Upload trainer_state.json with huggingface_hub

trainer_state.json

@@ -10,153 +10,153 @@
   "log_history": [
     {
       "epoch": 0.008247422680412371,
-      "grad_norm": 16.375,
+      "grad_norm": 16.75,
       "learning_rate": 6.666666666666667e-06,
-      "loss": 0.8858,
+      "loss": 0.8829,
       "num_input_tokens_seen": 1413808,
       "step": 4
     },
     {
       "epoch": 0.016494845360824743,
-      "grad_norm": 4.1875,
+      "grad_norm": 3.90625,
       "learning_rate": 1.3333333333333333e-05,
-      "loss": 0.4088,
+      "loss": 0.4033,
       "num_input_tokens_seen": 2866496,
       "step": 8
     },
     {
       "epoch": 0.024742268041237112,
-      "grad_norm": 3.03125,
+      "grad_norm": 5.0,
       "learning_rate": 2e-05,
-      "loss": 0.2731,
+      "loss": 0.2755,
       "num_input_tokens_seen": 4305104,
       "step": 12
     },
     {
       "epoch": 0.032989690721649485,
-      "grad_norm": 1.6484375,
+      "grad_norm": 2.453125,
       "learning_rate": 2.6666666666666667e-05,
-      "loss": 0.2563,
+      "loss": 0.2531,
       "num_input_tokens_seen": 5594128,
       "step": 16
     },
     {
       "epoch": 0.041237113402061855,
-      "grad_norm": 1.46875,
+      "grad_norm": 1.484375,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.2536,
+      "loss": 0.2585,
       "num_input_tokens_seen": 6683376,
       "step": 20
     },
     {
       "epoch": 0.049484536082474224,
-      "grad_norm": 1.2421875,
+      "grad_norm": 1.328125,
       "learning_rate": 4e-05,
-      "loss": 0.229,
+      "loss": 0.2308,
       "num_input_tokens_seen": 8030336,
       "step": 24
     },
     {
       "epoch": 0.0577319587628866,
-      "grad_norm": 1.2578125,
+      "grad_norm": 1.3671875,
       "learning_rate": 3.9992569962849926e-05,
-      "loss": 0.2212,
+      "loss": 0.2221,
       "num_input_tokens_seen": 9395728,
       "step": 28
     },
     {
       "epoch": 0.06597938144329897,
-      "grad_norm": 1.3671875,
+      "grad_norm": 1.3125,
       "learning_rate": 3.99702853719449e-05,
-      "loss": 0.2275,
+      "loss": 0.2259,
       "num_input_tokens_seen": 10689344,
       "step": 32
     },
     {
       "epoch": 0.07422680412371134,
-      "grad_norm": 1.328125,
+      "grad_norm": 1.234375,
       "learning_rate": 3.9933162784818745e-05,
-      "loss": 0.2262,
+      "loss": 0.2201,
       "num_input_tokens_seen": 11936704,
       "step": 36
     },
     {
       "epoch": 0.08247422680412371,
-      "grad_norm": 1.484375,
+      "grad_norm": 1.359375,
       "learning_rate": 3.988122978369162e-05,
-      "loss": 0.2254,
+      "loss": 0.2242,
       "num_input_tokens_seen": 13217248,
       "step": 40
     },
     {
       "epoch": 0.09072164948453608,
-      "grad_norm": 1.15625,
+      "grad_norm": 1.203125,
       "learning_rate": 3.981452495497628e-05,
-      "loss": 0.2186,
+      "loss": 0.2213,
       "num_input_tokens_seen": 14587328,
       "step": 44
     },
     {
       "epoch": 0.09896907216494845,
-      "grad_norm": 1.109375,
+      "grad_norm": 1.0859375,
       "learning_rate": 3.973309786060829e-05,
-      "loss": 0.1971,
+      "loss": 0.1958,
       "num_input_tokens_seen": 15976464,
       "step": 48
     },
     {
       "epoch": 0.10721649484536082,
-      "grad_norm": 1.125,
+      "grad_norm": 1.09375,
       "learning_rate": 3.963700900122124e-05,
-      "loss": 0.2231,
+      "loss": 0.2136,
       "num_input_tokens_seen": 17262576,
       "step": 52
     },
     {
       "epoch": 0.1154639175257732,
-      "grad_norm": 0.9765625,
+      "grad_norm": 1.03125,
       "learning_rate": 3.952632977119465e-05,
-      "loss": 0.2029,
+      "loss": 0.2059,
       "num_input_tokens_seen": 18801264,
       "step": 56
     },
     {
       "epoch": 0.12371134020618557,
-      "grad_norm": 1.0625,
+      "grad_norm": 1.0859375,
       "learning_rate": 3.9401142405607594e-05,
-      "loss": 0.2033,
+      "loss": 0.197,
       "num_input_tokens_seen": 20158000,
       "step": 60
     },
     {
       "epoch": 0.13195876288659794,
-      "grad_norm": 1.09375,
+      "grad_norm": 1.2109375,
       "learning_rate": 3.9261539919137776e-05,
-      "loss": 0.2278,
+      "loss": 0.2273,
       "num_input_tokens_seen": 21322240,
       "step": 64
     },
     {
       "epoch": 0.1402061855670103,
-      "grad_norm": 1.0,
+      "grad_norm": 0.98828125,
       "learning_rate": 3.9107626036951266e-05,
-      "loss": 0.1998,
+      "loss": 0.1971,
       "num_input_tokens_seen": 22631360,
       "step": 68
     },
     {
       "epoch": 0.14845360824742268,
-      "grad_norm": 1.109375,
+      "grad_norm": 1.125,
       "learning_rate": 3.8939515117634326e-05,
-      "loss": 0.2148,
+      "loss": 0.2194,
       "num_input_tokens_seen": 23848496,
       "step": 72
     },
     {
       "epoch": 0.15670103092783505,
-      "grad_norm": 1.1484375,
+      "grad_norm": 1.1953125,
       "learning_rate": 3.875733206822452e-05,
-      "loss": 0.2246,
+      "loss": 0.2215,
       "num_input_tokens_seen": 25148336,
       "step": 76
     },
@@ -164,842 +164,842 @@
       "epoch": 0.16494845360824742,
       "grad_norm": 1.203125,
       "learning_rate": 3.8561212251404406e-05,
-      "loss": 0.2056,
+      "loss": 0.1989,
       "num_input_tokens_seen": 26427264,
       "step": 80
     },
     {
       "epoch": 0.1731958762886598,
-      "grad_norm": 1.1328125,
+      "grad_norm": 1.203125,
       "learning_rate": 3.835130138492644e-05,
-      "loss": 0.203,
+      "loss": 0.2072,
       "num_input_tokens_seen": 27833024,
       "step": 84
     },
     {
       "epoch": 0.18144329896907216,
-      "grad_norm": 1.1015625,
+      "grad_norm": 1.1171875,
       "learning_rate": 3.812775543334425e-05,
-      "loss": 0.1912,
+      "loss": 0.2013,
       "num_input_tokens_seen": 29273008,
       "step": 88
     },
     {
       "epoch": 0.18969072164948453,
-      "grad_norm": 1.2265625,
+      "grad_norm": 1.1484375,
       "learning_rate": 3.789074049213033e-05,
-      "loss": 0.2182,
-      "num_input_tokens_seen": 30624112,
+      "loss": 0.2119,
+      "num_input_tokens_seen": 30628416,
       "step": 92
     },
     {
       "epoch": 0.1979381443298969,
-      "grad_norm": 1.1796875,
+      "grad_norm": 1.1328125,
       "learning_rate": 3.7640432664266514e-05,
-      "loss": 0.216,
-      "num_input_tokens_seen": 31857552,
+      "loss": 0.2213,
+      "num_input_tokens_seen": 31861856,
       "step": 96
     },
     {
       "epoch": 0.20618556701030927,
-      "grad_norm": 1.125,
+      "grad_norm": 1.0859375,
       "learning_rate": 3.737701792939881e-05,
-      "loss": 0.2065,
-      "num_input_tokens_seen": 33116768,
+      "loss": 0.2102,
+      "num_input_tokens_seen": 33121072,
       "step": 100
     },
     {
       "epoch": 0.21443298969072164,
-      "grad_norm": 1.0,
+      "grad_norm": 1.0859375,
       "learning_rate": 3.7100692005653796e-05,
-      "loss": 0.206,
-      "num_input_tokens_seen": 34461024,
+      "loss": 0.2052,
+      "num_input_tokens_seen": 34464560,
       "step": 104
     },
     {
       "epoch": 0.22268041237113403,
-      "grad_norm": 1.0703125,
+      "grad_norm": 1.09375,
       "learning_rate": 3.681166020421938e-05,
-      "loss": 0.1912,
-      "num_input_tokens_seen": 35915264,
+      "loss": 0.1942,
+      "num_input_tokens_seen": 35918800,
       "step": 108
     },
     {
       "epoch": 0.2309278350515464,
-      "grad_norm": 1.0625,
+      "grad_norm": 1.0546875,
       "learning_rate": 3.6510137276797786e-05,
-      "loss": 0.1952,
-      "num_input_tokens_seen": 37264080,
+      "loss": 0.1946,
+      "num_input_tokens_seen": 37267616,
       "step": 112
     },
     {
       "epoch": 0.23917525773195877,
-      "grad_norm": 1.109375,
+      "grad_norm": 1.078125,
       "learning_rate": 3.6196347256044236e-05,
-      "loss": 0.2273,
-      "num_input_tokens_seen": 38539072,
+      "loss": 0.2263,
+      "num_input_tokens_seen": 38542608,
       "step": 116
     },
     {
       "epoch": 0.24742268041237114,
-      "grad_norm": 1.109375,
+      "grad_norm": 1.015625,
       "learning_rate": 3.5870523289109886e-05,
-      "loss": 0.2041,
-      "num_input_tokens_seen": 39930480,
+      "loss": 0.2102,
+      "num_input_tokens_seen": 39934016,
       "step": 120
     },
     {
       "epoch": 0.2556701030927835,
-      "grad_norm": 1.15625,
+      "grad_norm": 1.0390625,
       "learning_rate": 3.553290746441261e-05,
-      "loss": 0.2065,
-      "num_input_tokens_seen": 41066544,
+      "loss": 0.2084,
+      "num_input_tokens_seen": 41070080,
       "step": 124
     },
     {
       "epoch": 0.2639175257731959,
       "grad_norm": 1.0390625,
       "learning_rate": 3.5183750631764406e-05,
-      "loss": 0.1979,
-      "num_input_tokens_seen": 42372160,
+      "loss": 0.1939,
+      "num_input_tokens_seen": 42375696,
       "step": 128
     },
     {
       "epoch": 0.2721649484536082,
-      "grad_norm": 1.0703125,
+      "grad_norm": 1.0390625,
       "learning_rate": 3.4823312215989046e-05,
-      "loss": 0.2079,
-      "num_input_tokens_seen": 43644832,
+      "loss": 0.2027,
+      "num_input_tokens_seen": 43648368,
       "step": 132
     },
     {
       "epoch": 0.2804123711340206,
-      "grad_norm": 1.0078125,
+      "grad_norm": 1.03125,
       "learning_rate": 3.445186002416849e-05,
-      "loss": 0.2058,
-      "num_input_tokens_seen": 44948816,
+      "loss": 0.2093,
+      "num_input_tokens_seen": 44952352,
       "step": 136
     },
     {
       "epoch": 0.28865979381443296,
-      "grad_norm": 1.09375,
+      "grad_norm": 1.1484375,
       "learning_rate": 3.4069670046661197e-05,
-      "loss": 0.1857,
-      "num_input_tokens_seen": 46404048,
+      "loss": 0.1887,
+      "num_input_tokens_seen": 46407584,
       "step": 140
     },
     {
       "epoch": 0.29690721649484536,
       "grad_norm": 1.1015625,
       "learning_rate": 3.3677026252040306e-05,
-      "loss": 0.212,
-      "num_input_tokens_seen": 47646208,
+      "loss": 0.2109,
+      "num_input_tokens_seen": 47649744,
       "step": 144
     },
     {
       "epoch": 0.30515463917525776,
-      "grad_norm": 1.0234375,
+      "grad_norm": 0.98046875,
       "learning_rate": 3.327422037610389e-05,
-      "loss": 0.1983,
-      "num_input_tokens_seen": 49010928,
+      "loss": 0.2072,
+      "num_input_tokens_seen": 49014464,
       "step": 148
     },
     {
       "epoch": 0.3134020618556701,
-      "grad_norm": 0.9375,
+      "grad_norm": 0.97265625,
       "learning_rate": 3.286155170511419e-05,
-      "loss": 0.197,
-      "num_input_tokens_seen": 50440128,
+      "loss": 0.2046,
+      "num_input_tokens_seen": 50443616,
       "step": 152
     },
     {
       "epoch": 0.3216494845360825,
-      "grad_norm": 1.140625,
+      "grad_norm": 1.1328125,
       "learning_rate": 3.2439326853426824e-05,
-      "loss": 0.2028,
-      "num_input_tokens_seen": 51797840,
+      "loss": 0.211,
+      "num_input_tokens_seen": 51801328,
       "step": 156
     },
     {
       "epoch": 0.32989690721649484,
-      "grad_norm": 0.984375,
+      "grad_norm": 0.9765625,
       "learning_rate": 3.200785953567517e-05,
-      "loss": 0.196,
-      "num_input_tokens_seen": 53109456,
+      "loss": 0.1977,
+      "num_input_tokens_seen": 53112944,
       "step": 160
     },
     {
       "epoch": 0.33814432989690724,
-      "grad_norm": 1.0390625,
+      "grad_norm": 1.03125,
       "learning_rate": 3.156747033367922e-05,
-      "loss": 0.2016,
-      "num_input_tokens_seen": 54440768,
+      "loss": 0.2001,
+      "num_input_tokens_seen": 54444256,
       "step": 164
     },
     {
       "epoch": 0.3463917525773196,
-      "grad_norm": 0.97265625,
+      "grad_norm": 1.046875,
       "learning_rate": 3.1118486458252094e-05,
-      "loss": 0.1975,
-      "num_input_tokens_seen": 55879424,
+      "loss": 0.2,
+      "num_input_tokens_seen": 55882912,
       "step": 168
     },
     {
       "epoch": 0.354639175257732,
       "grad_norm": 1.0234375,
       "learning_rate": 3.0661241506081236e-05,
-      "loss": 0.1965,
-      "num_input_tokens_seen": 57154384,
+      "loss": 0.1997,
+      "num_input_tokens_seen": 57157872,
       "step": 172
     },
     {
       "epoch": 0.3628865979381443,
-      "grad_norm": 0.95703125,
+      "grad_norm": 1.0078125,
       "learning_rate": 3.019607521186475e-05,
-      "loss": 0.2078,
-      "num_input_tokens_seen": 58470672,
+      "loss": 0.2085,
+      "num_input_tokens_seen": 58474160,
       "step": 176
     },
     {
       "epoch": 0.3711340206185567,
-      "grad_norm": 1.0625,
+      "grad_norm": 1.0859375,
       "learning_rate": 2.972333319588736e-05,
-      "loss": 0.2092,
-      "num_input_tokens_seen": 59684416,
+      "loss": 0.2093,
+      "num_input_tokens_seen": 59687904,
       "step": 180
     },
     {
       "epoch": 0.37938144329896906,
-      "grad_norm": 0.96484375,
+      "grad_norm": 0.99609375,
       "learning_rate": 2.9243366707223165e-05,
-      "loss": 0.2018,
-      "num_input_tokens_seen": 61002832,
+      "loss": 0.1963,
+      "num_input_tokens_seen": 61006320,
       "step": 184
     },
     {
       "epoch": 0.38762886597938145,
-      "grad_norm": 1.0546875,
+      "grad_norm": 0.98828125,
       "learning_rate": 2.875653236275632e-05,
-      "loss": 0.2072,
-      "num_input_tokens_seen": 62262064,
+      "loss": 0.2001,
+      "num_input_tokens_seen": 62265552,
       "step": 188
     },
     {
       "epoch": 0.3958762886597938,
-      "grad_norm": 0.92578125,
+      "grad_norm": 0.9609375,
       "learning_rate": 2.8263191882213362e-05,
-      "loss": 0.1936,
-      "num_input_tokens_seen": 63678896,
+      "loss": 0.1948,
+      "num_input_tokens_seen": 63682384,
       "step": 192
     },
     {
       "epoch": 0.4041237113402062,
-      "grad_norm": 0.9609375,
+      "grad_norm": 0.953125,
       "learning_rate": 2.7763711819404098e-05,
-      "loss": 0.2069,
-      "num_input_tokens_seen": 64844672,
+      "loss": 0.2048,
+      "num_input_tokens_seen": 64848160,
       "step": 196
     },
     {
       "epoch": 0.41237113402061853,
-      "grad_norm": 1.0546875,
+      "grad_norm": 0.98046875,
       "learning_rate": 2.7258463289870764e-05,
-      "loss": 0.1924,
-      "num_input_tokens_seen": 66274544,
+      "loss": 0.192,
+      "num_input_tokens_seen": 66278032,
       "step": 200
     },
     {
       "epoch": 0.42061855670103093,
-      "grad_norm": 0.88671875,
+      "grad_norm": 0.80859375,
       "learning_rate": 2.6747821695147806e-05,
-      "loss": 0.1949,
-      "num_input_tokens_seen": 67683072,
+      "loss": 0.1933,
+      "num_input_tokens_seen": 67686560,
       "step": 204
     },
     {
       "epoch": 0.4288659793814433,
-      "grad_norm": 1.125,
+      "grad_norm": 1.0703125,
       "learning_rate": 2.623216644383715e-05,
-      "loss": 0.2092,
-      "num_input_tokens_seen": 68860288,
+      "loss": 0.2094,
+      "num_input_tokens_seen": 68863776,
       "step": 208
     },
     {
       "epoch": 0.43711340206185567,
-      "grad_norm": 1.078125,
+      "grad_norm": 1.0859375,
       "learning_rate": 2.5711880669706172e-05,
-      "loss": 0.1959,
-      "num_input_tokens_seen": 70182736,
+      "loss": 0.1964,
+      "num_input_tokens_seen": 70186224,
       "step": 212
     },
     {
       "epoch": 0.44536082474226807,
-      "grad_norm": 0.9296875,
+      "grad_norm": 0.9765625,
       "learning_rate": 2.5187350947017918e-05,
-      "loss": 0.2101,
-      "num_input_tokens_seen": 71494624,
+      "loss": 0.2042,
+      "num_input_tokens_seen": 71498112,
       "step": 216
     },
     {
       "epoch": 0.4536082474226804,
-      "grad_norm": 0.86328125,
+      "grad_norm": 0.89453125,
       "learning_rate": 2.4658967003304986e-05,
-      "loss": 0.1925,
-      "num_input_tokens_seen": 72877248,
+      "loss": 0.1908,
+      "num_input_tokens_seen": 72880736,
       "step": 220
     },
     {
       "epoch": 0.4618556701030928,
-      "grad_norm": 1.0078125,
+      "grad_norm": 1.0,
       "learning_rate": 2.4127121429800498e-05,
-      "loss": 0.1841,
-      "num_input_tokens_seen": 74118560,
+      "loss": 0.187,
+      "num_input_tokens_seen": 74122048,
       "step": 224
     },
     {
       "epoch": 0.47010309278350515,
-      "grad_norm": 0.88671875,
+      "grad_norm": 0.9296875,
       "learning_rate": 2.3592209389741372e-05,
-      "loss": 0.174,
-      "num_input_tokens_seen": 75598912,
+      "loss": 0.1778,
+      "num_input_tokens_seen": 75602400,
       "step": 228
     },
     {
       "epoch": 0.47835051546391755,
-      "grad_norm": 1.0234375,
+      "grad_norm": 1.0546875,
       "learning_rate": 2.30546283247606e-05,
       "loss": 0.207,
-      "num_input_tokens_seen": 76742752,
+      "num_input_tokens_seen": 76746240,
       "step": 232
     },
     {
       "epoch": 0.4865979381443299,
-      "grad_norm": 1.015625,
+      "grad_norm": 0.984375,
       "learning_rate": 2.251477765958655e-05,
-      "loss": 0.1932,
-      "num_input_tokens_seen": 78206256,
+      "loss": 0.1911,
+      "num_input_tokens_seen": 78209744,
       "step": 236
     },
     {
       "epoch": 0.4948453608247423,
-      "grad_norm": 1.0546875,
+      "grad_norm": 1.125,
       "learning_rate": 2.1973058505269007e-05,
-      "loss": 0.1946,
-      "num_input_tokens_seen": 79491408,
+      "loss": 0.1935,
+      "num_input_tokens_seen": 79494896,
       "step": 240
     },
     {
       "epoch": 0.5030927835051546,
       "grad_norm": 1.0546875,
       "learning_rate": 2.1429873361152124e-05,
-      "loss": 0.1975,
-      "num_input_tokens_seen": 80718320,
+      "loss": 0.1977,
+      "num_input_tokens_seen": 80721808,
       "step": 244
     },
     {
       "epoch": 0.511340206185567,
-      "grad_norm": 0.91015625,
+      "grad_norm": 0.96875,
       "learning_rate": 2.088562581581592e-05,
-      "loss": 0.1964,
-      "num_input_tokens_seen": 81915456,
+      "loss": 0.1956,
+      "num_input_tokens_seen": 81918944,
       "step": 248
     },
     {
       "epoch": 0.5195876288659794,
-      "grad_norm": 1.1015625,
+      "grad_norm": 1.125,
       "learning_rate": 2.0340720247208447e-05,
-      "loss": 0.191,
-      "num_input_tokens_seen": 83180624,
+      "loss": 0.1912,
+      "num_input_tokens_seen": 83184064,
       "step": 252
     },
     {
       "epoch": 0.5278350515463918,
-      "grad_norm": 0.90234375,
+      "grad_norm": 1.03125,
       "learning_rate": 1.9795561522191523e-05,
-      "loss": 0.1832,
-      "num_input_tokens_seen": 84571536,
+      "loss": 0.1843,
+      "num_input_tokens_seen": 84574976,
       "step": 256
     },
     {
       "epoch": 0.5360824742268041,
-      "grad_norm": 1.046875,
+      "grad_norm": 1.0078125,
       "learning_rate": 1.9250554695723107e-05,
-      "loss": 0.1964,
-      "num_input_tokens_seen": 85841328,
+      "loss": 0.1942,
+      "num_input_tokens_seen": 85844768,
       "step": 260
     },
     {
       "epoch": 0.5443298969072164,
-      "grad_norm": 0.953125,
+      "grad_norm": 0.9765625,
       "learning_rate": 1.8706104709899964e-05,
-      "loss": 0.1875,
-      "num_input_tokens_seen": 87241616,
+      "loss": 0.1922,
+      "num_input_tokens_seen": 87245056,
       "step": 264
     },
     {
       "epoch": 0.5525773195876289,
-      "grad_norm": 0.92578125,
+      "grad_norm": 1.0234375,
       "learning_rate": 1.816261609308419e-05,
-      "loss": 0.1809,
-      "num_input_tokens_seen": 88600352,
+      "loss": 0.182,
+      "num_input_tokens_seen": 88603792,
       "step": 268
     },
     {
       "epoch": 0.5608247422680412,
-      "grad_norm": 0.87109375,
+      "grad_norm": 0.90625,
       "learning_rate": 1.7620492659337155e-05,
-      "loss": 0.1793,
-      "num_input_tokens_seen": 90051376,
+      "loss": 0.1879,
+      "num_input_tokens_seen": 90054816,
       "step": 272
     },
     {
       "epoch": 0.5690721649484536,
-      "grad_norm": 1.0390625,
+      "grad_norm": 0.9453125,
       "learning_rate": 1.7080137208384122e-05,
-      "loss": 0.1865,
-      "num_input_tokens_seen": 91429472,
+      "loss": 0.1809,
+      "num_input_tokens_seen": 91432912,
       "step": 276
     },
     {
       "epoch": 0.5773195876288659,
-      "grad_norm": 0.9140625,
+      "grad_norm": 0.921875,
       "learning_rate": 1.6541951226332565e-05,
-      "loss": 0.1745,
-      "num_input_tokens_seen": 92791856,
+      "loss": 0.1735,
+      "num_input_tokens_seen": 92795296,
       "step": 280
     },
     {
       "epoch": 0.5855670103092784,
-      "grad_norm": 0.875,
+      "grad_norm": 0.9296875,
       "learning_rate": 1.600633458736653e-05,
-      "loss": 0.1925,
-      "num_input_tokens_seen": 94068304,
+      "loss": 0.1915,
+      "num_input_tokens_seen": 94071744,
       "step": 284
     },
     {
       "epoch": 0.5938144329896907,
-      "grad_norm": 0.98828125,
+      "grad_norm": 0.94921875,
       "learning_rate": 1.5473685256638572e-05,
-      "loss": 0.1903,
-      "num_input_tokens_seen": 95338656,
+      "loss": 0.1895,
+      "num_input_tokens_seen": 95342096,
       "step": 288
     },
     {
       "epoch": 0.6020618556701031,
-      "grad_norm": 0.90625,
+      "grad_norm": 0.98828125,
       "learning_rate": 1.4944398994580232e-05,
-      "loss": 0.1834,
-      "num_input_tokens_seen": 96565872,
+      "loss": 0.1869,
+      "num_input_tokens_seen": 96569312,
       "step": 292
     },
     {
       "epoch": 0.6103092783505155,
-      "grad_norm": 1.0,
+      "grad_norm": 1.03125,
       "learning_rate": 1.4418869062850514e-05,
-      "loss": 0.211,
-      "num_input_tokens_seen": 97845776,
+      "loss": 0.2004,
+      "num_input_tokens_seen": 97849216,
       "step": 296
     },
     {
       "epoch": 0.6185567010309279,
-      "grad_norm": 0.92578125,
+      "grad_norm": 0.93359375,
       "learning_rate": 1.3897485932141042e-05,
-      "loss": 0.1872,
-      "num_input_tokens_seen": 99080048,
+      "loss": 0.1865,
+      "num_input_tokens_seen": 99083488,
       "step": 300
     },
     {
       "epoch": 0.6268041237113402,
-      "grad_norm": 0.88671875,
+      "grad_norm": 0.91796875,
       "learning_rate": 1.3380636992054878e-05,
-      "loss": 0.17,
-      "num_input_tokens_seen": 100563184,
+      "loss": 0.1769,
+      "num_input_tokens_seen": 100566624,
       "step": 304
     },
     {
       "epoch": 0.6350515463917525,
-      "grad_norm": 0.9375,
+      "grad_norm": 0.86328125,
       "learning_rate": 1.2868706263274602e-05,
-      "loss": 0.1935,
-      "num_input_tokens_seen": 101820432,
+      "loss": 0.1969,
+      "num_input_tokens_seen": 101823872,
       "step": 308
     },
     {
       "epoch": 0.643298969072165,
-      "grad_norm": 0.88671875,
+      "grad_norm": 0.96484375,
       "learning_rate": 1.236207411223353e-05,
-      "loss": 0.1833,
-      "num_input_tokens_seen": 103280736,
+      "loss": 0.1767,
+      "num_input_tokens_seen": 103284176,
       "step": 312
     },
     {
       "epoch": 0.6515463917525773,
-      "grad_norm": 0.88671875,
+      "grad_norm": 0.83984375,
       "learning_rate": 1.1861116968502015e-05,
-      "loss": 0.1815,
-      "num_input_tokens_seen": 104563920,
+      "loss": 0.1799,
+      "num_input_tokens_seen": 104567360,
       "step": 316
     },
     {
       "epoch": 0.6597938144329897,
-      "grad_norm": 0.875,
+      "grad_norm": 0.92578125,
       "learning_rate": 1.136620704509892e-05,
-      "loss": 0.1816,
-      "num_input_tokens_seen": 105869408,
+      "loss": 0.1856,
+      "num_input_tokens_seen": 105872848,
       "step": 320
     },
     {
       "epoch": 0.668041237113402,
-      "grad_norm": 0.9296875,
+      "grad_norm": 0.86328125,
       "learning_rate": 1.087771206193593e-05,
-      "loss": 0.1837,
-      "num_input_tokens_seen": 107213792,
+      "loss": 0.1791,
+      "num_input_tokens_seen": 107217232,
       "step": 324
     },
     {
       "epoch": 0.6762886597938145,
-      "grad_norm": 0.96484375,
+      "grad_norm": 0.9609375,
       "learning_rate": 1.0395994972600285e-05,
-      "loss": 0.1775,
-      "num_input_tokens_seen": 108623536,
+      "loss": 0.1806,
+      "num_input_tokens_seen": 108626976,
       "step": 328
     },
     {
       "epoch": 0.6845360824742268,
-      "grad_norm": 0.97265625,
+      "grad_norm": 0.9609375,
       "learning_rate": 9.921413694678959e-06,
-      "loss": 0.2035,
-      "num_input_tokens_seen": 109750560,
+      "loss": 0.2018,
+      "num_input_tokens_seen": 109754000,
       "step": 332
     },
     {
       "epoch": 0.6927835051546392,
-      "grad_norm": 0.921875,
+      "grad_norm": 0.94921875,
       "learning_rate": 9.454320843824512e-06,
-      "loss": 0.1862,
-      "num_input_tokens_seen": 111023152,
+      "loss": 0.1848,
+      "num_input_tokens_seen": 111026592,
       "step": 336
     },
     {
       "epoch": 0.7010309278350515,
-      "grad_norm": 0.95703125,
+      "grad_norm": 0.96484375,
       "learning_rate": 8.995063471760377e-06,
-      "loss": 0.1927,
-      "num_input_tokens_seen": 112284320,
+      "loss": 0.1885,
+      "num_input_tokens_seen": 112287760,
       "step": 340
     },
     {
       "epoch": 0.709278350515464,
-      "grad_norm": 0.96484375,
+      "grad_norm": 0.98828125,
       "learning_rate": 8.543982808420156e-06,
-      "loss": 0.1856,
-      "num_input_tokens_seen": 113630688,
+      "loss": 0.1838,
+      "num_input_tokens_seen": 113634128,
       "step": 344
     },
     {
       "epoch": 0.7175257731958763,
-      "grad_norm": 0.953125,
+      "grad_norm": 0.984375,
       "learning_rate": 8.101414008412469e-06,
-      "loss": 0.1792,
-      "num_input_tokens_seen": 114946320,
+      "loss": 0.1842,
+      "num_input_tokens_seen": 114949760,
       "step": 348
     },
     {
       "epoch": 0.7257731958762886,
-      "grad_norm": 0.94921875,
+      "grad_norm": 0.96484375,
       "learning_rate": 7.667685901999875e-06,
-      "loss": 0.1891,
-      "num_input_tokens_seen": 116220208,
+      "loss": 0.1935,
+      "num_input_tokens_seen": 116223648,
       "step": 352
     },
     {
       "epoch": 0.734020618556701,
-      "grad_norm": 0.8671875,
+      "grad_norm": 0.86328125,
       "learning_rate": 7.24312075077674e-06,
-      "loss": 0.1891,
-      "num_input_tokens_seen": 117614672,
+      "loss": 0.1866,
+      "num_input_tokens_seen": 117618112,
       "step": 356
     },
     {
       "epoch": 0.7422680412371134,
-      "grad_norm": 1.0859375,
+      "grad_norm": 0.875,
       "learning_rate": 6.828034008227678e-06,
-      "loss": 0.1714,
-      "num_input_tokens_seen": 118996816,
+      "loss": 0.1751,
+      "num_input_tokens_seen": 119000256,
       "step": 360
     },
     {
       "epoch": 0.7505154639175258,
-      "grad_norm": 0.90234375,
+      "grad_norm": 0.9765625,
       "learning_rate": 6.422734085344464e-06,
-      "loss": 0.1871,
-      "num_input_tokens_seen": 120229232,
+      "loss": 0.1796,
+      "num_input_tokens_seen": 120232672,
       "step": 364
     },
     {
       "epoch": 0.7587628865979381,
-      "grad_norm": 0.8203125,
+      "grad_norm": 0.83984375,
       "learning_rate": 6.027522121475482e-06,
-      "loss": 0.1795,
-      "num_input_tokens_seen": 121495936,
+      "loss": 0.1783,
+      "num_input_tokens_seen": 121499376,
       "step": 368
     },
     {
       "epoch": 0.7670103092783506,
       "grad_norm": 0.95703125,
       "learning_rate": 5.642691760578116e-06,
-      "loss": 0.1833,
-      "num_input_tokens_seen": 122787872,
+      "loss": 0.1856,
+      "num_input_tokens_seen": 122791312,
       "step": 372
     },
     {
       "epoch": 0.7752577319587629,
-      "grad_norm": 0.83984375,
+      "grad_norm": 0.85546875,
       "learning_rate": 5.268528933040147e-06,
-      "loss": 0.1673,
-      "num_input_tokens_seen": 124257600,
+      "loss": 0.1674,
+      "num_input_tokens_seen": 124261040,
       "step": 376
     },
     {
       "epoch": 0.7835051546391752,
-      "grad_norm": 0.875,
+      "grad_norm": 0.82421875,
       "learning_rate": 4.905311643232464e-06,
-      "loss": 0.1763,
-      "num_input_tokens_seen": 125705408,
+      "loss": 0.1773,
+      "num_input_tokens_seen": 125708848,
       "step": 380
     },
     {
       "epoch": 0.7917525773195876,
-      "grad_norm": 0.9765625,
+      "grad_norm": 0.9375,
       "learning_rate": 4.553309762950739e-06,
-      "loss": 0.1877,
-      "num_input_tokens_seen": 126862272,
+      "loss": 0.1905,
+      "num_input_tokens_seen": 126865712,
       "step": 384
     },
     {
       "epoch": 0.8,
-      "grad_norm": 0.90625,
+      "grad_norm": 0.99609375,
       "learning_rate": 4.212784830899725e-06,
-      "loss": 0.1795,
-      "num_input_tokens_seen": 128153600,
+      "loss": 0.1793,
+      "num_input_tokens_seen": 128157040,
       "step": 388
     },
     {
       "epoch": 0.8082474226804124,
-      "grad_norm": 0.828125,
+      "grad_norm": 0.84375,
       "learning_rate": 3.8839898583689725e-06,
-      "loss": 0.1803,
-      "num_input_tokens_seen": 129461872,
+      "loss": 0.1812,
+      "num_input_tokens_seen": 129465312,
       "step": 392
     },
     {
       "epoch": 0.8164948453608247,
-      "grad_norm": 0.88671875,
+      "grad_norm": 0.890625,
       "learning_rate": 3.567169141244562e-06,
-      "loss": 0.179,
-      "num_input_tokens_seen": 130662064,
+      "loss": 0.1813,
+      "num_input_tokens_seen": 130665504,
       "step": 396
     },
     {
       "epoch": 0.8247422680412371,
-      "grad_norm": 0.8359375,
+      "grad_norm": 0.92578125,
       "learning_rate": 3.262558078496301e-06,
-      "loss": 0.1679,
-      "num_input_tokens_seen": 131997568,
+      "loss": 0.1727,
+      "num_input_tokens_seen": 132002896,
       "step": 400
     },
     {
       "epoch": 0.8329896907216495,
-      "grad_norm": 0.94921875,
+      "grad_norm": 0.8828125,
       "learning_rate": 2.9703829972754407e-06,
-      "loss": 0.1974,
-      "num_input_tokens_seen": 133415744,
+      "loss": 0.1858,
+      "num_input_tokens_seen": 133415088,
       "step": 404
     },
     {
       "epoch": 0.8412371134020619,
-      "grad_norm": 0.828125,
+      "grad_norm": 0.90234375,
       "learning_rate": 2.69086098475277e-06,
-      "loss": 0.1699,
-      "num_input_tokens_seen": 134815840,
+      "loss": 0.1707,
+      "num_input_tokens_seen": 134815184,
       "step": 408
     },
     {
       "epoch": 0.8494845360824742,
-      "grad_norm": 0.9921875,
+      "grad_norm": 0.97265625,
       "learning_rate": 2.4241997268220096e-06,
-      "loss": 0.1815,
-      "num_input_tokens_seen": 136262128,
+      "loss": 0.1822,
+      "num_input_tokens_seen": 136261472,
       "step": 412
     },
     {
       "epoch": 0.8577319587628865,
-      "grad_norm": 0.96484375,
+      "grad_norm": 0.98828125,
       "learning_rate": 2.1705973537884615e-06,
-      "loss": 0.1781,
-      "num_input_tokens_seen": 137430160,
+      "loss": 0.1809,
+      "num_input_tokens_seen": 137429504,
       "step": 416
     },
     {
       "epoch": 0.865979381443299,
-      "grad_norm": 0.95703125,
+      "grad_norm": 0.9375,
       "learning_rate": 1.9302422931574183e-06,
-      "loss": 0.1899,
-      "num_input_tokens_seen": 138709200,
+      "loss": 0.1885,
+      "num_input_tokens_seen": 138708544,
       "step": 420
     },
     {
       "epoch": 0.8742268041237113,
       "grad_norm": 0.83984375,
       "learning_rate": 1.7033131296318473e-06,
-      "loss": 0.1795,
-      "num_input_tokens_seen": 140033680,
+      "loss": 0.1687,
+      "num_input_tokens_seen": 140033024,
       "step": 424
     },
     {
       "epoch": 0.8824742268041237,
-      "grad_norm": 0.8203125,
+      "grad_norm": 0.859375,
       "learning_rate": 1.4899784724232968e-06,
-      "loss": 0.1749,
-      "num_input_tokens_seen": 141348848,
+      "loss": 0.1748,
+      "num_input_tokens_seen": 141348192,
       "step": 428
     },
     {
       "epoch": 0.8907216494845361,
-      "grad_norm": 0.83984375,
+      "grad_norm": 0.859375,
       "learning_rate": 1.2903968299746094e-06,
-      "loss": 0.171,
-      "num_input_tokens_seen": 142797664,
+      "loss": 0.1716,
+      "num_input_tokens_seen": 142800048,
       "step": 432
     },
     {
       "epoch": 0.8989690721649485,
-      "grad_norm": 0.89453125,
+      "grad_norm": 0.94140625,
       "learning_rate": 1.104716492187574e-06,
-      "loss": 0.1812,
-      "num_input_tokens_seen": 144154208,
+      "loss": 0.1841,
+      "num_input_tokens_seen": 144156592,
       "step": 436
     },
     {
       "epoch": 0.9072164948453608,
-      "grad_norm": 1.03125,
+      "grad_norm": 0.984375,
       "learning_rate": 9.330754202429726e-07,
-      "loss": 0.1891,
-      "num_input_tokens_seen": 145332560,
+      "loss": 0.1855,
+      "num_input_tokens_seen": 145334944,
       "step": 440
     },
     {
       "epoch": 0.9154639175257732,
-      "grad_norm": 0.98046875,
+      "grad_norm": 1.0625,
       "learning_rate": 7.756011440948996e-07,
-      "loss": 0.1902,
-      "num_input_tokens_seen": 146527344,
+      "loss": 0.1895,
+      "num_input_tokens_seen": 146529728,
       "step": 444
     },
     {
       "epoch": 0.9237113402061856,
-      "grad_norm": 1.0,
+      "grad_norm": 1.03125,
       "learning_rate": 6.324106677155573e-07,
-      "loss": 0.1793,
-      "num_input_tokens_seen": 147821568,
+      "loss": 0.1841,
+      "num_input_tokens_seen": 147823952,
       "step": 448
     },
     {
       "epoch": 0.931958762886598,
       "grad_norm": 0.88671875,
       "learning_rate": 5.036103821608485e-07,
-      "loss": 0.1844,
-      "num_input_tokens_seen": 149191664,
+      "loss": 0.1838,
+      "num_input_tokens_seen": 149194048,
       "step": 452
     },
     {
       "epoch": 0.9402061855670103,
-      "grad_norm": 0.8515625,
+      "grad_norm": 0.83984375,
       "learning_rate": 3.892959865214363e-07,
-      "loss": 0.1795,
-      "num_input_tokens_seen": 150526864,
+      "loss": 0.1773,
+      "num_input_tokens_seen": 150529248,
       "step": 456
     },
     {
       "epoch": 0.9484536082474226,
-      "grad_norm": 0.9453125,
+      "grad_norm": 1.1640625,
       "learning_rate": 2.8955241681795534e-07,
-      "loss": 0.1859,
-      "num_input_tokens_seen": 151861952,
+      "loss": 0.1863,
+      "num_input_tokens_seen": 151864336,
       "step": 460
     },
     {
       "epoch": 0.9567010309278351,
-      "grad_norm": 0.71875,
+      "grad_norm": 0.765625,
       "learning_rate": 2.044537828932458e-07,
-      "loss": 0.1787,
-      "num_input_tokens_seen": 153201488,
+      "loss": 0.1803,
+      "num_input_tokens_seen": 153203872,
       "step": 464
     },
     {
       "epoch": 0.9649484536082474,
-      "grad_norm": 0.87109375,
+      "grad_norm": 0.8828125,
       "learning_rate": 1.3406331334845813e-07,
-      "loss": 0.1884,
-      "num_input_tokens_seen": 154511184,
+      "loss": 0.1869,
+      "num_input_tokens_seen": 154513568,
       "step": 468
     },
     {
       "epoch": 0.9731958762886598,
-      "grad_norm": 0.796875,
+      "grad_norm": 0.8046875,
       "learning_rate": 7.843330856396103e-08,
-      "loss": 0.1858,
-      "num_input_tokens_seen": 155737200,
+      "loss": 0.1818,
+      "num_input_tokens_seen": 155739584,
       "step": 472
     },
     {
       "epoch": 0.9814432989690721,
-      "grad_norm": 0.8671875,
+      "grad_norm": 0.89453125,
       "learning_rate": 3.760510183997701e-08,
-      "loss": 0.183,
-      "num_input_tokens_seen": 157084960,
+      "loss": 0.1826,
+      "num_input_tokens_seen": 157087344,
       "step": 476
     },
     {
       "epoch": 0.9896907216494846,
-      "grad_norm": 0.93359375,
+      "grad_norm": 0.8984375,
       "learning_rate": 1.160902868577951e-08,
-      "loss": 0.1908,
-      "num_input_tokens_seen": 158349904,
+      "loss": 0.1905,
+      "num_input_tokens_seen": 158352288,
       "step": 480
     },
     {
       "epoch": 0.9979381443298969,
-      "grad_norm": 0.8984375,
+      "grad_norm": 0.86328125,
       "learning_rate": 4.64404280295927e-10,
-      "loss": 0.1764,
-      "num_input_tokens_seen": 159669152,
+      "loss": 0.1794,
+      "num_input_tokens_seen": 159671536,
       "step": 484
     },
     {
       "epoch": 1.0,
-      "eval_loss": 0.0952233299612999,
-      "eval_runtime": 83.4314,
-      "eval_samples_per_second": 12.453,
-      "eval_steps_per_second": 0.396,
-      "num_input_tokens_seen": 160041824,
+      "eval_loss": 0.09546061605215073,
+      "eval_runtime": 83.6615,
+      "eval_samples_per_second": 12.419,
+      "eval_steps_per_second": 0.394,
+      "num_input_tokens_seen": 160044208,
       "step": 485
     },
     {
       "epoch": 1.0,
-      "num_input_tokens_seen": 160041824,
+      "num_input_tokens_seen": 160044208,
       "step": 485,
-      "total_flos": 9.01334912177537e+17,
-      "train_loss": 0.20323730206366666,
-      "train_runtime": 14397.8817,
-      "train_samples_per_second": 4.306,
+      "total_flos": 9.013493089079132e+17,
+      "train_loss": 0.20329311268845784,
+      "train_runtime": 14400.7959,
+      "train_samples_per_second": 4.305,
       "train_steps_per_second": 0.034,
-      "train_tokens_per_second": 1386.226
+      "train_tokens_per_second": 1385.946
     }
   ],
   "logging_steps": 4,
   "max_steps": 485,
-  "num_input_tokens_seen": 160041824,
+  "num_input_tokens_seen": 160044208,
   "num_train_epochs": 1,
   "save_steps": 0,
   "stateful_callbacks": {
@@ -1014,7 +1014,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 9.01334912177537e+17,
+  "total_flos": 9.013493089079132e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null