Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,9 +20,9 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "down_proj",
     "qkv_proj",
     "o_proj",
+    "down_proj",
     "gate_up_proj"
   ],
   "task_type": "CAUSAL_LM",

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57220ba9a17fa6e4dba34f0bc82eabe7b1dba570ca47ce468ea5aa1bf4fd0748
+oid sha256:5e6d6588f31866bb890d7e0dc36fcda6e114720f8103076ba1ed36936a0cda75
 size 201361312

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64b5843437d5b65c7b6e205e61ab53ff2d7abfa4d98031aa7639591a9142a426
+oid sha256:1a419bd68a4498a5117c41daec6342b57ed7935b308a0f062ad4df6b411dfb5d
 size 402868986

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.20182333886623383,
+  "best_metric": 0.2011301964521408,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.02594033722438392,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.0005188067444876783,
-      "grad_norm": 7.257709980010986,
+      "grad_norm": 7.088029384613037,
       "learning_rate": 5e-05,
       "loss": 0.8421,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.0005188067444876783,
       "eval_loss": 0.9019366502761841,
-      "eval_runtime": 437.3032,
-      "eval_samples_per_second": 29.693,
-      "eval_steps_per_second": 3.714,
+      "eval_runtime": 439.5345,
+      "eval_samples_per_second": 29.543,
+      "eval_steps_per_second": 3.695,
       "step": 1
     },
     {
       "epoch": 0.0010376134889753567,
-      "grad_norm": 9.514312744140625,
+      "grad_norm": 9.270331382751465,
       "learning_rate": 0.0001,
       "loss": 0.7793,
       "step": 2
     },
     {
       "epoch": 0.0015564202334630351,
-      "grad_norm": 13.657303810119629,
+      "grad_norm": 13.402799606323242,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 0.8538,
+      "loss": 0.8548,
       "step": 3
     },
     {
       "epoch": 0.0020752269779507134,
-      "grad_norm": 13.498005867004395,
+      "grad_norm": 13.478734016418457,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 0.7789,
+      "loss": 0.7784,
       "step": 4
     },
     {
       "epoch": 0.0025940337224383916,
-      "grad_norm": 15.755859375,
+      "grad_norm": 15.679061889648438,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 0.6017,
+      "loss": 0.596,
       "step": 5
     },
     {
       "epoch": 0.0031128404669260703,
-      "grad_norm": 13.434879302978516,
+      "grad_norm": 13.199936866760254,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 0.4315,
+      "loss": 0.4246,
       "step": 6
     },
     {
       "epoch": 0.0036316472114137485,
-      "grad_norm": 8.647991180419922,
+      "grad_norm": 8.329848289489746,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 0.2934,
+      "loss": 0.2895,
       "step": 7
     },
     {
       "epoch": 0.004150453955901427,
-      "grad_norm": 5.960851192474365,
+      "grad_norm": 5.857521057128906,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 0.2742,
+      "loss": 0.2731,
       "step": 8
     },
     {
       "epoch": 0.004669260700389105,
-      "grad_norm": 5.692524433135986,
+      "grad_norm": 5.562647342681885,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 0.254,
+      "loss": 0.2526,
       "step": 9
     },
     {
       "epoch": 0.005188067444876783,
-      "grad_norm": 6.573745250701904,
+      "grad_norm": 6.4964423179626465,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 0.266,
+      "loss": 0.2667,
       "step": 10
     },
     {
       "epoch": 0.005706874189364461,
-      "grad_norm": 7.913269996643066,
+      "grad_norm": 7.804101467132568,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 0.2703,
+      "loss": 0.2699,
       "step": 11
     },
     {
       "epoch": 0.0062256809338521405,
-      "grad_norm": 7.290318489074707,
+      "grad_norm": 7.244912147521973,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 0.202,
+      "loss": 0.2022,
       "step": 12
     },
     {
       "epoch": 0.006744487678339819,
-      "grad_norm": 10.77153491973877,
+      "grad_norm": 10.575514793395996,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 0.5624,
+      "loss": 0.5618,
       "step": 13
     },
     {
       "epoch": 0.007263294422827497,
-      "grad_norm": 13.511357307434082,
+      "grad_norm": 13.16678524017334,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 0.5687,
+      "loss": 0.564,
       "step": 14
     },
     {
       "epoch": 0.007782101167315175,
-      "grad_norm": 9.808670043945312,
+      "grad_norm": 9.636844635009766,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 0.4122,
+      "loss": 0.4095,
       "step": 15
     },
     {
       "epoch": 0.008300907911802853,
-      "grad_norm": 7.290343761444092,
+      "grad_norm": 7.153684139251709,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 0.2995,
+      "loss": 0.2974,
       "step": 16
     },
     {
       "epoch": 0.008819714656290532,
-      "grad_norm": 8.092347145080566,
+      "grad_norm": 7.988699913024902,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 0.3212,
+      "loss": 0.3181,
       "step": 17
     },
     {
       "epoch": 0.00933852140077821,
-      "grad_norm": 9.879920959472656,
+      "grad_norm": 9.76743221282959,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 0.2751,
+      "loss": 0.2727,
       "step": 18
     },
     {
       "epoch": 0.009857328145265888,
-      "grad_norm": 7.612329483032227,
+      "grad_norm": 7.5284318923950195,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 0.2201,
+      "loss": 0.2183,
       "step": 19
     },
     {
       "epoch": 0.010376134889753566,
-      "grad_norm": 5.5150322914123535,
+      "grad_norm": 5.3470563888549805,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 0.1909,
+      "loss": 0.1889,
       "step": 20
     },
     {
       "epoch": 0.010894941634241245,
-      "grad_norm": 3.7366225719451904,
+      "grad_norm": 3.7029356956481934,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 0.1765,
+      "loss": 0.1763,
       "step": 21
     },
     {
       "epoch": 0.011413748378728923,
-      "grad_norm": 4.036696434020996,
+      "grad_norm": 4.031704425811768,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 0.1671,
+      "loss": 0.1656,
       "step": 22
     },
     {
       "epoch": 0.011932555123216601,
-      "grad_norm": 5.270599365234375,
+      "grad_norm": 5.219779014587402,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 0.1808,
+      "loss": 0.1802,
       "step": 23
     },
     {
       "epoch": 0.012451361867704281,
-      "grad_norm": 8.023921012878418,
+      "grad_norm": 7.983980178833008,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 0.2025,
+      "loss": 0.2019,
       "step": 24
     },
     {
       "epoch": 0.01297016861219196,
-      "grad_norm": 8.021526336669922,
+      "grad_norm": 8.021879196166992,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 0.2274,
+      "loss": 0.2279,
       "step": 25
     },
     {
       "epoch": 0.01297016861219196,
-      "eval_loss": 0.235469788312912,
-      "eval_runtime": 437.5793,
-      "eval_samples_per_second": 29.675,
-      "eval_steps_per_second": 3.711,
+      "eval_loss": 0.23452582955360413,
+      "eval_runtime": 439.5012,
+      "eval_samples_per_second": 29.545,
+      "eval_steps_per_second": 3.695,
       "step": 25
     },
     {
       "epoch": 0.013488975356679637,
-      "grad_norm": 5.151494979858398,
+      "grad_norm": 5.008545398712158,
       "learning_rate": 5e-05,
-      "loss": 0.625,
+      "loss": 0.624,
       "step": 26
     },
     {
       "epoch": 0.014007782101167316,
-      "grad_norm": 5.601419448852539,
+      "grad_norm": 5.509274959564209,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 0.4202,
+      "loss": 0.4169,
       "step": 27
     },
     {
       "epoch": 0.014526588845654994,
-      "grad_norm": 5.864795207977295,
+      "grad_norm": 5.751987934112549,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 0.2736,
+      "loss": 0.2727,
       "step": 28
     },
     {
       "epoch": 0.015045395590142672,
-      "grad_norm": 4.344137668609619,
+      "grad_norm": 4.345045566558838,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 0.2721,
+      "loss": 0.2713,
       "step": 29
     },
     {
       "epoch": 0.01556420233463035,
-      "grad_norm": 3.193909168243408,
+      "grad_norm": 3.158082962036133,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 0.2,
+      "loss": 0.1986,
       "step": 30
     },
     {
       "epoch": 0.016083009079118027,
-      "grad_norm": 2.997635841369629,
+      "grad_norm": 2.939302444458008,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 0.1727,
+      "loss": 0.1714,
       "step": 31
     },
     {
       "epoch": 0.016601815823605707,
-      "grad_norm": 2.729241371154785,
+      "grad_norm": 2.697280168533325,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 0.1658,
+      "loss": 0.1648,
       "step": 32
     },
     {
       "epoch": 0.017120622568093387,
-      "grad_norm": 2.756976842880249,
+      "grad_norm": 2.7456414699554443,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 0.1545,
+      "loss": 0.1538,
       "step": 33
     },
     {
       "epoch": 0.017639429312581063,
-      "grad_norm": 2.4652767181396484,
+      "grad_norm": 2.4092962741851807,
       "learning_rate": 2.500000000000001e-05,
       "loss": 0.1408,
       "step": 34
     },
     {
       "epoch": 0.018158236057068743,
-      "grad_norm": 2.768718957901001,
+      "grad_norm": 2.7168354988098145,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 0.1777,
+      "loss": 0.178,
       "step": 35
     },
     {
       "epoch": 0.01867704280155642,
-      "grad_norm": 2.951350212097168,
+      "grad_norm": 2.9179553985595703,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 0.1508,
+      "loss": 0.1499,
       "step": 36
     },
     {
       "epoch": 0.0191958495460441,
-      "grad_norm": 4.257266521453857,
+      "grad_norm": 4.161209583282471,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 0.1418,
+      "loss": 0.141,
       "step": 37
     },
     {
       "epoch": 0.019714656290531776,
-      "grad_norm": 4.354798793792725,
+      "grad_norm": 4.350818634033203,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 0.421,
+      "loss": 0.4201,
       "step": 38
     },
     {
       "epoch": 0.020233463035019456,
-      "grad_norm": 3.913156032562256,
+      "grad_norm": 3.8931922912597656,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 0.378,
+      "loss": 0.3758,
       "step": 39
     },
     {
       "epoch": 0.020752269779507133,
-      "grad_norm": 3.5373024940490723,
+      "grad_norm": 3.4323060512542725,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 0.2926,
+      "loss": 0.2915,
       "step": 40
     },
     {
       "epoch": 0.021271076523994813,
-      "grad_norm": 4.939075946807861,
+      "grad_norm": 4.832789897918701,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 0.1924,
+      "loss": 0.1918,
       "step": 41
     },
     {
       "epoch": 0.02178988326848249,
-      "grad_norm": 3.2499287128448486,
+      "grad_norm": 3.2131426334381104,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 0.2142,
+      "loss": 0.2135,
       "step": 42
     },
     {
       "epoch": 0.02230869001297017,
-      "grad_norm": 2.4298489093780518,
+      "grad_norm": 2.4344329833984375,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 0.1793,
+      "loss": 0.1791,
       "step": 43
     },
     {
       "epoch": 0.022827496757457846,
-      "grad_norm": 2.3704288005828857,
+      "grad_norm": 2.4006295204162598,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 0.1485,
+      "loss": 0.1481,
       "step": 44
     },
     {
       "epoch": 0.023346303501945526,
-      "grad_norm": 2.584822177886963,
+      "grad_norm": 2.5179152488708496,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 0.1585,
+      "loss": 0.1581,
       "step": 45
     },
     {
       "epoch": 0.023865110246433202,
-      "grad_norm": 2.569350481033325,
+      "grad_norm": 2.5411667823791504,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 0.1276,
+      "loss": 0.1277,
       "step": 46
     },
     {
       "epoch": 0.024383916990920882,
-      "grad_norm": 2.4440035820007324,
+      "grad_norm": 2.382852792739868,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 0.1723,
+      "loss": 0.1709,
       "step": 47
     },
     {
       "epoch": 0.024902723735408562,
-      "grad_norm": 2.5331945419311523,
+      "grad_norm": 2.456338405609131,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 0.1476,
+      "loss": 0.1471,
       "step": 48
     },
     {
       "epoch": 0.02542153047989624,
-      "grad_norm": 3.334847927093506,
+      "grad_norm": 3.2760050296783447,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 0.1711,
+      "loss": 0.1705,
       "step": 49
     },
     {
       "epoch": 0.02594033722438392,
-      "grad_norm": 3.929615020751953,
+      "grad_norm": 3.9281301498413086,
       "learning_rate": 0.0,
-      "loss": 0.1231,
+      "loss": 0.1235,
       "step": 50
     },
     {
       "epoch": 0.02594033722438392,
-      "eval_loss": 0.20182333886623383,
-      "eval_runtime": 437.3533,
-      "eval_samples_per_second": 29.69,
-      "eval_steps_per_second": 3.713,
+      "eval_loss": 0.2011301964521408,
+      "eval_runtime": 439.5685,
+      "eval_samples_per_second": 29.54,
+      "eval_steps_per_second": 3.695,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b4c867fc22c662d2eb55140f540ec9341dba7a58b87b7576d7182430af5852d
+oid sha256:32c9a617800bcf00617a974e9c1ac0e7ad0ab0a36eece5b682859f66cccab4d2
 size 6776