Training in progress, step 200, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:07a03b826407206fe6e5f14fca113c3f559c3cff04d924fd3ac5453addd96cef
+oid sha256:495448502f763b86d28dda524b7eb5eceebcee1f4b70e09714c53c29d0af4a94
 size 1204678496

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:642367ed92e204c923279bdf6bb638684b6dba12ac6c577fad7b0638120a6da9
-size 341366172
+oid sha256:3adc770c65652ac004a4500adf7ed16738e9acb3816ccbdac986e98a23139520
+size 341366620

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca2909acf1948419fcf13aff2756604c948567401ae936261f4764d4beb350d5
+oid sha256:4d941d1a1ef2faca0ae11de7ca6a381a98c904eb294ce6b248f89c9fb94663aa
 size 1064

last-checkpoint/trainer_state.json

@@ -10,1522 +10,1522 @@
   "log_history": [
     {
       "epoch": 0.0,
-      "learning_rate": 0.0,
-      "loss": 3.2602,
+      "learning_rate": 2.5e-05,
+      "loss": 0.8445,
       "step": 1
     },
     {
       "epoch": 0.01,
-      "learning_rate": 0.0,
-      "loss": 4.1154,
+      "learning_rate": 5e-05,
+      "loss": 0.8413,
       "step": 2
     },
     {
       "epoch": 0.01,
-      "learning_rate": 0.0,
-      "loss": 3.959,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.7449,
       "step": 3
     },
     {
       "epoch": 0.02,
-      "learning_rate": 2.5e-05,
-      "loss": 4.3797,
+      "learning_rate": 0.0001,
+      "loss": 0.6177,
       "step": 4
     },
     {
       "epoch": 0.02,
-      "learning_rate": 5e-05,
-      "loss": 4.3625,
+      "learning_rate": 0.000125,
+      "loss": 0.7529,
       "step": 5
     },
     {
       "epoch": 0.02,
-      "eval_loss": 3.1120874881744385,
-      "eval_runtime": 52.7995,
-      "eval_samples_per_second": 4.735,
-      "eval_steps_per_second": 0.606,
+      "eval_loss": 0.6046672463417053,
+      "eval_runtime": 54.0642,
+      "eval_samples_per_second": 4.624,
+      "eval_steps_per_second": 0.592,
       "step": 5
     },
     {
       "epoch": 0.02,
-      "learning_rate": 7.500000000000001e-05,
-      "loss": 2.8,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 0.525,
       "step": 6
     },
     {
       "epoch": 0.03,
-      "learning_rate": 7.500000000000001e-05,
-      "loss": 2.0775,
+      "learning_rate": 0.000175,
+      "loss": 0.7356,
       "step": 7
     },
     {
       "epoch": 0.03,
-      "learning_rate": 0.0001,
-      "loss": 1.9897,
+      "learning_rate": 0.0002,
+      "loss": 0.522,
       "step": 8
     },
     {
       "epoch": 0.04,
-      "learning_rate": 0.0001,
-      "loss": 1.4163,
+      "learning_rate": 0.00019917355371900828,
+      "loss": 0.795,
       "step": 9
     },
     {
       "epoch": 0.04,
-      "learning_rate": 0.000125,
-      "loss": 1.3422,
+      "learning_rate": 0.00019834710743801655,
+      "loss": 0.4396,
       "step": 10
     },
     {
       "epoch": 0.04,
-      "eval_loss": 1.08388090133667,
-      "eval_runtime": 53.6216,
-      "eval_samples_per_second": 4.662,
-      "eval_steps_per_second": 0.597,
+      "eval_loss": 0.6954616904258728,
+      "eval_runtime": 53.7057,
+      "eval_samples_per_second": 4.655,
+      "eval_steps_per_second": 0.596,
       "step": 10
     },
     {
       "epoch": 0.04,
-      "learning_rate": 0.00015000000000000001,
-      "loss": 1.2575,
+      "learning_rate": 0.00019752066115702482,
+      "loss": 0.6409,
       "step": 11
     },
     {
       "epoch": 0.05,
-      "learning_rate": 0.000175,
-      "loss": 1.1459,
+      "learning_rate": 0.0001966942148760331,
+      "loss": 0.4477,
       "step": 12
     },
     {
       "epoch": 0.05,
-      "learning_rate": 0.0002,
-      "loss": 0.9636,
+      "learning_rate": 0.00019586776859504133,
+      "loss": 0.4022,
       "step": 13
     },
     {
       "epoch": 0.06,
-      "learning_rate": 0.00019917355371900828,
-      "loss": 1.001,
+      "learning_rate": 0.0001950413223140496,
+      "loss": 0.473,
       "step": 14
     },
     {
       "epoch": 0.06,
-      "learning_rate": 0.00019834710743801655,
-      "loss": 1.1057,
+      "learning_rate": 0.00019421487603305787,
+      "loss": 0.7096,
       "step": 15
     },
     {
       "epoch": 0.06,
-      "eval_loss": 0.8665942549705505,
-      "eval_runtime": 53.7019,
-      "eval_samples_per_second": 4.655,
-      "eval_steps_per_second": 0.596,
+      "eval_loss": 0.9140636324882507,
+      "eval_runtime": 53.7517,
+      "eval_samples_per_second": 4.651,
+      "eval_steps_per_second": 0.595,
       "step": 15
     },
     {
       "epoch": 0.06,
-      "learning_rate": 0.00019752066115702482,
-      "loss": 0.9466,
+      "learning_rate": 0.0001933884297520661,
+      "loss": 0.5573,
       "step": 16
     },
     {
       "epoch": 0.07,
-      "learning_rate": 0.0001966942148760331,
-      "loss": 0.7701,
+      "learning_rate": 0.00019256198347107438,
+      "loss": 0.4636,
       "step": 17
     },
     {
       "epoch": 0.07,
-      "learning_rate": 0.00019586776859504133,
-      "loss": 0.8989,
+      "learning_rate": 0.00019173553719008265,
+      "loss": 0.5834,
       "step": 18
     },
     {
       "epoch": 0.08,
-      "learning_rate": 0.0001950413223140496,
-      "loss": 0.9402,
+      "learning_rate": 0.00019090909090909092,
+      "loss": 0.5678,
       "step": 19
     },
     {
       "epoch": 0.08,
-      "learning_rate": 0.00019421487603305787,
-      "loss": 0.7351,
+      "learning_rate": 0.0001900826446280992,
+      "loss": 0.5193,
       "step": 20
     },
     {
       "epoch": 0.08,
-      "eval_loss": 0.8300237059593201,
-      "eval_runtime": 53.6887,
-      "eval_samples_per_second": 4.656,
-      "eval_steps_per_second": 0.596,
+      "eval_loss": 0.9248924851417542,
+      "eval_runtime": 53.5727,
+      "eval_samples_per_second": 4.667,
+      "eval_steps_per_second": 0.597,
       "step": 20
     },
     {
       "epoch": 0.08,
-      "learning_rate": 0.0001933884297520661,
-      "loss": 0.8839,
+      "learning_rate": 0.00018925619834710743,
+      "loss": 0.5513,
       "step": 21
     },
     {
       "epoch": 0.09,
-      "learning_rate": 0.00019256198347107438,
-      "loss": 0.6989,
+      "learning_rate": 0.0001884297520661157,
+      "loss": 0.4522,
       "step": 22
     },
     {
       "epoch": 0.09,
-      "learning_rate": 0.00019173553719008265,
-      "loss": 0.79,
+      "learning_rate": 0.00018760330578512397,
+      "loss": 0.53,
       "step": 23
     },
     {
       "epoch": 0.1,
-      "learning_rate": 0.00019090909090909092,
-      "loss": 0.8464,
+      "learning_rate": 0.00018677685950413224,
+      "loss": 0.5113,
       "step": 24
     },
     {
       "epoch": 0.1,
-      "learning_rate": 0.0001900826446280992,
-      "loss": 0.8272,
+      "learning_rate": 0.0001859504132231405,
+      "loss": 0.4743,
       "step": 25
     },
     {
       "epoch": 0.1,
-      "eval_loss": 0.729701042175293,
-      "eval_runtime": 53.6573,
-      "eval_samples_per_second": 4.659,
-      "eval_steps_per_second": 0.596,
+      "eval_loss": 0.8713067173957825,
+      "eval_runtime": 53.5406,
+      "eval_samples_per_second": 4.669,
+      "eval_steps_per_second": 0.598,
       "step": 25
     },
     {
       "epoch": 0.1,
-      "learning_rate": 0.00018925619834710743,
-      "loss": 0.8048,
+      "learning_rate": 0.00018512396694214878,
+      "loss": 0.5393,
       "step": 26
     },
     {
       "epoch": 0.11,
-      "learning_rate": 0.0001884297520661157,
-      "loss": 0.7717,
+      "learning_rate": 0.00018429752066115705,
+      "loss": 0.5302,
       "step": 27
     },
     {
       "epoch": 0.11,
-      "learning_rate": 0.00018760330578512397,
-      "loss": 0.7662,
+      "learning_rate": 0.00018347107438016532,
+      "loss": 0.9226,
       "step": 28
     },
     {
       "epoch": 0.12,
-      "learning_rate": 0.00018677685950413224,
-      "loss": 0.782,
+      "learning_rate": 0.00018264462809917356,
+      "loss": 0.6233,
       "step": 29
     },
     {
       "epoch": 0.12,
-      "learning_rate": 0.0001859504132231405,
-      "loss": 0.619,
+      "learning_rate": 0.00018181818181818183,
+      "loss": 0.6283,
       "step": 30
     },
     {
       "epoch": 0.12,
-      "eval_loss": 0.7578176259994507,
-      "eval_runtime": 53.7263,
-      "eval_samples_per_second": 4.653,
-      "eval_steps_per_second": 0.596,
+      "eval_loss": 0.9399847388267517,
+      "eval_runtime": 53.5962,
+      "eval_samples_per_second": 4.665,
+      "eval_steps_per_second": 0.597,
       "step": 30
     },
     {
       "epoch": 0.12,
-      "learning_rate": 0.00018512396694214878,
-      "loss": 0.7644,
+      "learning_rate": 0.00018099173553719008,
+      "loss": 0.4825,
       "step": 31
     },
     {
       "epoch": 0.13,
-      "learning_rate": 0.00018429752066115705,
-      "loss": 0.6039,
+      "learning_rate": 0.00018016528925619835,
+      "loss": 0.4793,
       "step": 32
     },
     {
       "epoch": 0.13,
-      "learning_rate": 0.00018347107438016532,
-      "loss": 0.6441,
+      "learning_rate": 0.00017933884297520662,
+      "loss": 0.4345,
       "step": 33
     },
     {
       "epoch": 0.14,
-      "learning_rate": 0.00018264462809917356,
-      "loss": 0.5813,
+      "learning_rate": 0.00017851239669421489,
+      "loss": 0.342,
       "step": 34
     },
     {
       "epoch": 0.14,
-      "learning_rate": 0.00018181818181818183,
-      "loss": 0.7727,
+      "learning_rate": 0.00017768595041322316,
+      "loss": 0.6307,
       "step": 35
     },
     {
       "epoch": 0.14,
-      "eval_loss": 0.7339813113212585,
-      "eval_runtime": 53.801,
-      "eval_samples_per_second": 4.647,
-      "eval_steps_per_second": 0.595,
+      "eval_loss": 0.9868490695953369,
+      "eval_runtime": 53.6775,
+      "eval_samples_per_second": 4.657,
+      "eval_steps_per_second": 0.596,
       "step": 35
     },
     {
       "epoch": 0.14,
-      "learning_rate": 0.00018099173553719008,
-      "loss": 0.7294,
+      "learning_rate": 0.00017685950413223143,
+      "loss": 0.6323,
       "step": 36
     },
     {
       "epoch": 0.15,
-      "learning_rate": 0.00018016528925619835,
-      "loss": 0.8192,
+      "learning_rate": 0.00017603305785123967,
+      "loss": 0.6918,
       "step": 37
     },
     {
       "epoch": 0.15,
-      "learning_rate": 0.00017933884297520662,
-      "loss": 0.7274,
+      "learning_rate": 0.00017520661157024794,
+      "loss": 0.5309,
       "step": 38
     },
     {
       "epoch": 0.16,
-      "learning_rate": 0.00017851239669421489,
-      "loss": 0.6377,
+      "learning_rate": 0.0001743801652892562,
+      "loss": 0.3449,
       "step": 39
     },
     {
       "epoch": 0.16,
-      "learning_rate": 0.00017851239669421489,
-      "loss": 0.8137,
+      "learning_rate": 0.00017355371900826448,
+      "loss": 0.529,
       "step": 40
     },
     {
       "epoch": 0.16,
-      "eval_loss": 0.8961383104324341,
-      "eval_runtime": 53.7166,
+      "eval_loss": 0.8907827138900757,
+      "eval_runtime": 53.7206,
       "eval_samples_per_second": 4.654,
       "eval_steps_per_second": 0.596,
       "step": 40
     },
     {
       "epoch": 0.16,
-      "learning_rate": 0.00017768595041322316,
-      "loss": 1.0662,
+      "learning_rate": 0.00017272727272727275,
+      "loss": 0.5317,
       "step": 41
     },
     {
       "epoch": 0.17,
-      "learning_rate": 0.00017685950413223143,
-      "loss": 0.9434,
+      "learning_rate": 0.00017190082644628102,
+      "loss": 0.6033,
       "step": 42
     },
     {
       "epoch": 0.17,
-      "learning_rate": 0.00017603305785123967,
-      "loss": 1.4569,
+      "learning_rate": 0.00017107438016528926,
+      "loss": 0.5922,
       "step": 43
     },
     {
       "epoch": 0.18,
-      "learning_rate": 0.00017603305785123967,
-      "loss": 1.4323,
+      "learning_rate": 0.00017024793388429753,
+      "loss": 0.9378,
       "step": 44
     },
     {
       "epoch": 0.18,
-      "learning_rate": 0.00017520661157024794,
-      "loss": 1.3056,
+      "learning_rate": 0.00016942148760330577,
+      "loss": 0.6055,
       "step": 45
     },
     {
       "epoch": 0.18,
-      "eval_loss": 5.5059051513671875,
-      "eval_runtime": 53.5845,
-      "eval_samples_per_second": 4.666,
-      "eval_steps_per_second": 0.597,
+      "eval_loss": 0.8834822773933411,
+      "eval_runtime": 53.6693,
+      "eval_samples_per_second": 4.658,
+      "eval_steps_per_second": 0.596,
       "step": 45
     },
     {
       "epoch": 0.18,
-      "learning_rate": 0.0001743801652892562,
-      "loss": 1.8405,
+      "learning_rate": 0.00016859504132231404,
+      "loss": 0.6727,
       "step": 46
     },
     {
       "epoch": 0.19,
-      "learning_rate": 0.00017355371900826448,
-      "loss": 1.229,
+      "learning_rate": 0.0001677685950413223,
+      "loss": 0.547,
       "step": 47
     },
     {
       "epoch": 0.19,
-      "learning_rate": 0.00017272727272727275,
-      "loss": 0.8605,
+      "learning_rate": 0.00016694214876033058,
+      "loss": 0.552,
       "step": 48
     },
     {
       "epoch": 0.2,
-      "learning_rate": 0.00017190082644628102,
-      "loss": 0.781,
+      "learning_rate": 0.00016611570247933885,
+      "loss": 0.4931,
       "step": 49
     },
     {
       "epoch": 0.2,
-      "learning_rate": 0.00017107438016528926,
-      "loss": 0.9167,
+      "learning_rate": 0.00016528925619834712,
+      "loss": 0.7009,
       "step": 50
     },
     {
       "epoch": 0.2,
-      "eval_loss": 1.111301302909851,
-      "eval_runtime": 53.8007,
-      "eval_samples_per_second": 4.647,
-      "eval_steps_per_second": 0.595,
+      "eval_loss": 0.9503442049026489,
+      "eval_runtime": 53.6719,
+      "eval_samples_per_second": 4.658,
+      "eval_steps_per_second": 0.596,
       "step": 50
     },
     {
       "epoch": 0.2,
-      "learning_rate": 0.00017107438016528926,
-      "loss": 4.9629,
+      "learning_rate": 0.0001644628099173554,
+      "loss": 1.7286,
       "step": 51
     },
     {
       "epoch": 0.21,
-      "learning_rate": 0.00017024793388429753,
-      "loss": 3.8133,
+      "learning_rate": 0.00016363636363636366,
+      "loss": 1.1259,
       "step": 52
     },
     {
       "epoch": 0.21,
-      "learning_rate": 0.00016942148760330577,
-      "loss": 1.5526,
+      "learning_rate": 0.0001628099173553719,
+      "loss": 1.0129,
       "step": 53
     },
     {
       "epoch": 0.22,
-      "learning_rate": 0.00016859504132231404,
-      "loss": 1.3449,
+      "learning_rate": 0.00016198347107438017,
+      "loss": 0.7712,
       "step": 54
     },
     {
       "epoch": 0.22,
-      "learning_rate": 0.0001677685950413223,
-      "loss": 1.3002,
+      "learning_rate": 0.00016115702479338844,
+      "loss": 0.9015,
       "step": 55
     },
     {
       "epoch": 0.22,
-      "eval_loss": 0.8309872150421143,
-      "eval_runtime": 53.6711,
-      "eval_samples_per_second": 4.658,
-      "eval_steps_per_second": 0.596,
+      "eval_loss": 0.7898709177970886,
+      "eval_runtime": 53.7797,
+      "eval_samples_per_second": 4.649,
+      "eval_steps_per_second": 0.595,
       "step": 55
     },
     {
       "epoch": 0.22,
-      "learning_rate": 0.00016694214876033058,
-      "loss": 5.1895,
+      "learning_rate": 0.0001603305785123967,
+      "loss": 0.9435,
       "step": 56
     },
     {
       "epoch": 0.23,
-      "learning_rate": 0.00016611570247933885,
-      "loss": 2.9415,
+      "learning_rate": 0.00015950413223140498,
+      "loss": 0.9349,
       "step": 57
     },
     {
       "epoch": 0.23,
-      "learning_rate": 0.00016528925619834712,
-      "loss": 2.8453,
+      "learning_rate": 0.00015867768595041322,
+      "loss": 1.0034,
       "step": 58
     },
     {
       "epoch": 0.24,
-      "learning_rate": 0.0001644628099173554,
-      "loss": 3.2052,
+      "learning_rate": 0.0001578512396694215,
+      "loss": 0.7622,
       "step": 59
     },
     {
       "epoch": 0.24,
-      "learning_rate": 0.00016363636363636366,
-      "loss": 1.4997,
+      "learning_rate": 0.00015702479338842976,
+      "loss": 0.7044,
       "step": 60
     },
     {
       "epoch": 0.24,
-      "eval_loss": 0.9820745587348938,
-      "eval_runtime": 53.7166,
+      "eval_loss": 0.7550108432769775,
+      "eval_runtime": 53.7223,
       "eval_samples_per_second": 4.654,
       "eval_steps_per_second": 0.596,
       "step": 60
     },
     {
       "epoch": 0.24,
-      "learning_rate": 0.0001628099173553719,
-      "loss": 1.1877,
+      "learning_rate": 0.000156198347107438,
+      "loss": 0.6556,
       "step": 61
     },
     {
       "epoch": 0.25,
-      "learning_rate": 0.00016198347107438017,
-      "loss": 1.1343,
+      "learning_rate": 0.00015537190082644627,
+      "loss": 0.7031,
       "step": 62
     },
     {
       "epoch": 0.25,
-      "learning_rate": 0.00016115702479338844,
-      "loss": 1.606,
+      "learning_rate": 0.00015454545454545454,
+      "loss": 0.8668,
       "step": 63
     },
     {
       "epoch": 0.26,
-      "learning_rate": 0.0001603305785123967,
-      "loss": 1.4458,
+      "learning_rate": 0.00015371900826446281,
+      "loss": 0.7272,
       "step": 64
     },
     {
       "epoch": 0.26,
-      "learning_rate": 0.00015950413223140498,
-      "loss": 1.3983,
+      "learning_rate": 0.00015289256198347108,
+      "loss": 0.6895,
       "step": 65
     },
     {
       "epoch": 0.26,
-      "eval_loss": 1.4925471544265747,
-      "eval_runtime": 53.6349,
-      "eval_samples_per_second": 4.661,
-      "eval_steps_per_second": 0.597,
+      "eval_loss": 0.7527397274971008,
+      "eval_runtime": 53.8077,
+      "eval_samples_per_second": 4.646,
+      "eval_steps_per_second": 0.595,
       "step": 65
     },
     {
       "epoch": 0.26,
-      "learning_rate": 0.00015867768595041322,
-      "loss": 2.0347,
+      "learning_rate": 0.00015206611570247935,
+      "loss": 0.7011,
       "step": 66
     },
     {
       "epoch": 0.27,
-      "learning_rate": 0.0001578512396694215,
-      "loss": 1.783,
+      "learning_rate": 0.00015123966942148762,
+      "loss": 0.8521,
       "step": 67
     },
     {
       "epoch": 0.27,
-      "learning_rate": 0.00015702479338842976,
-      "loss": 1.0935,
+      "learning_rate": 0.0001504132231404959,
+      "loss": 0.5923,
       "step": 68
     },
     {
       "epoch": 0.28,
-      "learning_rate": 0.000156198347107438,
-      "loss": 0.893,
+      "learning_rate": 0.00014958677685950414,
+      "loss": 0.4891,
       "step": 69
     },
     {
       "epoch": 0.28,
-      "learning_rate": 0.00015537190082644627,
-      "loss": 0.8647,
+      "learning_rate": 0.0001487603305785124,
+      "loss": 0.6437,
       "step": 70
     },
     {
       "epoch": 0.28,
-      "eval_loss": 1.0354112386703491,
-      "eval_runtime": 53.718,
-      "eval_samples_per_second": 4.654,
+      "eval_loss": 0.7655532360076904,
+      "eval_runtime": 53.7258,
+      "eval_samples_per_second": 4.653,
       "eval_steps_per_second": 0.596,
       "step": 70
     },
     {
       "epoch": 0.28,
-      "learning_rate": 0.00015454545454545454,
-      "loss": 1.1792,
+      "learning_rate": 0.00014793388429752067,
+      "loss": 0.8369,
       "step": 71
     },
     {
       "epoch": 0.29,
-      "learning_rate": 0.00015371900826446281,
-      "loss": 1.0682,
+      "learning_rate": 0.00014710743801652894,
+      "loss": 0.7403,
       "step": 72
     },
     {
       "epoch": 0.29,
-      "learning_rate": 0.00015289256198347108,
-      "loss": 0.9638,
+      "learning_rate": 0.0001462809917355372,
+      "loss": 0.6789,
       "step": 73
     },
     {
       "epoch": 0.3,
-      "learning_rate": 0.00015206611570247935,
-      "loss": 0.8753,
+      "learning_rate": 0.00014545454545454546,
+      "loss": 0.5445,
       "step": 74
     },
     {
       "epoch": 0.3,
-      "learning_rate": 0.00015123966942148762,
-      "loss": 0.8439,
+      "learning_rate": 0.00014462809917355373,
+      "loss": 0.5606,
       "step": 75
     },
     {
       "epoch": 0.3,
-      "eval_loss": 0.9328717589378357,
-      "eval_runtime": 53.9147,
-      "eval_samples_per_second": 4.637,
-      "eval_steps_per_second": 0.594,
+      "eval_loss": 0.7676512002944946,
+      "eval_runtime": 53.6793,
+      "eval_samples_per_second": 4.657,
+      "eval_steps_per_second": 0.596,
       "step": 75
     },
     {
       "epoch": 0.3,
-      "learning_rate": 0.0001504132231404959,
-      "loss": 0.7914,
+      "learning_rate": 0.000143801652892562,
+      "loss": 0.5263,
       "step": 76
     },
     {
       "epoch": 0.31,
-      "learning_rate": 0.00014958677685950414,
-      "loss": 2.5149,
+      "learning_rate": 0.00014297520661157024,
+      "loss": 0.5318,
       "step": 77
     },
     {
       "epoch": 0.31,
-      "learning_rate": 0.0001487603305785124,
-      "loss": 0.9441,
+      "learning_rate": 0.0001421487603305785,
+      "loss": 0.5521,
       "step": 78
     },
     {
       "epoch": 0.32,
-      "learning_rate": 0.00014793388429752067,
-      "loss": 0.67,
+      "learning_rate": 0.00014132231404958678,
+      "loss": 0.508,
       "step": 79
     },
     {
       "epoch": 0.32,
-      "learning_rate": 0.00014710743801652894,
-      "loss": 0.9213,
+      "learning_rate": 0.00014049586776859505,
+      "loss": 0.5276,
       "step": 80
     },
     {
       "epoch": 0.32,
-      "eval_loss": 0.7875626683235168,
-      "eval_runtime": 53.7442,
-      "eval_samples_per_second": 4.652,
-      "eval_steps_per_second": 0.595,
+      "eval_loss": 0.8956261277198792,
+      "eval_runtime": 53.6781,
+      "eval_samples_per_second": 4.657,
+      "eval_steps_per_second": 0.596,
       "step": 80
     },
     {
       "epoch": 0.32,
-      "learning_rate": 0.0001462809917355372,
-      "loss": 0.6597,
+      "learning_rate": 0.00013966942148760332,
+      "loss": 0.4192,
       "step": 81
     },
     {
       "epoch": 0.33,
-      "learning_rate": 0.00014545454545454546,
-      "loss": 0.7725,
+      "learning_rate": 0.0001388429752066116,
+      "loss": 1.0631,
       "step": 82
     },
     {
       "epoch": 0.33,
-      "learning_rate": 0.00014462809917355373,
-      "loss": 0.6193,
+      "learning_rate": 0.00013801652892561986,
+      "loss": 0.5133,
       "step": 83
     },
     {
       "epoch": 0.34,
-      "learning_rate": 0.000143801652892562,
-      "loss": 2.3156,
+      "learning_rate": 0.00013719008264462813,
+      "loss": 0.5695,
       "step": 84
     },
     {
       "epoch": 0.34,
-      "learning_rate": 0.00014297520661157024,
-      "loss": 0.6892,
+      "learning_rate": 0.00013636363636363637,
+      "loss": 0.4937,
       "step": 85
     },
     {
       "epoch": 0.34,
-      "eval_loss": 0.9291159510612488,
-      "eval_runtime": 53.7672,
-      "eval_samples_per_second": 4.65,
+      "eval_loss": 0.8561215400695801,
+      "eval_runtime": 53.744,
+      "eval_samples_per_second": 4.652,
       "eval_steps_per_second": 0.595,
       "step": 85
     },
     {
       "epoch": 0.34,
-      "learning_rate": 0.0001421487603305785,
-      "loss": 1.5728,
+      "learning_rate": 0.00013553719008264464,
+      "loss": 0.7045,
       "step": 86
     },
     {
       "epoch": 0.35,
-      "learning_rate": 0.00014132231404958678,
-      "loss": 0.9787,
+      "learning_rate": 0.00013471074380165288,
+      "loss": 0.5249,
       "step": 87
     },
     {
       "epoch": 0.35,
-      "learning_rate": 0.00014049586776859505,
-      "loss": 0.7264,
+      "learning_rate": 0.00013388429752066115,
+      "loss": 0.5327,
       "step": 88
     },
     {
       "epoch": 0.36,
-      "learning_rate": 0.00013966942148760332,
-      "loss": 0.8636,
+      "learning_rate": 0.00013305785123966942,
+      "loss": 0.6264,
       "step": 89
     },
     {
       "epoch": 0.36,
-      "learning_rate": 0.0001388429752066116,
-      "loss": 0.8998,
+      "learning_rate": 0.0001322314049586777,
+      "loss": 0.4968,
       "step": 90
     },
     {
       "epoch": 0.36,
-      "eval_loss": 1.0169296264648438,
-      "eval_runtime": 53.7695,
-      "eval_samples_per_second": 4.649,
+      "eval_loss": 0.8243765830993652,
+      "eval_runtime": 53.784,
+      "eval_samples_per_second": 4.648,
       "eval_steps_per_second": 0.595,
       "step": 90
     },
     {
       "epoch": 0.36,
-      "learning_rate": 0.00013801652892561986,
-      "loss": 0.6554,
+      "learning_rate": 0.00013140495867768596,
+      "loss": 0.4186,
       "step": 91
     },
     {
       "epoch": 0.37,
-      "learning_rate": 0.00013719008264462813,
-      "loss": 0.5781,
+      "learning_rate": 0.00013057851239669423,
+      "loss": 0.4288,
       "step": 92
     },
     {
       "epoch": 0.37,
-      "learning_rate": 0.00013636363636363637,
-      "loss": 0.8252,
+      "learning_rate": 0.00012975206611570247,
+      "loss": 0.4699,
       "step": 93
     },
     {
       "epoch": 0.38,
-      "learning_rate": 0.00013553719008264464,
-      "loss": 0.653,
+      "learning_rate": 0.00012892561983471074,
+      "loss": 0.4096,
       "step": 94
     },
     {
       "epoch": 0.38,
-      "learning_rate": 0.00013471074380165288,
-      "loss": 0.6295,
+      "learning_rate": 0.000128099173553719,
+      "loss": 0.3972,
       "step": 95
     },
     {
       "epoch": 0.38,
-      "eval_loss": 0.9279801249504089,
-      "eval_runtime": 53.8354,
-      "eval_samples_per_second": 4.644,
-      "eval_steps_per_second": 0.594,
+      "eval_loss": 0.836593747138977,
+      "eval_runtime": 53.6868,
+      "eval_samples_per_second": 4.657,
+      "eval_steps_per_second": 0.596,
       "step": 95
     },
     {
       "epoch": 0.38,
-      "learning_rate": 0.00013388429752066115,
-      "loss": 0.7676,
+      "learning_rate": 0.00012727272727272728,
+      "loss": 0.5188,
       "step": 96
     },
     {
       "epoch": 0.39,
-      "learning_rate": 0.00013305785123966942,
-      "loss": 0.5702,
+      "learning_rate": 0.00012644628099173555,
+      "loss": 0.3747,
       "step": 97
     },
     {
       "epoch": 0.39,
-      "learning_rate": 0.0001322314049586777,
-      "loss": 0.9051,
+      "learning_rate": 0.00012561983471074382,
+      "loss": 0.5992,
       "step": 98
     },
     {
       "epoch": 0.4,
-      "learning_rate": 0.00013140495867768596,
-      "loss": 0.5604,
+      "learning_rate": 0.0001247933884297521,
+      "loss": 0.3338,
       "step": 99
     },
     {
       "epoch": 0.4,
-      "learning_rate": 0.00013057851239669423,
-      "loss": 0.9297,
+      "learning_rate": 0.00012396694214876033,
+      "loss": 0.8653,
       "step": 100
     },
     {
       "epoch": 0.4,
-      "eval_loss": 0.9151987433433533,
-      "eval_runtime": 53.7965,
-      "eval_samples_per_second": 4.647,
-      "eval_steps_per_second": 0.595,
+      "eval_loss": 0.8624871969223022,
+      "eval_runtime": 53.7207,
+      "eval_samples_per_second": 4.654,
+      "eval_steps_per_second": 0.596,
       "step": 100
     },
     {
       "epoch": 0.4,
-      "learning_rate": 0.00012975206611570247,
-      "loss": 1.6051,
+      "learning_rate": 0.0001231404958677686,
+      "loss": 1.2187,
       "step": 101
     },
     {
       "epoch": 0.41,
-      "learning_rate": 0.00012892561983471074,
-      "loss": 1.4374,
+      "learning_rate": 0.00012231404958677685,
+      "loss": 0.9642,
       "step": 102
     },
     {
       "epoch": 0.41,
-      "learning_rate": 0.000128099173553719,
-      "loss": 1.0713,
+      "learning_rate": 0.00012148760330578513,
+      "loss": 0.7793,
       "step": 103
     },
     {
       "epoch": 0.42,
-      "learning_rate": 0.00012727272727272728,
-      "loss": 1.3361,
+      "learning_rate": 0.0001206611570247934,
+      "loss": 0.899,
       "step": 104
     },
     {
       "epoch": 0.42,
-      "learning_rate": 0.00012644628099173555,
-      "loss": 1.0197,
+      "learning_rate": 0.00011983471074380165,
+      "loss": 0.6753,
       "step": 105
     },
     {
       "epoch": 0.42,
-      "eval_loss": 4.240027904510498,
-      "eval_runtime": 53.6073,
-      "eval_samples_per_second": 4.664,
-      "eval_steps_per_second": 0.597,
+      "eval_loss": 0.7977741956710815,
+      "eval_runtime": 53.7442,
+      "eval_samples_per_second": 4.652,
+      "eval_steps_per_second": 0.595,
       "step": 105
     },
     {
       "epoch": 0.42,
-      "learning_rate": 0.00012644628099173555,
-      "loss": 1.2143,
+      "learning_rate": 0.00011900826446280992,
+      "loss": 1.2854,
       "step": 106
     },
     {
       "epoch": 0.43,
-      "learning_rate": 0.00012561983471074382,
-      "loss": 1.4475,
+      "learning_rate": 0.0001181818181818182,
+      "loss": 0.9619,
       "step": 107
     },
     {
       "epoch": 0.43,
-      "learning_rate": 0.0001247933884297521,
-      "loss": 1.4733,
+      "learning_rate": 0.00011735537190082646,
+      "loss": 0.879,
       "step": 108
     },
     {
       "epoch": 0.44,
-      "learning_rate": 0.00012396694214876033,
-      "loss": 1.6473,
+      "learning_rate": 0.0001165289256198347,
+      "loss": 1.0115,
       "step": 109
     },
     {
       "epoch": 0.44,
-      "learning_rate": 0.0001231404958677686,
-      "loss": 2.8045,
+      "learning_rate": 0.00011570247933884298,
+      "loss": 0.7888,
       "step": 110
     },
     {
       "epoch": 0.44,
-      "eval_loss": 5.459723949432373,
-      "eval_runtime": 53.3894,
-      "eval_samples_per_second": 4.683,
-      "eval_steps_per_second": 0.599,
+      "eval_loss": 0.755509078502655,
+      "eval_runtime": 53.7332,
+      "eval_samples_per_second": 4.653,
+      "eval_steps_per_second": 0.596,
       "step": 110
     },
     {
       "epoch": 0.44,
-      "learning_rate": 0.00012231404958677685,
-      "loss": 1.4439,
+      "learning_rate": 0.00011487603305785125,
+      "loss": 0.6777,
       "step": 111
     },
     {
       "epoch": 0.45,
-      "learning_rate": 0.00012148760330578513,
-      "loss": 2.3906,
+      "learning_rate": 0.0001140495867768595,
+      "loss": 0.876,
       "step": 112
     },
     {
       "epoch": 0.45,
-      "learning_rate": 0.0001206611570247934,
-      "loss": 1.1544,
+      "learning_rate": 0.00011322314049586777,
+      "loss": 0.5992,
       "step": 113
     },
     {
       "epoch": 0.46,
-      "learning_rate": 0.00011983471074380165,
-      "loss": 1.2571,
+      "learning_rate": 0.00011239669421487604,
+      "loss": 0.6918,
       "step": 114
     },
     {
       "epoch": 0.46,
-      "learning_rate": 0.00011900826446280992,
-      "loss": 1.3133,
+      "learning_rate": 0.00011157024793388431,
+      "loss": 0.7215,
       "step": 115
     },
     {
       "epoch": 0.46,
-      "eval_loss": 3.8302040100097656,
-      "eval_runtime": 53.6965,
-      "eval_samples_per_second": 4.656,
+      "eval_loss": 0.781696617603302,
+      "eval_runtime": 53.7115,
+      "eval_samples_per_second": 4.654,
       "eval_steps_per_second": 0.596,
       "step": 115
     },
     {
       "epoch": 0.46,
-      "learning_rate": 0.0001181818181818182,
-      "loss": 1.6594,
+      "learning_rate": 0.00011074380165289258,
+      "loss": 0.728,
       "step": 116
     },
     {
       "epoch": 0.47,
-      "learning_rate": 0.00011735537190082646,
-      "loss": 1.0288,
+      "learning_rate": 0.00010991735537190082,
+      "loss": 0.6974,
       "step": 117
     },
     {
       "epoch": 0.47,
-      "learning_rate": 0.0001165289256198347,
-      "loss": 1.0129,
+      "learning_rate": 0.00010909090909090909,
+      "loss": 0.5451,
       "step": 118
     },
     {
       "epoch": 0.48,
-      "learning_rate": 0.00011570247933884298,
-      "loss": 1.0464,
+      "learning_rate": 0.00010826446280991735,
+      "loss": 0.5469,
       "step": 119
     },
     {
       "epoch": 0.48,
-      "learning_rate": 0.00011487603305785125,
-      "loss": 0.7648,
+      "learning_rate": 0.00010743801652892562,
+      "loss": 0.4957,
       "step": 120
     },
     {
       "epoch": 0.48,
-      "eval_loss": 1.3478542566299438,
-      "eval_runtime": 53.8122,
-      "eval_samples_per_second": 4.646,
-      "eval_steps_per_second": 0.595,
+      "eval_loss": 0.7713479399681091,
+      "eval_runtime": 53.7009,
+      "eval_samples_per_second": 4.655,
+      "eval_steps_per_second": 0.596,
       "step": 120
     },
     {
       "epoch": 0.48,
-      "learning_rate": 0.0001140495867768595,
-      "loss": 0.8218,
+      "learning_rate": 0.00010661157024793389,
+      "loss": 0.5928,
       "step": 121
     },
     {
       "epoch": 0.49,
-      "learning_rate": 0.00011322314049586777,
-      "loss": 0.9217,
+      "learning_rate": 0.00010578512396694216,
+      "loss": 0.6318,
       "step": 122
     },
     {
       "epoch": 0.49,
-      "learning_rate": 0.00011239669421487604,
-      "loss": 0.7243,
+      "learning_rate": 0.00010495867768595043,
+      "loss": 0.4274,
       "step": 123
     },
     {
       "epoch": 0.5,
-      "learning_rate": 0.00011157024793388431,
-      "loss": 0.8348,
+      "learning_rate": 0.0001041322314049587,
+      "loss": 0.6478,
       "step": 124
     },
     {
       "epoch": 0.5,
-      "learning_rate": 0.00011074380165289258,
-      "loss": 0.8907,
+      "learning_rate": 0.00010330578512396694,
+      "loss": 0.5175,
       "step": 125
     },
     {
       "epoch": 0.5,
-      "eval_loss": 1.0467392206192017,
-      "eval_runtime": 53.8255,
-      "eval_samples_per_second": 4.645,
-      "eval_steps_per_second": 0.595,
+      "eval_loss": 0.7830690145492554,
+      "eval_runtime": 53.6903,
+      "eval_samples_per_second": 4.656,
+      "eval_steps_per_second": 0.596,
       "step": 125
     },
     {
       "epoch": 0.5,
-      "learning_rate": 0.00010991735537190082,
-      "loss": 0.8285,
+      "learning_rate": 0.00010247933884297521,
+      "loss": 0.5796,
       "step": 126
     },
     {
       "epoch": 0.51,
-      "learning_rate": 0.00010909090909090909,
-      "loss": 0.661,
+      "learning_rate": 0.00010165289256198347,
+      "loss": 0.5363,
       "step": 127
     },
     {
       "epoch": 0.51,
-      "learning_rate": 0.00010826446280991735,
-      "loss": 0.8155,
+      "learning_rate": 0.00010082644628099174,
+      "loss": 0.3942,
       "step": 128
     },
     {
       "epoch": 0.52,
-      "learning_rate": 0.00010743801652892562,
-      "loss": 0.8824,
+      "learning_rate": 0.0001,
+      "loss": 0.5536,
       "step": 129
     },
     {
       "epoch": 0.52,
-      "learning_rate": 0.00010661157024793389,
-      "loss": 0.7232,
+      "learning_rate": 9.917355371900827e-05,
+      "loss": 0.3535,
       "step": 130
     },
     {
       "epoch": 0.52,
-      "eval_loss": 0.7986066937446594,
-      "eval_runtime": 53.8077,
-      "eval_samples_per_second": 4.646,
-      "eval_steps_per_second": 0.595,
+      "eval_loss": 0.7811296582221985,
+      "eval_runtime": 53.6917,
+      "eval_samples_per_second": 4.656,
+      "eval_steps_per_second": 0.596,
       "step": 130
     },
     {
       "epoch": 0.52,
-      "learning_rate": 0.00010578512396694216,
-      "loss": 0.7196,
+      "learning_rate": 9.834710743801654e-05,
+      "loss": 0.4034,
       "step": 131
     },
     {
       "epoch": 0.53,
-      "learning_rate": 0.00010495867768595043,
-      "loss": 0.6013,
+      "learning_rate": 9.75206611570248e-05,
+      "loss": 0.4504,
       "step": 132
     },
     {
       "epoch": 0.53,
-      "learning_rate": 0.0001041322314049587,
-      "loss": 0.6572,
+      "learning_rate": 9.669421487603306e-05,
+      "loss": 0.3688,
       "step": 133
     },
     {
       "epoch": 0.54,
-      "learning_rate": 0.00010330578512396694,
-      "loss": 0.6946,
+      "learning_rate": 9.586776859504133e-05,
+      "loss": 0.4402,
       "step": 134
     },
     {
       "epoch": 0.54,
-      "learning_rate": 0.00010247933884297521,
-      "loss": 0.6709,
+      "learning_rate": 9.50413223140496e-05,
+      "loss": 0.5129,
       "step": 135
     },
     {
       "epoch": 0.54,
-      "eval_loss": 0.8022701144218445,
-      "eval_runtime": 53.7039,
-      "eval_samples_per_second": 4.655,
+      "eval_loss": 0.7998033761978149,
+      "eval_runtime": 53.6501,
+      "eval_samples_per_second": 4.66,
       "eval_steps_per_second": 0.596,
       "step": 135
     },
     {
       "epoch": 0.54,
-      "learning_rate": 0.00010165289256198347,
-      "loss": 0.4935,
+      "learning_rate": 9.421487603305785e-05,
+      "loss": 0.408,
       "step": 136
     },
     {
       "epoch": 0.55,
-      "learning_rate": 0.00010082644628099174,
-      "loss": 0.6626,
+      "learning_rate": 9.338842975206612e-05,
+      "loss": 0.4771,
       "step": 137
     },
     {
       "epoch": 0.55,
-      "learning_rate": 0.0001,
-      "loss": 1.0252,
+      "learning_rate": 9.256198347107439e-05,
+      "loss": 0.6095,
       "step": 138
     },
     {
       "epoch": 0.56,
-      "learning_rate": 9.917355371900827e-05,
-      "loss": 0.9888,
+      "learning_rate": 9.173553719008266e-05,
+      "loss": 0.6053,
       "step": 139
     },
     {
       "epoch": 0.56,
-      "learning_rate": 9.834710743801654e-05,
-      "loss": 1.0212,
+      "learning_rate": 9.090909090909092e-05,
+      "loss": 0.531,
       "step": 140
     },
     {
       "epoch": 0.56,
-      "eval_loss": 0.863627552986145,
-      "eval_runtime": 53.6581,
-      "eval_samples_per_second": 4.659,
+      "eval_loss": 0.9052188992500305,
+      "eval_runtime": 53.686,
+      "eval_samples_per_second": 4.657,
       "eval_steps_per_second": 0.596,
       "step": 140
     },
     {
       "epoch": 0.56,
-      "learning_rate": 9.75206611570248e-05,
-      "loss": 0.7159,
+      "learning_rate": 9.008264462809917e-05,
+      "loss": 0.4578,
       "step": 141
     },
     {
       "epoch": 0.57,
-      "learning_rate": 9.669421487603306e-05,
-      "loss": 0.578,
+      "learning_rate": 8.925619834710744e-05,
+      "loss": 0.4666,
       "step": 142
     },
     {
       "epoch": 0.57,
-      "learning_rate": 9.586776859504133e-05,
-      "loss": 0.6738,
+      "learning_rate": 8.842975206611571e-05,
+      "loss": 0.4442,
       "step": 143
     },
     {
       "epoch": 0.58,
-      "learning_rate": 9.50413223140496e-05,
-      "loss": 0.965,
+      "learning_rate": 8.760330578512397e-05,
+      "loss": 0.5553,
       "step": 144
     },
     {
       "epoch": 0.58,
-      "learning_rate": 9.421487603305785e-05,
-      "loss": 0.6165,
+      "learning_rate": 8.677685950413224e-05,
+      "loss": 0.4091,
       "step": 145
     },
     {
       "epoch": 0.58,
-      "eval_loss": 0.9543225765228271,
-      "eval_runtime": 53.8162,
-      "eval_samples_per_second": 4.645,
+      "eval_loss": 0.8757880926132202,
+      "eval_runtime": 53.7782,
+      "eval_samples_per_second": 4.649,
       "eval_steps_per_second": 0.595,
       "step": 145
     },
     {
       "epoch": 0.58,
-      "learning_rate": 9.338842975206612e-05,
-      "loss": 0.7398,
+      "learning_rate": 8.595041322314051e-05,
+      "loss": 0.5266,
       "step": 146
     },
     {
       "epoch": 0.59,
-      "learning_rate": 9.256198347107439e-05,
-      "loss": 1.6066,
+      "learning_rate": 8.512396694214876e-05,
+      "loss": 0.717,
       "step": 147
     },
     {
       "epoch": 0.59,
-      "learning_rate": 9.173553719008266e-05,
-      "loss": 0.7194,
+      "learning_rate": 8.429752066115702e-05,
+      "loss": 0.3758,
       "step": 148
     },
     {
       "epoch": 0.6,
-      "learning_rate": 9.090909090909092e-05,
-      "loss": 0.6199,
+      "learning_rate": 8.347107438016529e-05,
+      "loss": 0.4382,
       "step": 149
     },
     {
       "epoch": 0.6,
-      "learning_rate": 9.008264462809917e-05,
-      "loss": 0.6934,
+      "learning_rate": 8.264462809917356e-05,
+      "loss": 0.4992,
       "step": 150
     },
     {
       "epoch": 0.6,
-      "eval_loss": 0.8588433861732483,
-      "eval_runtime": 53.6877,
-      "eval_samples_per_second": 4.657,
+      "eval_loss": 0.8280476927757263,
+      "eval_runtime": 53.6749,
+      "eval_samples_per_second": 4.658,
       "eval_steps_per_second": 0.596,
       "step": 150
     },
     {
       "epoch": 0.6,
-      "learning_rate": 8.925619834710744e-05,
-      "loss": 1.1822,
+      "learning_rate": 8.181818181818183e-05,
+      "loss": 0.9833,
       "step": 151
     },
     {
       "epoch": 0.61,
-      "learning_rate": 8.842975206611571e-05,
-      "loss": 1.1988,
+      "learning_rate": 8.099173553719009e-05,
+      "loss": 1.0995,
       "step": 152
     },
     {
       "epoch": 0.61,
-      "learning_rate": 8.760330578512397e-05,
-      "loss": 0.9174,
+      "learning_rate": 8.016528925619836e-05,
+      "loss": 0.6434,
       "step": 153
     },
     {
       "epoch": 0.62,
-      "learning_rate": 8.677685950413224e-05,
-      "loss": 0.865,
+      "learning_rate": 7.933884297520661e-05,
+      "loss": 0.6236,
       "step": 154
     },
     {
       "epoch": 0.62,
-      "learning_rate": 8.595041322314051e-05,
-      "loss": 0.7286,
+      "learning_rate": 7.851239669421488e-05,
+      "loss": 0.4561,
       "step": 155
     },
     {
       "epoch": 0.62,
-      "eval_loss": 0.8376938104629517,
-      "eval_runtime": 53.7041,
-      "eval_samples_per_second": 4.655,
-      "eval_steps_per_second": 0.596,
+      "eval_loss": 0.7679440379142761,
+      "eval_runtime": 53.5737,
+      "eval_samples_per_second": 4.666,
+      "eval_steps_per_second": 0.597,
       "step": 155
     },
     {
       "epoch": 0.62,
-      "learning_rate": 8.512396694214876e-05,
-      "loss": 0.96,
+      "learning_rate": 7.768595041322314e-05,
+      "loss": 0.689,
       "step": 156
     },
     {
       "epoch": 0.63,
-      "learning_rate": 8.429752066115702e-05,
-      "loss": 1.0063,
+      "learning_rate": 7.685950413223141e-05,
+      "loss": 0.6266,
       "step": 157
     },
     {
       "epoch": 0.63,
-      "learning_rate": 8.347107438016529e-05,
-      "loss": 0.9759,
+      "learning_rate": 7.603305785123968e-05,
+      "loss": 0.4888,
       "step": 158
     },
     {
       "epoch": 0.64,
-      "learning_rate": 8.264462809917356e-05,
-      "loss": 0.9462,
+      "learning_rate": 7.520661157024795e-05,
+      "loss": 0.5136,
       "step": 159
     },
     {
       "epoch": 0.64,
-      "learning_rate": 8.181818181818183e-05,
-      "loss": 0.9358,
+      "learning_rate": 7.43801652892562e-05,
+      "loss": 0.6694,
       "step": 160
     },
     {
       "epoch": 0.64,
-      "eval_loss": 0.8625456094741821,
-      "eval_runtime": 53.7827,
-      "eval_samples_per_second": 4.648,
-      "eval_steps_per_second": 0.595,
+      "eval_loss": 0.7545422315597534,
+      "eval_runtime": 53.6881,
+      "eval_samples_per_second": 4.657,
+      "eval_steps_per_second": 0.596,
       "step": 160
     },
     {
       "epoch": 0.64,
-      "learning_rate": 8.099173553719009e-05,
-      "loss": 0.8784,
+      "learning_rate": 7.355371900826447e-05,
+      "loss": 0.7067,
       "step": 161
     },
     {
       "epoch": 0.65,
-      "learning_rate": 8.016528925619836e-05,
-      "loss": 0.7743,
+      "learning_rate": 7.272727272727273e-05,
+      "loss": 0.4744,
       "step": 162
     },
     {
       "epoch": 0.65,
-      "learning_rate": 7.933884297520661e-05,
-      "loss": 0.8533,
+      "learning_rate": 7.1900826446281e-05,
+      "loss": 0.5205,
       "step": 163
     },
     {
       "epoch": 0.66,
-      "learning_rate": 7.851239669421488e-05,
-      "loss": 0.7121,
+      "learning_rate": 7.107438016528925e-05,
+      "loss": 0.5893,
       "step": 164
     },
     {
       "epoch": 0.66,
-      "learning_rate": 7.768595041322314e-05,
-      "loss": 0.9879,
+      "learning_rate": 7.024793388429752e-05,
+      "loss": 0.6063,
       "step": 165
     },
     {
       "epoch": 0.66,
-      "eval_loss": 0.8257442712783813,
-      "eval_runtime": 53.633,
-      "eval_samples_per_second": 4.661,
-      "eval_steps_per_second": 0.597,
+      "eval_loss": 0.7204523682594299,
+      "eval_runtime": 53.6716,
+      "eval_samples_per_second": 4.658,
+      "eval_steps_per_second": 0.596,
       "step": 165
     },
     {
       "epoch": 0.66,
-      "learning_rate": 7.685950413223141e-05,
-      "loss": 0.751,
+      "learning_rate": 6.94214876033058e-05,
+      "loss": 0.5689,
       "step": 166
     },
     {
       "epoch": 0.67,
-      "learning_rate": 7.603305785123968e-05,
-      "loss": 0.697,
+      "learning_rate": 6.859504132231406e-05,
+      "loss": 0.4719,
       "step": 167
     },
     {
       "epoch": 0.67,
-      "learning_rate": 7.520661157024795e-05,
-      "loss": 0.8949,
+      "learning_rate": 6.776859504132232e-05,
+      "loss": 0.4556,
       "step": 168
     },
     {
       "epoch": 0.68,
-      "learning_rate": 7.43801652892562e-05,
-      "loss": 0.7368,
+      "learning_rate": 6.694214876033058e-05,
+      "loss": 0.398,
       "step": 169
     },
     {
       "epoch": 0.68,
-      "learning_rate": 7.355371900826447e-05,
-      "loss": 0.8358,
+      "learning_rate": 6.611570247933885e-05,
+      "loss": 0.4753,
       "step": 170
     },
     {
       "epoch": 0.68,
-      "eval_loss": 0.7519776821136475,
-      "eval_runtime": 53.7106,
-      "eval_samples_per_second": 4.655,
+      "eval_loss": 0.768771231174469,
+      "eval_runtime": 53.6543,
+      "eval_samples_per_second": 4.659,
       "eval_steps_per_second": 0.596,
       "step": 170
     },
     {
       "epoch": 0.68,
-      "learning_rate": 7.272727272727273e-05,
-      "loss": 0.7448,
+      "learning_rate": 6.528925619834711e-05,
+      "loss": 0.5509,
       "step": 171
     },
     {
       "epoch": 0.69,
-      "learning_rate": 7.1900826446281e-05,
-      "loss": 0.7534,
+      "learning_rate": 6.446280991735537e-05,
+      "loss": 0.4325,
       "step": 172
     },
     {
       "epoch": 0.69,
-      "learning_rate": 7.107438016528925e-05,
-      "loss": 0.6142,
+      "learning_rate": 6.363636363636364e-05,
+      "loss": 0.3852,
       "step": 173
     },
     {
       "epoch": 0.7,
-      "learning_rate": 7.024793388429752e-05,
-      "loss": 0.6749,
+      "learning_rate": 6.280991735537191e-05,
+      "loss": 0.4855,
       "step": 174
     },
     {
       "epoch": 0.7,
-      "learning_rate": 6.94214876033058e-05,
-      "loss": 0.8757,
+      "learning_rate": 6.198347107438017e-05,
+      "loss": 0.7439,
       "step": 175
     },
     {
       "epoch": 0.7,
-      "eval_loss": 0.7619721293449402,
-      "eval_runtime": 53.8101,
-      "eval_samples_per_second": 4.646,
-      "eval_steps_per_second": 0.595,
+      "eval_loss": 0.7854070067405701,
+      "eval_runtime": 53.656,
+      "eval_samples_per_second": 4.659,
+      "eval_steps_per_second": 0.596,
       "step": 175
     },
     {
       "epoch": 0.7,
-      "learning_rate": 6.859504132231406e-05,
-      "loss": 0.8771,
+      "learning_rate": 6.115702479338842e-05,
+      "loss": 0.7033,
       "step": 176
     },
     {
       "epoch": 0.71,
-      "learning_rate": 6.776859504132232e-05,
-      "loss": 0.7568,
+      "learning_rate": 6.03305785123967e-05,
+      "loss": 0.6599,
       "step": 177
     },
     {
       "epoch": 0.71,
-      "learning_rate": 6.694214876033058e-05,
-      "loss": 0.5086,
+      "learning_rate": 5.950413223140496e-05,
+      "loss": 0.4345,
       "step": 178
     },
     {
       "epoch": 0.72,
-      "learning_rate": 6.611570247933885e-05,
-      "loss": 0.6883,
+      "learning_rate": 5.867768595041323e-05,
+      "loss": 0.4249,
       "step": 179
     },
     {
       "epoch": 0.72,
-      "learning_rate": 6.528925619834711e-05,
-      "loss": 0.6506,
+      "learning_rate": 5.785123966942149e-05,
+      "loss": 0.5298,
       "step": 180
     },
     {
       "epoch": 0.72,
-      "eval_loss": 0.7380254864692688,
-      "eval_runtime": 53.6926,
-      "eval_samples_per_second": 4.656,
-      "eval_steps_per_second": 0.596,
+      "eval_loss": 0.7402228713035583,
+      "eval_runtime": 53.6144,
+      "eval_samples_per_second": 4.663,
+      "eval_steps_per_second": 0.597,
       "step": 180
     },
     {
       "epoch": 0.72,
-      "learning_rate": 6.446280991735537e-05,
-      "loss": 0.832,
+      "learning_rate": 5.702479338842975e-05,
+      "loss": 0.527,
       "step": 181
     },
     {
       "epoch": 0.73,
-      "learning_rate": 6.363636363636364e-05,
-      "loss": 0.5172,
+      "learning_rate": 5.619834710743802e-05,
+      "loss": 0.4173,
       "step": 182
     },
     {
       "epoch": 0.73,
-      "learning_rate": 6.280991735537191e-05,
-      "loss": 0.5754,
+      "learning_rate": 5.537190082644629e-05,
+      "loss": 0.3894,
       "step": 183
     },
     {
       "epoch": 0.74,
-      "learning_rate": 6.198347107438017e-05,
-      "loss": 0.5498,
+      "learning_rate": 5.4545454545454546e-05,
+      "loss": 0.5203,
       "step": 184
     },
     {
       "epoch": 0.74,
-      "learning_rate": 6.115702479338842e-05,
-      "loss": 0.7684,
+      "learning_rate": 5.371900826446281e-05,
+      "loss": 0.5605,
       "step": 185
     },
     {
       "epoch": 0.74,
-      "eval_loss": 0.7799345254898071,
-      "eval_runtime": 53.7341,
-      "eval_samples_per_second": 4.653,
-      "eval_steps_per_second": 0.596,
+      "eval_loss": 0.7166436314582825,
+      "eval_runtime": 53.5756,
+      "eval_samples_per_second": 4.666,
+      "eval_steps_per_second": 0.597,
       "step": 185
     },
     {
       "epoch": 0.74,
-      "learning_rate": 6.03305785123967e-05,
-      "loss": 0.5429,
+      "learning_rate": 5.289256198347108e-05,
+      "loss": 0.4387,
       "step": 186
     },
     {
       "epoch": 0.75,
-      "learning_rate": 5.950413223140496e-05,
-      "loss": 0.4125,
+      "learning_rate": 5.206611570247935e-05,
+      "loss": 0.2797,
       "step": 187
     },
     {
       "epoch": 0.75,
-      "learning_rate": 5.867768595041323e-05,
-      "loss": 0.5034,
+      "learning_rate": 5.1239669421487605e-05,
+      "loss": 0.3081,
       "step": 188
     },
     {
       "epoch": 0.76,
-      "learning_rate": 5.785123966942149e-05,
-      "loss": 0.7265,
+      "learning_rate": 5.041322314049587e-05,
+      "loss": 0.4857,
       "step": 189
     },
     {
       "epoch": 0.76,
-      "learning_rate": 5.702479338842975e-05,
-      "loss": 0.6579,
+      "learning_rate": 4.958677685950414e-05,
+      "loss": 0.4996,
       "step": 190
     },
     {
       "epoch": 0.76,
-      "eval_loss": 0.7613645792007446,
-      "eval_runtime": 53.7419,
-      "eval_samples_per_second": 4.652,
-      "eval_steps_per_second": 0.595,
+      "eval_loss": 0.7330756783485413,
+      "eval_runtime": 53.5984,
+      "eval_samples_per_second": 4.664,
+      "eval_steps_per_second": 0.597,
       "step": 190
     },
     {
       "epoch": 0.76,
-      "learning_rate": 5.619834710743802e-05,
-      "loss": 0.6208,
+      "learning_rate": 4.87603305785124e-05,
+      "loss": 0.428,
       "step": 191
     },
     {
       "epoch": 0.77,
-      "learning_rate": 5.537190082644629e-05,
-      "loss": 0.5012,
+      "learning_rate": 4.793388429752066e-05,
+      "loss": 0.3813,
       "step": 192
     },
     {
       "epoch": 0.77,
-      "learning_rate": 5.4545454545454546e-05,
-      "loss": 0.6489,
+      "learning_rate": 4.7107438016528926e-05,
+      "loss": 0.5674,
       "step": 193
     },
     {
       "epoch": 0.78,
-      "learning_rate": 5.371900826446281e-05,
-      "loss": 2.2796,
+      "learning_rate": 4.6280991735537196e-05,
+      "loss": 0.5904,
       "step": 194
     },
     {
       "epoch": 0.78,
-      "learning_rate": 5.289256198347108e-05,
-      "loss": 0.6158,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": 0.395,
       "step": 195
     },
     {
       "epoch": 0.78,
-      "eval_loss": 0.7108221054077148,
-      "eval_runtime": 53.7335,
-      "eval_samples_per_second": 4.653,
-      "eval_steps_per_second": 0.596,
+      "eval_loss": 0.7636235356330872,
+      "eval_runtime": 53.6407,
+      "eval_samples_per_second": 4.661,
+      "eval_steps_per_second": 0.597,
       "step": 195
     },
     {
       "epoch": 0.78,
-      "learning_rate": 5.206611570247935e-05,
-      "loss": 0.6332,
+      "learning_rate": 4.462809917355372e-05,
+      "loss": 0.415,
       "step": 196
     },
     {
       "epoch": 0.79,
-      "learning_rate": 5.1239669421487605e-05,
-      "loss": 0.467,
+      "learning_rate": 4.3801652892561984e-05,
+      "loss": 0.3077,
       "step": 197
     },
     {
       "epoch": 0.79,
-      "learning_rate": 5.041322314049587e-05,
-      "loss": 0.5199,
+      "learning_rate": 4.2975206611570254e-05,
+      "loss": 0.4011,
       "step": 198
     },
     {
       "epoch": 0.8,
-      "learning_rate": 4.958677685950414e-05,
-      "loss": 0.4777,
+      "learning_rate": 4.214876033057851e-05,
+      "loss": 0.3334,
       "step": 199
     },
     {
       "epoch": 0.8,
-      "learning_rate": 4.87603305785124e-05,
-      "loss": 0.6286,
+      "learning_rate": 4.132231404958678e-05,
+      "loss": 0.4171,
       "step": 200
     },
     {
       "epoch": 0.8,
-      "eval_loss": 0.7811810374259949,
-      "eval_runtime": 53.7947,
-      "eval_samples_per_second": 4.647,
-      "eval_steps_per_second": 0.595,
+      "eval_loss": 0.7737492322921753,
+      "eval_runtime": 53.6156,
+      "eval_samples_per_second": 4.663,
+      "eval_steps_per_second": 0.597,
       "step": 200
     }
   ],