch": 0.13006327652302435,
+      "grad_norm": 2.6591241359710693,
+      "learning_rate": 0.0009709988622506973,
+      "loss": 1.6390057802200317,
+      "step": 2816
+    },
+    {
+      "epoch": 0.1418872107523902,
+      "grad_norm": 2.2664096355438232,
+      "learning_rate": 0.000964275131968659,
+      "loss": 1.5896209478378296,
+      "step": 3072
+    },
+    {
+      "epoch": 0.1418872107523902,
+      "eval_bleu": 0.02088055114550435,
+      "eval_cos0_loss": 0.24901231800176238,
+      "eval_cos_loss": 0.30351526511314253,
+      "eval_lcos_loss": 0.3036922022496184,
+      "eval_lmse_loss": 1.0799240483540922,
+      "eval_loss": 1.5640880138906714,
+      "eval_mse0_loss": 0.9713506719020948,
+      "eval_mse_loss": 1.0769013045313152,
+      "step": 3072
+    },
+    {
+      "epoch": 0.1418872107523902,
+      "eval_bleu": 0.02088055114550435,
+      "eval_cos0_loss": 0.24901231800176238,
+      "eval_cos_loss": 0.30351526511314253,
+      "eval_lcos_loss": 0.3036922022496184,
+      "eval_lmse_loss": 1.0799240483540922,
+      "eval_loss": 1.5640880138906714,
+      "eval_mse0_loss": 0.9713506719020948,
+      "eval_mse_loss": 1.0769013045313152,
+      "eval_runtime": 609.2065,
+      "eval_samples_per_second": 45.95,
+      "eval_steps_per_second": 0.719,
+      "step": 3072
+    },
+    {
+      "epoch": 0.15371114498175603,
+      "grad_norm": 4.887567043304443,
+      "learning_rate": 0.0009568794565203123,
+      "loss": 1.548627495765686,
+      "step": 3328
+    },
+    {
+      "epoch": 0.1655350792111219,
+      "grad_norm": 2.5831308364868164,
+      "learning_rate": 0.0009488225396630347,
+      "loss": 1.5119283199310303,
+      "step": 3584
+    },
+    {
+      "epoch": 0.17735901344048774,
+      "grad_norm": 2.4399147033691406,
+      "learning_rate": 0.0009401160421685646,
+      "loss": 1.4889769554138184,
+      "step": 3840
+    },
+    {
+      "epoch": 0.18918294766985358,
+      "grad_norm": 3.813706636428833,
+      "learning_rate": 0.0009307725649463714,
+      "loss": 1.4559980630874634,
+      "step": 4096
+    },
+    {
+      "epoch": 0.18918294766985358,
+      "eval_bleu": 0.04036824125921908,
+      "eval_cos0_loss": 0.22499571029590146,
+      "eval_cos_loss": 0.2870930704760225,
+      "eval_lcos_loss": 0.28663786027801635,
+      "eval_lmse_loss": 1.0095530485725839,
+      "eval_loss": 1.4431848909756908,
+      "eval_mse0_loss": 0.8669073336744961,
+      "eval_mse_loss": 1.009909397526963,
+      "step": 4096
+    },
+    {
+      "epoch": 0.18918294766985358,
+      "eval_bleu": 0.04036824125921908,
+      "eval_cos0_loss": 0.22499571029590146,
+      "eval_cos_loss": 0.2870930704760225,
+      "eval_lcos_loss": 0.28663786027801635,
+      "eval_lmse_loss": 1.0095530485725839,
+      "eval_loss": 1.4431848909756908,
+      "eval_mse0_loss": 0.8669073336744961,
+      "eval_mse_loss": 1.009909397526963,
+      "eval_runtime": 610.7117,
+      "eval_samples_per_second": 45.837,
+      "eval_steps_per_second": 0.717,
+      "step": 4096
+    },
+    {
+      "epoch": 0.20100688189921945,
+      "grad_norm": 2.183851957321167,
+      "learning_rate": 0.0009208056308063659,
+      "loss": 1.4380117654800415,
+      "step": 4352
+    },
+    {
+      "epoch": 0.2128308161285853,
+      "grad_norm": 2.0409207344055176,
+      "learning_rate": 0.0009102296648873445,
+      "loss": 1.4116824865341187,
+      "step": 4608
+    },
+    {
+      "epoch": 0.22465475035795113,
+      "grad_norm": 3.0817489624023438,
+      "learning_rate": 0.0008990599737794927,
+      "loss": 1.3894058465957642,
+      "step": 4864
+    },
+    {
+      "epoch": 0.236478684587317,
+      "grad_norm": 2.6919968128204346,
+      "learning_rate": 0.0008873127233711644,
+      "loss": 1.3580043315887451,
+      "step": 5120
+    },
+    {
+      "epoch": 0.236478684587317,
+      "eval_bleu": 0.05952291809050901,
+      "eval_cos0_loss": 0.20102498530661134,
+      "eval_cos_loss": 0.2724936866120661,
+      "eval_lcos_loss": 0.27243535475779884,
+      "eval_lmse_loss": 0.9554824699822082,
+      "eval_loss": 1.3415723780518798,
+      "eval_mse0_loss": 0.7718455541079448,
+      "eval_mse_loss": 0.955816726978511,
+      "step": 5120
+    },
+    {
+      "epoch": 0.236478684587317,
+      "eval_bleu": 0.05952291809050901,
+      "eval_cos0_loss": 0.20102498530661134,
+      "eval_cos_loss": 0.2724936866120661,
+      "eval_lcos_loss": 0.27243535475779884,
+      "eval_lmse_loss": 0.9554824699822082,
+      "eval_loss": 1.3415723780518798,
+      "eval_mse0_loss": 0.7718455541079448,
+      "eval_mse_loss": 0.955816726978511,
+      "eval_runtime": 608.2141,
+      "eval_samples_per_second": 46.025,
+      "eval_steps_per_second": 0.72,
+      "step": 5120
+    },
+    {
+      "epoch": 0.24830261881668284,
+      "grad_norm": 4.273655891418457,
+      "learning_rate": 0.0008750049154520011,
+      "loss": 1.3323787450790405,
+      "step": 5376
+    },
+    {
+      "epoch": 0.2601265530460487,
+      "grad_norm": 2.282118320465088,
+      "learning_rate": 0.0008621543631062487,
+      "loss": 1.3110138177871704,
+      "step": 5632
+    },
+    {
+      "epoch": 0.27195048727541454,
+      "grad_norm": 3.549677848815918,
+      "learning_rate": 0.0008487796649318904,
+      "loss": 1.2956509590148926,
+      "step": 5888
+    },
+    {
+      "epoch": 0.2837744215047804,
+      "grad_norm": 2.1778132915496826,
+      "learning_rate": 0.0008349001781229053,
+      "loss": 1.2801446914672852,
+      "step": 6144
+    },
+    {
+      "epoch": 0.2837744215047804,
+      "eval_bleu": 0.10035188687728737,
+      "eval_cos0_loss": 0.18733559718959408,
+      "eval_cos_loss": 0.26199433217718177,
+      "eval_lcos_loss": 0.2604647331518125,
+      "eval_lmse_loss": 0.9071085341444843,
+      "eval_loss": 1.2658131002835487,
+      "eval_mse0_loss": 0.7129845402828635,
+      "eval_mse_loss": 0.911533124370662,
+      "step": 6144
+    },
+    {
+      "epoch": 0.2837744215047804,
+      "eval_bleu": 0.10035188687728737,
+      "eval_cos0_loss": 0.18733559718959408,
+      "eval_cos_loss": 0.26199433217718177,
+      "eval_lcos_loss": 0.2604647331518125,
+      "eval_lmse_loss": 0.9071085341444843,
+      "eval_loss": 1.2658131002835487,
+      "eval_mse0_loss": 0.7129845402828635,
+      "eval_mse_loss": 0.911533124370662,
+      "eval_runtime": 629.5224,
+      "eval_samples_per_second": 44.467,
+      "eval_steps_per_second": 0.696,
+      "step": 6144
+    },
+    {
+      "epoch": 0.2955983557341462,
+      "grad_norm": 2.317173957824707,
+      "learning_rate": 0.0008205359904536107,
+      "loss": 1.2554467916488647,
+      "step": 6400
+    },
+    {
+      "epoch": 0.30742228996351206,
+      "grad_norm": 2.557554006576538,
+      "learning_rate": 0.0008057078912056363,
+      "loss": 1.249757170677185,
+      "step": 6656
+    },
+    {
+      "epoch": 0.3192462241928779,
+      "grad_norm": 1.894919991493225,
+      "learning_rate": 0.0007904373410796086,
+      "loss": 1.2362620830535889,
+      "step": 6912
+    },
+    {
+      "epoch": 0.3310701584222438,
+      "grad_norm": 1.9156849384307861,
+      "learning_rate": 0.0007747464411350876,
+      "loss": 1.2148041725158691,
+      "step": 7168
+    },
+    {
+      "epoch": 0.3310701584222438,
+      "eval_bleu": 0.1203407967843517,
+      "eval_cos0_loss": 0.1777811785607033,
+      "eval_cos_loss": 0.2516806123599614,
+      "eval_lcos_loss": 0.25071116482422234,
+      "eval_lmse_loss": 0.8678208071075074,
+      "eval_loss": 1.2057547844163903,
+      "eval_mse0_loss": 0.6730798853042463,
+      "eval_mse_loss": 0.8706088818643736,
+      "step": 7168
+    },
+    {
+      "epoch": 0.3310701584222438,
+      "eval_bleu": 0.1203407967843517,
+      "eval_cos0_loss": 0.1777811785607033,
+      "eval_cos_loss": 0.2516806123599614,
+      "eval_lcos_loss": 0.25071116482422234,
+      "eval_lmse_loss": 0.8678208071075074,
+      "eval_loss": 1.2057547844163903,
+      "eval_mse0_loss": 0.6730798853042463,
+      "eval_mse_loss": 0.8706088818643736,
+      "eval_runtime": 621.8627,
+      "eval_samples_per_second": 45.015,
+      "eval_steps_per_second": 0.704,
+      "step": 7168
+    },
+    {
+      "epoch": 0.34289409265160964,
+      "grad_norm": 3.0442354679107666,
+      "learning_rate": 0.000758657900803716,
+      "loss": 1.2075221538543701,
+      "step": 7424
+    },
+    {
+      "epoch": 0.3547180268809755,
+      "grad_norm": 2.7499632835388184,
+      "learning_rate": 0.000742195005021869,
+      "loss": 1.1966240406036377,
+      "step": 7680
+    },
+    {
+      "epoch": 0.3665419611103413,
+      "grad_norm": 4.639651775360107,
+      "learning_rate": 0.0007253815805303786,
+      "loss": 1.1882134675979614,
+      "step": 7936
+    },
+    {
+      "epoch": 0.37836589533970716,
+      "grad_norm": 2.148693323135376,
+      "learning_rate": 0.0007082419613901028,
+      "loss": 1.177225947380066,
+      "step": 8192
+    },
+    {
+      "epoch": 0.37836589533970716,
+      "eval_bleu": 0.1438011815022915,
+      "eval_cos0_loss": 0.17181862849895269,
+      "eval_cos_loss": 0.24644448921152445,
+      "eval_lcos_loss": 0.24526936169628683,
+      "eval_lmse_loss": 0.846808816880396,
+      "eval_loss": 1.1733890529636923,
+      "eval_mse0_loss": 0.6489214906681618,
+      "eval_mse_loss": 0.8510477972901576,
+      "step": 8192
+    },
+    {
+      "epoch": 0.37836589533970716,
+      "eval_bleu": 0.1438011815022915,
+      "eval_cos0_loss": 0.17181862849895269,
+      "eval_cos_loss": 0.24644448921152445,
+      "eval_lcos_loss": 0.24526936169628683,
+      "eval_lmse_loss": 0.846808816880396,
+      "eval_loss": 1.1733890529636923,
+      "eval_mse0_loss": 0.6489214906681618,
+      "eval_mse_loss": 0.8510477972901576,
+      "eval_runtime": 622.9709,
+      "eval_samples_per_second": 44.935,
+      "eval_steps_per_second": 0.703,
+      "step": 8192
+    },
+    {
+      "epoch": 0.390189829569073,
+      "grad_norm": 2.8198015689849854,
+      "learning_rate": 0.0006908009537632514,
+      "loss": 1.1730091571807861,
+      "step": 8448
+    },
+    {
+      "epoch": 0.4020137637984389,
+      "grad_norm": 1.820741057395935,
+      "learning_rate": 0.0006730838000114403,
+      "loss": 1.1629409790039062,
+      "step": 8704
+    },
+    {
+      "epoch": 0.41383769802780473,
+      "grad_norm": 2.2142693996429443,
+      "learning_rate": 0.0006551161421624341,
+      "loss": 1.1535300016403198,
+      "step": 8960
+    },
+    {
+      "epoch": 0.4256616322571706,
+      "grad_norm": 2.3717238903045654,
+      "learning_rate": 0.0006369239847984517,
+      "loss": 1.1437243223190308,
+      "step": 9216
+    },
+    {
+      "epoch": 0.4256616322571706,
+      "eval_bleu": 0.15976181512208656,
+      "eval_cos0_loss": 0.16595067757449738,
+      "eval_cos_loss": 0.2388318737279879,
+      "eval_lcos_loss": 0.23841320053083168,
+      "eval_lmse_loss": 0.8197641533259388,
+      "eval_loss": 1.132960263180406,
+      "eval_mse0_loss": 0.6245599874894913,
+      "eval_mse_loss": 0.8215963885392228,
+      "step": 9216
+    },
+    {
+      "epoch": 0.4256616322571706,
+      "eval_bleu": 0.15976181512208656,
+      "eval_cos0_loss": 0.16595067757449738,
+      "eval_cos_loss": 0.2388318737279879,
+      "eval_lcos_loss": 0.23841320053083168,
+      "eval_lmse_loss": 0.8197641533259388,
+      "eval_loss": 1.132960263180406,
+      "eval_mse0_loss": 0.6245599874894913,
+      "eval_mse_loss": 0.8215963885392228,
+      "eval_runtime": 622.7806,
+      "eval_samples_per_second": 44.948,
+      "eval_steps_per_second": 0.703,
+      "step": 9216
+    },
+    {
+      "epoch": 0.4374855664865364,
+      "grad_norm": 3.3975112438201904,
+      "learning_rate": 0.0006185336574197479,
+      "loss": 1.1262043714523315,
+      "step": 9472
+    },
+    {
+      "epoch": 0.44930950071590225,
+      "grad_norm": 3.0337095260620117,
+      "learning_rate": 0.0005999717763379407,
+      "loss": 1.1239513158798218,
+      "step": 9728
+    },
+    {
+      "epoch": 0.4611334349452681,
+      "grad_norm": 1.9637302160263062,
+      "learning_rate": 0.0005812652061542363,
+      "loss": 1.1128973960876465,
+      "step": 9984
+    },
+    {
+      "epoch": 0.472957369174634,
+      "grad_norm": 2.980830669403076,
+      "learning_rate": 0.0005624410208783071,
+      "loss": 1.106379508972168,
+      "step": 10240
+    },
+    {
+      "epoch": 0.472957369174634,
+      "eval_bleu": 0.17188861377853284,
+      "eval_cos0_loss": 0.1620413672230015,
+      "eval_cos_loss": 0.23449426780552624,
+      "eval_lcos_loss": 0.2336000114951504,
+      "eval_lmse_loss": 0.8000838346949452,
+      "eval_loss": 1.1057640148080103,
+      "eval_mse0_loss": 0.6082026891784581,
+      "eval_mse_loss": 0.8032415092807926,
+      "step": 10240
+    },
+    {
+      "epoch": 0.472957369174634,
+      "eval_bleu": 0.17188861377853284,
+      "eval_cos0_loss": 0.1620413672230015,
+      "eval_cos_loss": 0.23449426780552624,
+      "eval_lcos_loss": 0.2336000114951504,
+      "eval_lmse_loss": 0.8000838346949452,
+      "eval_loss": 1.1057640148080103,
+      "eval_mse0_loss": 0.6082026891784581,
+      "eval_mse_loss": 0.8032415092807926,
+      "eval_runtime": 623.2144,
+      "eval_samples_per_second": 44.917,
+      "eval_steps_per_second": 0.703,
+      "step": 10240
+    },
+    {
+      "epoch": 0.48478130340399983,
+      "grad_norm": 2.48860502243042,
+      "learning_rate": 0.0005435264647440881,
+      "loss": 1.1007436513900757,
+      "step": 10496
+    },
+    {
+      "epoch": 0.49660523763336567,
+      "grad_norm": 2.1356256008148193,
+      "learning_rate": 0.000524548912779213,
+      "loss": 1.093902826309204,
+      "step": 10752
+    },
+    {
+      "epoch": 0.5084291718627315,
+      "grad_norm": 1.7593225240707397,
+      "learning_rate": 0.0005055358311851499,
+      "loss": 1.091771125793457,
+      "step": 11008
+    },
+    {
+      "epoch": 0.5202531060920974,
+      "grad_norm": 2.2384347915649414,
+      "learning_rate": 0.0004865147375853812,
+      "loss": 1.0827676057815552,
+      "step": 11264
+    },
+    {
+      "epoch": 0.5202531060920974,
+      "eval_bleu": 0.17610335609749747,
+      "eval_cos0_loss": 0.1587146663162262,
+      "eval_cos_loss": 0.2287932799014871,
+      "eval_lcos_loss": 0.22836553404048154,
+      "eval_lmse_loss": 0.7806114094442429,
+      "eval_loss": 1.0793486076947216,
+      "eval_mse0_loss": 0.5959821122966401,
+      "eval_mse_loss": 0.7821036917433891,
+      "step": 11264
+    },
+    {
+      "epoch": 0.5202531060920974,
+      "eval_bleu": 0.17610335609749747,
+      "eval_cos0_loss": 0.1587146663162262,
+      "eval_cos_loss": 0.2287932799014871,
+      "eval_lcos_loss": 0.22836553404048154,
+      "eval_lmse_loss": 0.7806114094442429,
+      "eval_loss": 1.0793486076947216,
+      "eval_mse0_loss": 0.5959821122966401,
+      "eval_mse_loss": 0.7821036917433891,
+      "eval_runtime": 628.9654,
+      "eval_samples_per_second": 44.506,
+      "eval_steps_per_second": 0.696,
+      "step": 11264
+    },
+    {
+      "epoch": 0.5320770403214632,
+      "grad_norm": 1.7046668529510498,
+      "learning_rate": 0.0004675131611991607,
+      "loss": 1.0763407945632935,
+      "step": 11520
+    },
+    {
+      "epoch": 0.5439009745508291,
+      "grad_norm": 2.0045571327209473,
+      "learning_rate": 0.0004485586029984899,
+      "loss": 1.0701327323913574,
+      "step": 11776
+    },
+    {
+      "epoch": 0.5557249087801949,
+      "grad_norm": 1.6229290962219238,
+      "learning_rate": 0.00042967849590597266,
+      "loss": 1.0641701221466064,
+      "step": 12032
+    },
+    {
+      "epoch": 0.5675488430095608,
+      "grad_norm": 2.0531442165374756,
+      "learning_rate": 0.0004109001650911621,
+      "loss": 1.0628031492233276,
+      "step": 12288
+    },
+    {
+      "epoch": 0.5675488430095608,
+      "eval_bleu": 0.19117070322403804,
+      "eval_cos0_loss": 0.1558652590943254,
+      "eval_cos_loss": 0.22613805709364207,
+      "eval_lcos_loss": 0.22542561093966165,
+      "eval_lmse_loss": 0.7683645386129754,
+      "eval_loss": 1.0618582483568149,
+      "eval_mse0_loss": 0.5838932091514814,
+      "eval_mse_loss": 0.7714587522151808,
+      "step": 12288
+    },
+    {
+      "epoch": 0.5675488430095608,
+      "eval_bleu": 0.19117070322403804,
+      "eval_cos0_loss": 0.1558652590943254,
+      "eval_cos_loss": 0.22613805709364207,
+      "eval_lcos_loss": 0.22542561093966165,
+      "eval_lmse_loss": 0.7683645386129754,
+      "eval_loss": 1.0618582483568149,
+      "eval_mse0_loss": 0.5838932091514814,
+      "eval_mse_loss": 0.7714587522151808,
+      "eval_runtime": 628.0916,
+      "eval_samples_per_second": 44.568,
+      "eval_steps_per_second": 0.697,
+      "step": 12288
+    },
+    {
+      "epoch": 0.5793727772389267,
+      "grad_norm": 1.896727442741394,
+      "learning_rate": 0.0003922507884228551,
+      "loss": 1.0593478679656982,
+      "step": 12544
+    },
+    {
+      "epoch": 0.5911967114682924,
+      "grad_norm": 1.6579889059066772,
+      "learning_rate": 0.00037375735713457723,
+      "loss": 1.052514672279358,
+      "step": 12800
+    },
+    {
+      "epoch": 0.6030206456976583,
+      "grad_norm": 2.1612658500671387,
+      "learning_rate": 0.00035544663676018276,
+      "loss": 1.0489237308502197,
+      "step": 13056
+    },
+    {
+      "epoch": 0.6148445799270241,
+      "grad_norm": 2.4297049045562744,
+      "learning_rate": 0.00033734512839611255,
+      "loss": 1.0462322235107422,
+      "step": 13312
+    },
+    {
+      "epoch": 0.6148445799270241,
+      "eval_bleu": 0.20081413615671828,
+      "eval_cos0_loss": 0.15379069088936942,
+      "eval_cos_loss": 0.22239228416251267,
+      "eval_lcos_loss": 0.22114114167347346,
+      "eval_lmse_loss": 0.7518216585187607,
+      "eval_loss": 1.0409937676501602,
+      "eval_mse0_loss": 0.5750884450461766,
+      "eval_mse_loss": 0.7550774360900601,
+      "step": 13312
+    },
+    {
+      "epoch": 0.6148445799270241,
+      "eval_bleu": 0.20081413615671828,
+      "eval_cos0_loss": 0.15379069088936942,
+      "eval_cos_loss": 0.22239228416251267,
+      "eval_lcos_loss": 0.22114114167347346,
+      "eval_lmse_loss": 0.7518216585187607,
+      "eval_loss": 1.0409937676501602,
+      "eval_mse0_loss": 0.5750884450461766,
+      "eval_mse_loss": 0.7550774360900601,
+      "eval_runtime": 592.8663,
+      "eval_samples_per_second": 47.216,
+      "eval_steps_per_second": 0.739,
+      "step": 13312
+    },
+    {
+      "epoch": 0.62666851415639,
+      "grad_norm": 1.7425934076309204,
+      "learning_rate": 0.0003194790303463687,
+      "loss": 1.0401134490966797,
+      "step": 13568
+    },
+    {
+      "epoch": 0.6384924483857558,
+      "grad_norm": 2.6148269176483154,
+      "learning_rate": 0.00030187420020572406,
+      "loss": 1.0375194549560547,
+      "step": 13824
+    },
+    {
+      "epoch": 0.6503163826151217,
+      "grad_norm": 2.0934908390045166,
+      "learning_rate": 0.00028455611743603626,
+      "loss": 1.032192587852478,
+      "step": 14080
+    },
+    {
+      "epoch": 0.6621403168444876,
+      "grad_norm": 2.202070713043213,
+      "learning_rate": 0.0002675498464898373,
+      "loss": 1.0297998189926147,
+      "step": 14336
+    },
+    {
+      "epoch": 0.6621403168444876,
+      "eval_bleu": 0.20754117009063805,
+      "eval_cos0_loss": 0.15258467177005663,
+      "eval_cos_loss": 0.22050917563645261,
+      "eval_lcos_loss": 0.21983005062221936,
+      "eval_lmse_loss": 0.7468018018763903,
+      "eval_loss": 1.0333503182620218,
+      "eval_mse0_loss": 0.5706300363965231,
+      "eval_mse_loss": 0.7492687921273654,
+      "step": 14336
+    },
+    {
+      "epoch": 0.6621403168444876,
+      "eval_bleu": 0.20754117009063805,
+      "eval_cos0_loss": 0.15258467177005663,
+      "eval_cos_loss": 0.22050917563645261,
+      "eval_lcos_loss": 0.21983005062221936,
+      "eval_lmse_loss": 0.7468018018763903,
+      "eval_loss": 1.0333503182620218,
+      "eval_mse0_loss": 0.5706300363965231,
+      "eval_mse_loss": 0.7492687921273654,
+      "eval_runtime": 587.051,
+      "eval_samples_per_second": 47.684,
+      "eval_steps_per_second": 0.746,
+      "step": 14336
+    },
+    {
+      "epoch": 0.6739642510738534,
+      "grad_norm": 2.002934455871582,
+      "learning_rate": 0.0002508800005345623,
+      "loss": 1.0300023555755615,
+      "step": 14592
+    },
+    {
+      "epoch": 0.6857881853032193,
+      "grad_norm": 1.974485158920288,
+      "learning_rate": 0.00023457070582992562,
+      "loss": 1.02817702293396,
+      "step": 14848
+    },
+    {
+      "epoch": 0.6976121195325851,
+      "grad_norm": 2.13757061958313,
+      "learning_rate": 0.00021864556680999692,
+      "loss": 1.023733139038086,
+      "step": 15104
+    },
+    {
+      "epoch": 0.709436053761951,
+      "grad_norm": 1.8265743255615234,
+      "learning_rate": 0.0002031276319205152,
+      "loss": 1.019716501235962,
+      "step": 15360
+    },
+    {
+      "epoch": 0.709436053761951,
+      "eval_bleu": 0.21009732847173415,
+      "eval_cos0_loss": 0.15065579374903412,
+      "eval_cos_loss": 0.2184738382424938,
+      "eval_lcos_loss": 0.21706832843284085,
+      "eval_lmse_loss": 0.7356251572092919,
+      "eval_loss": 1.0186850880379001,
+      "eval_mse0_loss": 0.5626335150847152,
+      "eval_mse_loss": 0.7391115056869646,
+      "step": 15360
+    },
+    {
+      "epoch": 0.709436053761951,
+      "eval_bleu": 0.21009732847173415,
+      "eval_cos0_loss": 0.15065579374903412,
+      "eval_cos_loss": 0.2184738382424938,
+      "eval_lcos_loss": 0.21706832843284085,
+      "eval_lmse_loss": 0.7356251572092919,
+      "eval_loss": 1.0186850880379001,
+      "eval_mse0_loss": 0.5626335150847152,
+      "eval_mse_loss": 0.7391115056869646,
+      "eval_runtime": 589.1266,
+      "eval_samples_per_second": 47.516,
+      "eval_steps_per_second": 0.743,
+      "step": 15360
+    }
+  ],
+  "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-v2.0c-discrete-condition/checkpoint-15360/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f63a8a711fb280c6320e95476c532b75c862f31d6a218a0d05079e32411a5ae1
+size 5201