Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "gate_proj",
-    "q_proj",
-    "k_proj",
     "up_proj",
+    "k_proj",
+    "gate_proj",
     "o_proj",
-    "down_proj"
+    "down_proj",
+    "v_proj",
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87c761f6a9e19c98c817c8efffb941c28e04619d8ce2bf1251e883c02e20b3ce
+oid sha256:453f679ff83f6cb78ca4a52079d84a61f6adf84723e10a078b4af6fd7b849cce
 size 524363632

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a34870135d1bccfe0202414b03ea26f6c8f7267bee90aa2a8729e2026a0255af
+oid sha256:30333b30ba5c00ca0ab3359bf31187eabd94f27746c78587bba588e5a2ee693e
 size 1049049122

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 1.4218698740005493,
+  "best_metric": 1.42130446434021,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.7920792079207921,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.015841584158415842,
-      "grad_norm": 115.84626770019531,
+      "grad_norm": 116.54348754882812,
       "learning_rate": 5e-05,
       "loss": 3.8832,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.015841584158415842,
       "eval_loss": 4.637433052062988,
-      "eval_runtime": 21.113,
-      "eval_samples_per_second": 20.177,
-      "eval_steps_per_second": 2.558,
+      "eval_runtime": 21.3372,
+      "eval_samples_per_second": 19.965,
+      "eval_steps_per_second": 2.531,
       "step": 1
     },
     {
       "epoch": 0.031683168316831684,
-      "grad_norm": 130.38941955566406,
+      "grad_norm": 131.15432739257812,
       "learning_rate": 0.0001,
       "loss": 4.2831,
       "step": 2
     },
     {
       "epoch": 0.047524752475247525,
-      "grad_norm": 125.5568618774414,
+      "grad_norm": 126.33002471923828,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 3.9435,
+      "loss": 3.9449,
       "step": 3
     },
     {
       "epoch": 0.06336633663366337,
-      "grad_norm": 71.17156219482422,
+      "grad_norm": 70.8408432006836,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 2.8621,
+      "loss": 2.861,
       "step": 4
     },
     {
       "epoch": 0.07920792079207921,
-      "grad_norm": 51.608543395996094,
+      "grad_norm": 51.910221099853516,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 2.3486,
+      "loss": 2.3513,
       "step": 5
     },
     {
       "epoch": 0.09504950495049505,
-      "grad_norm": 35.24829864501953,
+      "grad_norm": 35.98851013183594,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 2.1069,
+      "loss": 2.1084,
       "step": 6
     },
     {
       "epoch": 0.11089108910891089,
-      "grad_norm": 19.96686363220215,
+      "grad_norm": 20.054012298583984,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 1.7398,
+      "loss": 1.7399,
       "step": 7
     },
     {
       "epoch": 0.12673267326732673,
-      "grad_norm": 11.223455429077148,
+      "grad_norm": 11.261577606201172,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 1.5823,
+      "loss": 1.5846,
       "step": 8
     },
     {
       "epoch": 0.14257425742574256,
-      "grad_norm": 11.296300888061523,
+      "grad_norm": 11.138843536376953,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 1.6323,
+      "loss": 1.6313,
       "step": 9
     },
     {
       "epoch": 0.15841584158415842,
-      "grad_norm": 9.405316352844238,
+      "grad_norm": 9.569158554077148,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 1.4158,
+      "loss": 1.417,
       "step": 10
     },
     {
       "epoch": 0.17425742574257425,
-      "grad_norm": 7.900296211242676,
+      "grad_norm": 7.887810707092285,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 1.4083,
+      "loss": 1.4092,
       "step": 11
     },
     {
       "epoch": 0.1900990099009901,
-      "grad_norm": 8.70493221282959,
+      "grad_norm": 8.705619812011719,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 1.5026,
+      "loss": 1.503,
       "step": 12
     },
     {
       "epoch": 0.20594059405940593,
-      "grad_norm": 16.30130386352539,
+      "grad_norm": 16.14295768737793,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 1.7546,
+      "loss": 1.7561,
       "step": 13
     },
     {
       "epoch": 0.22178217821782178,
-      "grad_norm": 25.1881103515625,
+      "grad_norm": 24.8006649017334,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 1.9353,
+      "loss": 1.937,
       "step": 14
     },
     {
       "epoch": 0.2376237623762376,
-      "grad_norm": 13.436919212341309,
+      "grad_norm": 13.125953674316406,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 1.625,
+      "loss": 1.6222,
       "step": 15
     },
     {
       "epoch": 0.25346534653465347,
-      "grad_norm": 9.559671401977539,
+      "grad_norm": 9.406954765319824,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 1.4622,
+      "loss": 1.4597,
       "step": 16
     },
     {
       "epoch": 0.2693069306930693,
-      "grad_norm": 7.00495719909668,
+      "grad_norm": 6.961748123168945,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 1.4996,
+      "loss": 1.5003,
       "step": 17
     },
     {
       "epoch": 0.2851485148514851,
-      "grad_norm": 6.28257417678833,
+      "grad_norm": 6.327335357666016,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 1.3567,
+      "loss": 1.3565,
       "step": 18
     },
     {
       "epoch": 0.300990099009901,
-      "grad_norm": 6.954759120941162,
+      "grad_norm": 6.976326942443848,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 1.2845,
+      "loss": 1.2852,
       "step": 19
     },
     {
       "epoch": 0.31683168316831684,
-      "grad_norm": 5.887358665466309,
+      "grad_norm": 5.895344257354736,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 1.3541,
+      "loss": 1.3554,
       "step": 20
     },
     {
       "epoch": 0.3326732673267327,
-      "grad_norm": 7.681492328643799,
+      "grad_norm": 7.499822616577148,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 1.3665,
+      "loss": 1.3676,
       "step": 21
     },
     {
       "epoch": 0.3485148514851485,
-      "grad_norm": 9.990008354187012,
+      "grad_norm": 9.441739082336426,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 1.4321,
+      "loss": 1.4337,
       "step": 22
     },
     {
       "epoch": 0.36435643564356435,
-      "grad_norm": 7.844481945037842,
+      "grad_norm": 7.759869575500488,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 1.3991,
+      "loss": 1.4005,
       "step": 23
     },
     {
       "epoch": 0.3801980198019802,
-      "grad_norm": 7.6049370765686035,
+      "grad_norm": 7.4442925453186035,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 1.3329,
+      "loss": 1.3328,
       "step": 24
     },
     {
       "epoch": 0.39603960396039606,
-      "grad_norm": 10.985588073730469,
+      "grad_norm": 10.700361251831055,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 1.5425,
+      "loss": 1.544,
       "step": 25
     },
     {
       "epoch": 0.39603960396039606,
-      "eval_loss": 1.4600543975830078,
-      "eval_runtime": 21.1981,
-      "eval_samples_per_second": 20.096,
-      "eval_steps_per_second": 2.547,
+      "eval_loss": 1.4611538648605347,
+      "eval_runtime": 21.4271,
+      "eval_samples_per_second": 19.881,
+      "eval_steps_per_second": 2.52,
       "step": 25
     },
     {
       "epoch": 0.41188118811881186,
-      "grad_norm": 13.096388816833496,
+      "grad_norm": 12.989526748657227,
       "learning_rate": 5e-05,
-      "loss": 1.6191,
+      "loss": 1.6196,
       "step": 26
     },
     {
       "epoch": 0.4277227722772277,
-      "grad_norm": 12.10949993133545,
+      "grad_norm": 13.955931663513184,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 1.683,
+      "loss": 1.6851,
       "step": 27
     },
     {
       "epoch": 0.44356435643564357,
-      "grad_norm": 7.642582893371582,
+      "grad_norm": 7.539506912231445,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 1.4826,
+      "loss": 1.4837,
       "step": 28
     },
     {
       "epoch": 0.4594059405940594,
-      "grad_norm": 7.771426200866699,
+      "grad_norm": 7.674356937408447,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 1.4381,
+      "loss": 1.4401,
       "step": 29
     },
     {
       "epoch": 0.4752475247524752,
-      "grad_norm": 7.80220365524292,
+      "grad_norm": 7.777220249176025,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 1.2711,
+      "loss": 1.2722,
       "step": 30
     },
     {
       "epoch": 0.4910891089108911,
-      "grad_norm": 6.0962443351745605,
+      "grad_norm": 6.190223693847656,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 1.2835,
+      "loss": 1.2836,
       "step": 31
     },
     {
       "epoch": 0.5069306930693069,
-      "grad_norm": 4.596838474273682,
+      "grad_norm": 4.608152389526367,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 1.2574,
+      "loss": 1.2565,
       "step": 32
     },
     {
       "epoch": 0.5227722772277228,
-      "grad_norm": 4.949024677276611,
+      "grad_norm": 4.905404567718506,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 1.3384,
+      "loss": 1.3379,
       "step": 33
     },
     {
       "epoch": 0.5386138613861386,
-      "grad_norm": 4.323665142059326,
+      "grad_norm": 4.258622646331787,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 1.3154,
+      "loss": 1.314,
       "step": 34
     },
     {
       "epoch": 0.5544554455445545,
-      "grad_norm": 4.781197547912598,
+      "grad_norm": 4.795726299285889,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 1.3619,
+      "loss": 1.3602,
       "step": 35
     },
     {
       "epoch": 0.5702970297029702,
-      "grad_norm": 5.3685479164123535,
+      "grad_norm": 5.283386707305908,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 1.3155,
+      "loss": 1.3136,
       "step": 36
     },
     {
       "epoch": 0.5861386138613861,
-      "grad_norm": 8.885624885559082,
+      "grad_norm": 8.913880348205566,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 1.4451,
+      "loss": 1.4442,
       "step": 37
     },
     {
       "epoch": 0.601980198019802,
-      "grad_norm": 7.00028133392334,
+      "grad_norm": 7.143167972564697,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 1.5786,
+      "loss": 1.5798,
       "step": 38
     },
     {
       "epoch": 0.6178217821782178,
-      "grad_norm": 8.92757797241211,
+      "grad_norm": 9.055289268493652,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 1.6152,
+      "loss": 1.621,
       "step": 39
     },
     {
       "epoch": 0.6336633663366337,
-      "grad_norm": 6.305557727813721,
+      "grad_norm": 6.381430625915527,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 1.5112,
+      "loss": 1.5137,
       "step": 40
     },
     {
       "epoch": 0.6495049504950495,
-      "grad_norm": 6.068126201629639,
+      "grad_norm": 6.079700946807861,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 1.3797,
+      "loss": 1.3825,
       "step": 41
     },
     {
       "epoch": 0.6653465346534654,
-      "grad_norm": 5.500480651855469,
+      "grad_norm": 5.542724132537842,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 1.3315,
+      "loss": 1.3318,
       "step": 42
     },
     {
       "epoch": 0.6811881188118812,
-      "grad_norm": 4.893165111541748,
+      "grad_norm": 4.91347599029541,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 1.3253,
+      "loss": 1.3278,
       "step": 43
     },
     {
       "epoch": 0.697029702970297,
-      "grad_norm": 4.1639180183410645,
+      "grad_norm": 4.155934810638428,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 1.3478,
+      "loss": 1.3473,
       "step": 44
     },
     {
       "epoch": 0.7128712871287128,
-      "grad_norm": 4.112819194793701,
+      "grad_norm": 4.090728759765625,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 1.2916,
+      "loss": 1.2929,
       "step": 45
     },
     {
       "epoch": 0.7287128712871287,
-      "grad_norm": 4.722405433654785,
+      "grad_norm": 4.725870609283447,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 1.3223,
+      "loss": 1.3232,
       "step": 46
     },
     {
       "epoch": 0.7445544554455445,
-      "grad_norm": 5.680431842803955,
+      "grad_norm": 5.680789947509766,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 1.3769,
+      "loss": 1.3764,
       "step": 47
     },
     {
       "epoch": 0.7603960396039604,
-      "grad_norm": 5.624794960021973,
+      "grad_norm": 5.59313440322876,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 1.4066,
+      "loss": 1.4059,
       "step": 48
     },
     {
       "epoch": 0.7762376237623763,
-      "grad_norm": 7.075315475463867,
+      "grad_norm": 7.0038652420043945,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 1.3627,
+      "loss": 1.3611,
       "step": 49
     },
     {
       "epoch": 0.7920792079207921,
-      "grad_norm": 10.405349731445312,
+      "grad_norm": 10.242979049682617,
       "learning_rate": 0.0,
-      "loss": 1.3777,
+      "loss": 1.3739,
       "step": 50
     },
     {
       "epoch": 0.7920792079207921,
-      "eval_loss": 1.4218698740005493,
-      "eval_runtime": 21.1894,
-      "eval_samples_per_second": 20.104,
-      "eval_steps_per_second": 2.548,
+      "eval_loss": 1.42130446434021,
+      "eval_runtime": 21.4232,
+      "eval_samples_per_second": 19.885,
+      "eval_steps_per_second": 2.521,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac3e76440ec12a175bfc0815ce339f1f19754c1a4d3893054a2983def642f23f
+oid sha256:f73a5c49f83bb452292d035856867321abe71eecac39011cbbc7e22d4eae41b4
 size 6776