Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -21,10 +21,10 @@
   "revision": null,
   "target_modules": [
     "k_proj",
-    "down_proj",
-    "gate_proj",
     "q_proj",
+    "gate_proj",
     "v_proj",
+    "down_proj",
     "up_proj",
     "o_proj"
   ],

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:020b5754070bfaf7eaf5513f8adb989bb64a52299a64a292eef6230e71963075
+oid sha256:3446403d2f0c7e1833fe69ae3f77b5a43536a61c7a2973ebdce45154acfcf01f
 size 335604696

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1fa92f2210e69aa0a9c191f16d0e25a72774bf4e987c7510780097c2f3415770
+oid sha256:85e05a87b600fb59a55ed5531ab0ccad9e06f89761e2179afdb930d13d63f21b
 size 671466706

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.039766620844602585,
+  "best_metric": 0.03734762966632843,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 2.6996699669966997,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.052805280528052806,
-      "grad_norm": 240.08116149902344,
+      "grad_norm": 241.61077880859375,
       "learning_rate": 5e-05,
       "loss": 61.3338,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.052805280528052806,
       "eval_loss": 3.5627546310424805,
-      "eval_runtime": 4.3057,
-      "eval_samples_per_second": 29.728,
-      "eval_steps_per_second": 3.716,
+      "eval_runtime": 4.3596,
+      "eval_samples_per_second": 29.361,
+      "eval_steps_per_second": 3.67,
       "step": 1
     },
     {
       "epoch": 0.10561056105610561,
-      "grad_norm": 311.3833923339844,
+      "grad_norm": 313.4629821777344,
       "learning_rate": 0.0001,
       "loss": 55.4131,
       "step": 2
     },
     {
       "epoch": 0.15841584158415842,
-      "grad_norm": 186.2847137451172,
+      "grad_norm": 185.32693481445312,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 35.886,
+      "loss": 35.7674,
       "step": 3
     },
     {
       "epoch": 0.21122112211221122,
-      "grad_norm": 109.15618896484375,
+      "grad_norm": 110.76226806640625,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 20.0972,
+      "loss": 20.0764,
       "step": 4
     },
     {
       "epoch": 0.264026402640264,
-      "grad_norm": 88.25745391845703,
+      "grad_norm": 87.55223846435547,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 24.9318,
+      "loss": 24.8683,
       "step": 5
     },
     {
       "epoch": 0.31683168316831684,
-      "grad_norm": 140.07655334472656,
+      "grad_norm": 139.35531616210938,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 18.6597,
+      "loss": 18.6656,
       "step": 6
     },
     {
       "epoch": 0.3696369636963696,
-      "grad_norm": 134.88388061523438,
+      "grad_norm": 132.13233947753906,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 14.9552,
+      "loss": 14.8754,
       "step": 7
     },
     {
       "epoch": 0.42244224422442245,
-      "grad_norm": 63.23054122924805,
+      "grad_norm": 63.427032470703125,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 10.7139,
+      "loss": 10.6936,
       "step": 8
     },
     {
       "epoch": 0.4752475247524752,
-      "grad_norm": 71.05472564697266,
+      "grad_norm": 70.66545104980469,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 9.8416,
+      "loss": 9.8106,
       "step": 9
     },
     {
       "epoch": 0.528052805280528,
-      "grad_norm": 97.5541763305664,
+      "grad_norm": 97.06773376464844,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 19.9574,
+      "loss": 19.9034,
       "step": 10
     },
     {
       "epoch": 0.5808580858085809,
-      "grad_norm": 69.11842346191406,
+      "grad_norm": 69.93522644042969,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 11.0275,
+      "loss": 10.8521,
       "step": 11
     },
     {
       "epoch": 0.6336633663366337,
-      "grad_norm": 56.24506759643555,
+      "grad_norm": 57.24028396606445,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 10.1437,
+      "loss": 10.1543,
       "step": 12
     },
     {
       "epoch": 0.6864686468646864,
-      "grad_norm": 54.2360954284668,
+      "grad_norm": 57.92759323120117,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 7.8817,
+      "loss": 7.9742,
       "step": 13
     },
     {
       "epoch": 0.7392739273927392,
-      "grad_norm": 65.11487579345703,
+      "grad_norm": 64.80909729003906,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 9.9734,
+      "loss": 9.9122,
       "step": 14
     },
     {
       "epoch": 0.7920792079207921,
-      "grad_norm": 63.567726135253906,
+      "grad_norm": 62.86793518066406,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 8.4856,
+      "loss": 8.432,
       "step": 15
     },
     {
       "epoch": 0.8448844884488449,
-      "grad_norm": 54.104248046875,
+      "grad_norm": 56.4834098815918,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 5.8322,
+      "loss": 5.9763,
       "step": 16
     },
     {
       "epoch": 0.8976897689768977,
-      "grad_norm": 41.041969299316406,
+      "grad_norm": 40.859291076660156,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 3.4734,
+      "loss": 3.4547,
       "step": 17
     },
     {
       "epoch": 0.9504950495049505,
-      "grad_norm": 51.88878631591797,
+      "grad_norm": 50.14191436767578,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 3.058,
+      "loss": 2.9626,
       "step": 18
     },
     {
       "epoch": 1.033003300330033,
-      "grad_norm": 63.439964294433594,
+      "grad_norm": 71.62645721435547,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 5.4253,
+      "loss": 5.432,
       "step": 19
     },
     {
       "epoch": 1.0858085808580857,
-      "grad_norm": 48.61954116821289,
+      "grad_norm": 51.751678466796875,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 2.854,
+      "loss": 2.9714,
       "step": 20
     },
     {
       "epoch": 1.1386138613861387,
-      "grad_norm": 55.616329193115234,
+      "grad_norm": 71.75513458251953,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 2.0837,
+      "loss": 2.1652,
       "step": 21
     },
     {
       "epoch": 1.1914191419141915,
-      "grad_norm": 50.29981231689453,
+      "grad_norm": 51.177284240722656,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 1.5129,
+      "loss": 1.4909,
       "step": 22
     },
     {
       "epoch": 1.2442244224422443,
-      "grad_norm": 48.272666931152344,
+      "grad_norm": 72.12232208251953,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 1.6346,
+      "loss": 1.4707,
       "step": 23
     },
     {
       "epoch": 1.297029702970297,
-      "grad_norm": 82.75598907470703,
+      "grad_norm": 81.59396362304688,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 2.8828,
+      "loss": 2.5601,
       "step": 24
     },
     {
       "epoch": 1.3498349834983498,
-      "grad_norm": 55.31129837036133,
+      "grad_norm": 44.115966796875,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 0.9758,
+      "loss": 0.8231,
       "step": 25
     },
     {
       "epoch": 1.3498349834983498,
-      "eval_loss": 0.11382482945919037,
-      "eval_runtime": 4.569,
-      "eval_samples_per_second": 28.015,
-      "eval_steps_per_second": 3.502,
+      "eval_loss": 0.09059706330299377,
+      "eval_runtime": 4.6098,
+      "eval_samples_per_second": 27.767,
+      "eval_steps_per_second": 3.471,
       "step": 25
     },
     {
       "epoch": 1.4026402640264026,
-      "grad_norm": 80.16549682617188,
+      "grad_norm": 51.42778015136719,
       "learning_rate": 5e-05,
-      "loss": 0.9275,
+      "loss": 0.5969,
       "step": 26
     },
     {
       "epoch": 1.4554455445544554,
-      "grad_norm": 44.94060516357422,
+      "grad_norm": 41.340782165527344,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 0.6253,
+      "loss": 0.6936,
       "step": 27
     },
     {
       "epoch": 1.5082508250825084,
-      "grad_norm": 53.476905822753906,
+      "grad_norm": 62.70668411254883,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 2.6445,
+      "loss": 2.3933,
       "step": 28
     },
     {
       "epoch": 1.561056105610561,
-      "grad_norm": 72.94212341308594,
+      "grad_norm": 68.67789459228516,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 0.9836,
+      "loss": 0.8647,
       "step": 29
     },
     {
       "epoch": 1.613861386138614,
-      "grad_norm": 51.1599235534668,
+      "grad_norm": 48.94205856323242,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 0.539,
+      "loss": 0.5317,
       "step": 30
     },
     {
       "epoch": 1.6666666666666665,
-      "grad_norm": 18.60490608215332,
+      "grad_norm": 18.71625518798828,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 0.2169,
+      "loss": 0.1899,
       "step": 31
     },
     {
       "epoch": 1.7194719471947195,
-      "grad_norm": 47.54155349731445,
+      "grad_norm": 47.35069274902344,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 0.7477,
+      "loss": 0.7913,
       "step": 32
     },
     {
       "epoch": 1.7722772277227723,
-      "grad_norm": 68.19134521484375,
+      "grad_norm": 40.61697006225586,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 0.846,
+      "loss": 0.6418,
       "step": 33
     },
     {
       "epoch": 1.825082508250825,
-      "grad_norm": 39.737762451171875,
+      "grad_norm": 45.03499221801758,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 0.4499,
+      "loss": 0.4943,
       "step": 34
     },
     {
       "epoch": 1.8778877887788779,
-      "grad_norm": 22.386844635009766,
+      "grad_norm": 48.89573669433594,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 0.1428,
+      "loss": 0.2491,
       "step": 35
     },
     {
       "epoch": 1.9306930693069306,
-      "grad_norm": 8.85619068145752,
+      "grad_norm": 42.19164276123047,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 0.0899,
+      "loss": 0.1735,
       "step": 36
     },
     {
       "epoch": 2.0132013201320134,
-      "grad_norm": 13.05267333984375,
+      "grad_norm": 26.78213119506836,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 0.4039,
+      "loss": 0.4541,
       "step": 37
     },
     {
       "epoch": 2.066006600660066,
-      "grad_norm": 10.15993881225586,
+      "grad_norm": 43.349700927734375,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 0.2195,
+      "loss": 0.2041,
       "step": 38
     },
     {
       "epoch": 2.118811881188119,
-      "grad_norm": 4.588918685913086,
+      "grad_norm": 5.292726516723633,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 0.0582,
+      "loss": 0.0695,
       "step": 39
     },
     {
       "epoch": 2.1716171617161715,
-      "grad_norm": 17.074146270751953,
+      "grad_norm": 4.951777458190918,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 0.0406,
+      "loss": 0.0261,
       "step": 40
     },
     {
       "epoch": 2.2244224422442245,
-      "grad_norm": 1.4361339807510376,
+      "grad_norm": 29.308528900146484,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 0.0148,
+      "loss": 0.0913,
       "step": 41
     },
     {
       "epoch": 2.2772277227722775,
-      "grad_norm": 19.55823516845703,
+      "grad_norm": 20.512184143066406,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 0.3258,
+      "loss": 0.3224,
       "step": 42
     },
     {
       "epoch": 2.33003300330033,
-      "grad_norm": 23.317541122436523,
+      "grad_norm": 7.966917514801025,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 0.1395,
+      "loss": 0.0802,
       "step": 43
     },
     {
       "epoch": 2.382838283828383,
-      "grad_norm": 3.5347235202789307,
+      "grad_norm": 3.8203468322753906,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 0.0133,
+      "loss": 0.0214,
       "step": 44
     },
     {
       "epoch": 2.4356435643564356,
-      "grad_norm": 48.761112213134766,
+      "grad_norm": 3.0683605670928955,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 0.1465,
+      "loss": 0.0138,
       "step": 45
     },
     {
       "epoch": 2.4884488448844886,
-      "grad_norm": 64.77693176269531,
+      "grad_norm": 23.591899871826172,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 0.5884,
+      "loss": 0.5393,
       "step": 46
     },
     {
       "epoch": 2.541254125412541,
-      "grad_norm": 21.97736167907715,
+      "grad_norm": 18.360029220581055,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 0.0732,
+      "loss": 0.0586,
       "step": 47
     },
     {
       "epoch": 2.594059405940594,
-      "grad_norm": 19.152101516723633,
+      "grad_norm": 14.321728706359863,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 0.162,
+      "loss": 0.0914,
       "step": 48
     },
     {
       "epoch": 2.6468646864686467,
-      "grad_norm": 14.94710636138916,
+      "grad_norm": 16.951425552368164,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 0.1771,
+      "loss": 0.1885,
       "step": 49
     },
     {
       "epoch": 2.6996699669966997,
-      "grad_norm": 8.99813461303711,
+      "grad_norm": 31.718521118164062,
       "learning_rate": 0.0,
-      "loss": 0.2619,
+      "loss": 0.285,
       "step": 50
     },
     {
       "epoch": 2.6996699669966997,
-      "eval_loss": 0.039766620844602585,
-      "eval_runtime": 4.8436,
-      "eval_samples_per_second": 26.427,
-      "eval_steps_per_second": 3.303,
+      "eval_loss": 0.03734762966632843,
+      "eval_runtime": 4.8904,
+      "eval_samples_per_second": 26.174,
+      "eval_steps_per_second": 3.272,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fce2681dc79991c6f33e2d27573b5f6d6f5cf0f1fcde8d296e5d63aeff6c313b
+oid sha256:5dd421ee496f84b943e667379d94c279e81d013b511fd3641f3973aa3000a968
 size 6776