Training in progress, step 300, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b7a8bb8a4ae15ac03afd44a3577747a17cf8ae073c8e0a12f668e60bf4269c9
+oid sha256:fdfa536891b8ffc83f63e9ace650cf6e9f915075b64c129dc410645e45a1688a
 size 847599616

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d51750dfe21f1524b4e65e279067be3a9cf4264ff3dc3e8c306e71ffc8db8015
-size 1386414411
+oid sha256:ba78434b69e3d278adef8b3e6500747e43a121a034e99f8a3ff4d1bde8a22b60
+size 350603

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c587bc9603de883e44d0b68949a09abf17c1a2e4c3bc62449c91a13da9d5d454
+oid sha256:9079206c3c25e14fa2f8a19a24a32c2e5380266de4ccca26cb728acd92fae5e8
 size 14645

last-checkpoint/trainer_state.json

@@ -11,440 +11,440 @@
   "log_history": [
     {
       "epoch": 0.001,
-      "grad_norm": 2.5244109630584717,
+      "grad_norm": 0.27235767245292664,
       "learning_rate": 6.4e-08,
-      "loss": 10.941880798339843,
+      "loss": 10.786966705322266,
       "step": 5
     },
     {
       "epoch": 0.002,
-      "grad_norm": 2.4931907653808594,
+      "grad_norm": 0.36470845341682434,
       "learning_rate": 1.44e-07,
-      "loss": 10.940663146972657,
+      "loss": 10.004191589355468,
       "step": 10
     },
     {
       "epoch": 0.003,
-      "grad_norm": 2.397547721862793,
+      "grad_norm": 0.18794603645801544,
       "learning_rate": 2.24e-07,
-      "loss": 10.932537841796876,
+      "loss": 9.357711791992188,
       "step": 15
     },
     {
       "epoch": 0.004,
-      "grad_norm": 2.362260103225708,
+      "grad_norm": 0.15133747458457947,
       "learning_rate": 3.0399999999999997e-07,
-      "loss": 10.920089721679688,
+      "loss": 9.077755737304688,
       "step": 20
     },
     {
       "epoch": 0.005,
-      "grad_norm": 2.5390405654907227,
+      "grad_norm": 0.13880157470703125,
       "learning_rate": 3.84e-07,
-      "loss": 10.902597045898437,
+      "loss": 8.826399993896484,
       "step": 25
     },
     {
       "epoch": 0.006,
-      "grad_norm": 2.4956705570220947,
+      "grad_norm": 0.12539362907409668,
       "learning_rate": 4.64e-07,
-      "loss": 10.886011505126953,
+      "loss": 8.724310302734375,
       "step": 30
     },
     {
       "epoch": 0.007,
-      "grad_norm": 2.5255372524261475,
+      "grad_norm": 0.11178825795650482,
       "learning_rate": 5.44e-07,
-      "loss": 10.858057403564453,
+      "loss": 8.595573425292969,
       "step": 35
     },
     {
       "epoch": 0.008,
-      "grad_norm": 2.4960005283355713,
+      "grad_norm": 0.16203832626342773,
       "learning_rate": 6.24e-07,
-      "loss": 10.832856750488281,
+      "loss": 8.613536834716797,
       "step": 40
     },
     {
       "epoch": 0.009,
-      "grad_norm": 2.6297495365142822,
+      "grad_norm": 0.11954408138990402,
       "learning_rate": 7.04e-07,
-      "loss": 10.798202514648438,
+      "loss": 8.435651397705078,
       "step": 45
     },
     {
       "epoch": 0.01,
-      "grad_norm": 2.4866788387298584,
+      "grad_norm": 0.09399693459272385,
       "learning_rate": 7.84e-07,
-      "loss": 10.764683532714844,
+      "loss": 8.369234466552735,
       "step": 50
     },
     {
       "epoch": 0.011,
-      "grad_norm": 2.472733736038208,
+      "grad_norm": 0.08723891526460648,
       "learning_rate": 8.639999999999999e-07,
-      "loss": 10.72347640991211,
+      "loss": 8.335379791259765,
       "step": 55
     },
     {
       "epoch": 0.012,
-      "grad_norm": 2.432396650314331,
+      "grad_norm": 0.09454073011875153,
       "learning_rate": 9.439999999999999e-07,
-      "loss": 10.679368591308593,
+      "loss": 8.253787231445312,
       "step": 60
     },
     {
       "epoch": 0.013,
-      "grad_norm": 2.444960355758667,
+      "grad_norm": 0.09540031850337982,
       "learning_rate": 1.024e-06,
-      "loss": 10.622978973388673,
+      "loss": 8.186421966552734,
       "step": 65
     },
     {
       "epoch": 0.014,
-      "grad_norm": 2.3638627529144287,
+      "grad_norm": 0.11653102189302444,
       "learning_rate": 1.1040000000000001e-06,
-      "loss": 10.573637390136719,
+      "loss": 8.132817840576172,
       "step": 70
     },
     {
       "epoch": 0.015,
-      "grad_norm": 2.4232065677642822,
+      "grad_norm": 0.08094990253448486,
       "learning_rate": 1.1839999999999998e-06,
-      "loss": 10.517623901367188,
+      "loss": 8.107611846923827,
       "step": 75
     },
     {
       "epoch": 0.016,
-      "grad_norm": 2.4094903469085693,
+      "grad_norm": 0.10092103481292725,
       "learning_rate": 1.2639999999999999e-06,
-      "loss": 10.454130554199219,
+      "loss": 8.071889495849609,
       "step": 80
     },
     {
       "epoch": 0.017,
-      "grad_norm": 2.3315868377685547,
+      "grad_norm": 0.09236462414264679,
       "learning_rate": 1.344e-06,
-      "loss": 10.414918518066406,
+      "loss": 8.15390625,
       "step": 85
     },
     {
       "epoch": 0.018,
-      "grad_norm": 2.5305004119873047,
+      "grad_norm": 0.08973367512226105,
       "learning_rate": 1.4239999999999998e-06,
-      "loss": 10.341883087158203,
+      "loss": 8.073948669433594,
       "step": 90
     },
     {
       "epoch": 0.019,
-      "grad_norm": 2.3920392990112305,
+      "grad_norm": 0.08046946674585342,
       "learning_rate": 1.504e-06,
-      "loss": 10.270912170410156,
+      "loss": 7.942198944091797,
       "step": 95
     },
     {
       "epoch": 0.02,
-      "grad_norm": 2.678619861602783,
+      "grad_norm": 0.07841245085000992,
       "learning_rate": 1.584e-06,
-      "loss": 10.162237548828125,
+      "loss": 7.894329071044922,
       "step": 100
     },
     {
       "epoch": 0.021,
-      "grad_norm": 2.3744475841522217,
+      "grad_norm": 0.07948298752307892,
       "learning_rate": 1.6639999999999999e-06,
-      "loss": 10.103695678710938,
+      "loss": 7.908601379394531,
       "step": 105
     },
     {
       "epoch": 0.022,
-      "grad_norm": 2.334637403488159,
+      "grad_norm": 0.07720845937728882,
       "learning_rate": 1.744e-06,
-      "loss": 10.062642669677734,
+      "loss": 7.923919677734375,
       "step": 110
     },
     {
       "epoch": 0.023,
-      "grad_norm": 2.4162437915802,
+      "grad_norm": 0.09289873391389847,
       "learning_rate": 1.824e-06,
-      "loss": 10.005216979980469,
+      "loss": 7.866429138183594,
       "step": 115
     },
     {
       "epoch": 0.024,
-      "grad_norm": 2.38637113571167,
+      "grad_norm": 0.07957543432712555,
       "learning_rate": 1.904e-06,
-      "loss": 9.93200454711914,
+      "loss": 7.828727722167969,
       "step": 120
     },
     {
       "epoch": 0.025,
-      "grad_norm": 2.2896499633789062,
+      "grad_norm": 0.07522212713956833,
       "learning_rate": 1.984e-06,
-      "loss": 9.873931121826171,
+      "loss": 7.83117446899414,
       "step": 125
     },
     {
       "epoch": 0.026,
-      "grad_norm": 2.198643684387207,
+      "grad_norm": 0.09532520920038223,
       "learning_rate": 2.064e-06,
-      "loss": 9.849259185791016,
+      "loss": 7.8716987609863285,
       "step": 130
     },
     {
       "epoch": 0.027,
-      "grad_norm": 2.191349983215332,
+      "grad_norm": 0.07470740377902985,
       "learning_rate": 2.144e-06,
-      "loss": 9.790320587158202,
+      "loss": 7.834799194335938,
       "step": 135
     },
     {
       "epoch": 0.028,
-      "grad_norm": 2.097088575363159,
+      "grad_norm": 0.07192815095186234,
       "learning_rate": 2.2240000000000002e-06,
-      "loss": 9.746687316894532,
+      "loss": 7.736669921875,
       "step": 140
     },
     {
       "epoch": 0.029,
-      "grad_norm": 2.1079649925231934,
+      "grad_norm": 0.09201541543006897,
       "learning_rate": 2.304e-06,
-      "loss": 9.693339538574218,
+      "loss": 7.6827842712402346,
       "step": 145
     },
     {
       "epoch": 0.03,
-      "grad_norm": 2.0016579627990723,
+      "grad_norm": 0.09225732833147049,
       "learning_rate": 2.384e-06,
-      "loss": 9.655043792724609,
+      "loss": 7.745516967773438,
       "step": 150
     },
     {
       "epoch": 0.03,
-      "eval_accuracy": 0.0701001221001221,
-      "eval_loss": 9.63985824584961,
-      "eval_runtime": 10.5292,
-      "eval_samples_per_second": 9.497,
-      "eval_steps_per_second": 1.615,
+      "eval_accuracy": 0.11922344322344322,
+      "eval_loss": 7.651349067687988,
+      "eval_runtime": 10.4261,
+      "eval_samples_per_second": 9.591,
+      "eval_steps_per_second": 1.631,
       "step": 150
     },
     {
       "epoch": 0.031,
-      "grad_norm": 1.9648433923721313,
+      "grad_norm": 0.08881130814552307,
       "learning_rate": 2.464e-06,
-      "loss": 9.62866439819336,
+      "loss": 7.674343872070312,
       "step": 155
     },
     {
       "epoch": 0.032,
-      "grad_norm": 2.0246658325195312,
+      "grad_norm": 0.07704972475767136,
       "learning_rate": 2.544e-06,
-      "loss": 9.588599395751952,
+      "loss": 7.681053161621094,
       "step": 160
     },
     {
       "epoch": 0.033,
-      "grad_norm": 1.941605806350708,
+      "grad_norm": 0.09186960011720657,
       "learning_rate": 2.624e-06,
-      "loss": 9.559522247314453,
+      "loss": 7.690374755859375,
       "step": 165
     },
     {
       "epoch": 0.034,
-      "grad_norm": 2.1081161499023438,
+      "grad_norm": 0.08565036207437515,
       "learning_rate": 2.704e-06,
-      "loss": 9.523811340332031,
+      "loss": 7.603900146484375,
       "step": 170
     },
     {
       "epoch": 0.035,
-      "grad_norm": 1.9597957134246826,
+      "grad_norm": 0.1499108374118805,
       "learning_rate": 2.7839999999999995e-06,
-      "loss": 9.42935791015625,
+      "loss": 7.505958557128906,
       "step": 175
     },
     {
       "epoch": 0.036,
-      "grad_norm": 2.00820255279541,
+      "grad_norm": 0.06643059849739075,
       "learning_rate": 2.8639999999999996e-06,
-      "loss": 9.449478149414062,
+      "loss": 7.573754119873047,
       "step": 180
     },
     {
       "epoch": 0.037,
-      "grad_norm": 2.1553680896759033,
+      "grad_norm": 0.11821448802947998,
       "learning_rate": 2.9439999999999997e-06,
-      "loss": 9.507718658447265,
+      "loss": 7.752062225341797,
       "step": 185
     },
     {
       "epoch": 0.038,
-      "grad_norm": 2.015054225921631,
+      "grad_norm": 0.08827198296785355,
       "learning_rate": 3.0239999999999998e-06,
-      "loss": 9.442604064941406,
+      "loss": 7.5669189453125,
       "step": 190
     },
     {
       "epoch": 0.039,
-      "grad_norm": 1.9352600574493408,
+      "grad_norm": 0.07519616931676865,
       "learning_rate": 3.104e-06,
-      "loss": 9.40188980102539,
+      "loss": 7.571305847167968,
       "step": 195
     },
     {
       "epoch": 0.04,
-      "grad_norm": 1.9175664186477661,
+      "grad_norm": 0.075102798640728,
       "learning_rate": 3.184e-06,
-      "loss": 9.372711944580079,
+      "loss": 7.518638610839844,
       "step": 200
     },
     {
       "epoch": 0.041,
-      "grad_norm": 2.047138214111328,
+      "grad_norm": 0.10685888677835464,
       "learning_rate": 3.2639999999999996e-06,
-      "loss": 9.297854614257812,
+      "loss": 7.512000274658203,
       "step": 205
     },
     {
       "epoch": 0.042,
-      "grad_norm": 1.7625023126602173,
+      "grad_norm": 0.10541684180498123,
       "learning_rate": 3.3439999999999997e-06,
-      "loss": 9.354965209960938,
+      "loss": 7.564778137207031,
       "step": 210
     },
     {
       "epoch": 0.043,
-      "grad_norm": 1.9414070844650269,
+      "grad_norm": 0.09155019372701645,
       "learning_rate": 3.4239999999999997e-06,
-      "loss": 9.30640106201172,
+      "loss": 7.4674217224121096,
       "step": 215
     },
     {
       "epoch": 0.044,
-      "grad_norm": 1.8544589281082153,
+      "grad_norm": 0.11373615264892578,
       "learning_rate": 3.504e-06,
-      "loss": 9.286300659179688,
+      "loss": 7.450457000732422,
       "step": 220
     },
     {
       "epoch": 0.045,
-      "grad_norm": 1.864773154258728,
+      "grad_norm": 0.07444982975721359,
       "learning_rate": 3.584e-06,
-      "loss": 9.244451141357422,
+      "loss": 7.397786712646484,
       "step": 225
     },
     {
       "epoch": 0.046,
-      "grad_norm": 1.9134947061538696,
+      "grad_norm": 0.08052903413772583,
       "learning_rate": 3.664e-06,
-      "loss": 9.235625457763671,
+      "loss": 7.415688323974609,
       "step": 230
     },
     {
       "epoch": 0.047,
-      "grad_norm": 1.959318995475769,
+      "grad_norm": 0.08429361879825592,
       "learning_rate": 3.744e-06,
-      "loss": 9.202496337890626,
+      "loss": 7.411205291748047,
       "step": 235
     },
     {
       "epoch": 0.048,
-      "grad_norm": 1.792734146118164,
+      "grad_norm": 0.08501375466585159,
       "learning_rate": 3.823999999999999e-06,
-      "loss": 9.204552459716798,
+      "loss": 7.406707763671875,
       "step": 240
     },
     {
       "epoch": 0.049,
-      "grad_norm": 1.8003584146499634,
+      "grad_norm": 0.10457716882228851,
       "learning_rate": 3.903999999999999e-06,
-      "loss": 9.14830322265625,
+      "loss": 7.312602233886719,
       "step": 245
     },
     {
       "epoch": 0.05,
-      "grad_norm": 1.941810131072998,
+      "grad_norm": 0.09624126553535461,
       "learning_rate": 3.9839999999999995e-06,
-      "loss": 9.14439697265625,
+      "loss": 7.360851287841797,
       "step": 250
     },
     {
       "epoch": 0.051,
-      "grad_norm": 1.7709863185882568,
+      "grad_norm": 0.08660672605037689,
       "learning_rate": 3.99999300106024e-06,
-      "loss": 9.078700256347656,
+      "loss": 7.353035736083984,
       "step": 255
     },
     {
       "epoch": 0.052,
-      "grad_norm": 1.771018385887146,
+      "grad_norm": 0.08430126309394836,
       "learning_rate": 3.9999645679514235e-06,
-      "loss": 9.125299072265625,
+      "loss": 7.384159088134766,
       "step": 260
     },
     {
       "epoch": 0.053,
-      "grad_norm": 1.6554689407348633,
+      "grad_norm": 0.07826503366231918,
       "learning_rate": 3.999914263550513e-06,
-      "loss": 9.08728485107422,
+      "loss": 7.4195198059082035,
       "step": 265
     },
     {
       "epoch": 0.054,
-      "grad_norm": 1.889380931854248,
+      "grad_norm": 0.08577796816825867,
       "learning_rate": 3.9998420884076325e-06,
-      "loss": 9.034518432617187,
+      "loss": 7.362513732910156,
       "step": 270
     },
     {
       "epoch": 0.055,
-      "grad_norm": 1.7578399181365967,
+      "grad_norm": 0.0649741068482399,
       "learning_rate": 3.999748043312075e-06,
-      "loss": 9.05845947265625,
+      "loss": 7.275961303710938,
       "step": 275
     },
     {
       "epoch": 0.056,
-      "grad_norm": 1.7057148218154907,
+      "grad_norm": 0.07203389704227448,
       "learning_rate": 3.999632129292304e-06,
-      "loss": 9.022759246826173,
+      "loss": 7.263797760009766,
       "step": 280
     },
     {
       "epoch": 0.057,
-      "grad_norm": 1.7159546613693237,
+      "grad_norm": 0.12281067669391632,
       "learning_rate": 3.9994943476159364e-06,
-      "loss": 8.9812255859375,
+      "loss": 7.287098693847656,
       "step": 285
     },
     {
       "epoch": 0.058,
-      "grad_norm": 1.7239329814910889,
+      "grad_norm": 0.07184985280036926,
       "learning_rate": 3.999334699789731e-06,
-      "loss": 8.976101684570313,
+      "loss": 7.240232849121094,
       "step": 290
     },
     {
       "epoch": 0.059,
-      "grad_norm": 1.73225736618042,
+      "grad_norm": 0.07799801975488663,
       "learning_rate": 3.99915318755957e-06,
-      "loss": 8.98138885498047,
+      "loss": 7.2740531921386715,
       "step": 295
     },
     {
       "epoch": 0.06,
-      "grad_norm": 1.7068036794662476,
+      "grad_norm": 0.0761650875210762,
       "learning_rate": 3.9989498129104425e-06,
-      "loss": 8.94490966796875,
+      "loss": 7.299461364746094,
       "step": 300
     },
     {
       "epoch": 0.06,
-      "eval_accuracy": 0.08137484737484738,
-      "eval_loss": 8.956033706665039,
-      "eval_runtime": 10.504,
-      "eval_samples_per_second": 9.52,
-      "eval_steps_per_second": 1.618,
+      "eval_accuracy": 0.12382905982905983,
+      "eval_loss": 7.217157363891602,
+      "eval_runtime": 10.4259,
+      "eval_samples_per_second": 9.591,
+      "eval_steps_per_second": 1.631,
       "step": 300
     }
   ],