Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "gate_proj",
-    "o_proj",
-    "down_proj",
     "up_proj",
     "k_proj",
-    "v_proj"
+    "gate_proj",
+    "q_proj",
+    "down_proj",
+    "v_proj",
+    "o_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 sha256:8cdad695471ac5ebe1288a36a991fb68aa3a4df78e52765c1b084596c758dd4d
+oid sha256:1de24fc8f01deb18a95dcc6729ffe30a69cd909f7e3333ffc10f50ea664e1e25
 size 432223744

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:07084887e9f9ab1c4f8931c6516ecd1497bb7307b3232f0def48593f713e74a1
+oid sha256:36661c8f83cc9f1d8d841dfb010512a6375009ff360b84507b0108d6caf544be
 size 864785974

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.7666683197021484,
+  "best_metric": 0.7679101824760437,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 1.4652014652014653,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.029304029304029304,
-      "grad_norm": 42.63615417480469,
+      "grad_norm": 42.315101623535156,
       "learning_rate": 5e-05,
       "loss": 10.8108,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.029304029304029304,
       "eval_loss": 11.66943359375,
-      "eval_runtime": 12.0274,
-      "eval_samples_per_second": 19.123,
-      "eval_steps_per_second": 2.411,
+      "eval_runtime": 12.0921,
+      "eval_samples_per_second": 19.021,
+      "eval_steps_per_second": 2.398,
       "step": 1
     },
     {
       "epoch": 0.05860805860805861,
-      "grad_norm": 43.70536804199219,
+      "grad_norm": 43.577064514160156,
       "learning_rate": 0.0001,
       "loss": 11.4368,
       "step": 2
     },
     {
       "epoch": 0.08791208791208792,
-      "grad_norm": 30.874406814575195,
+      "grad_norm": 31.090862274169922,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 8.1072,
+      "loss": 8.1051,
       "step": 3
     },
     {
       "epoch": 0.11721611721611722,
-      "grad_norm": 17.766359329223633,
+      "grad_norm": 17.739784240722656,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 3.914,
+      "loss": 3.8912,
       "step": 4
     },
     {
       "epoch": 0.14652014652014653,
-      "grad_norm": 10.942605972290039,
+      "grad_norm": 10.802330017089844,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 2.2661,
+      "loss": 2.2566,
       "step": 5
     },
     {
       "epoch": 0.17582417582417584,
-      "grad_norm": 4.077508926391602,
+      "grad_norm": 4.046825408935547,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 1.1809,
+      "loss": 1.178,
       "step": 6
     },
     {
       "epoch": 0.20512820512820512,
-      "grad_norm": 4.200634479522705,
+      "grad_norm": 4.262223720550537,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 1.0111,
+      "loss": 1.0102,
       "step": 7
     },
     {
       "epoch": 0.23443223443223443,
-      "grad_norm": 2.5703258514404297,
+      "grad_norm": 2.372178792953491,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 0.9798,
+      "loss": 0.9741,
       "step": 8
     },
     {
       "epoch": 0.26373626373626374,
-      "grad_norm": 1.4646650552749634,
+      "grad_norm": 1.6314536333084106,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 0.8646,
+      "loss": 0.8676,
       "step": 9
     },
     {
       "epoch": 0.29304029304029305,
-      "grad_norm": 1.0245696306228638,
+      "grad_norm": 1.1375904083251953,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 0.7594,
+      "loss": 0.7656,
       "step": 10
     },
     {
       "epoch": 0.32234432234432236,
-      "grad_norm": 0.7607440948486328,
+      "grad_norm": 0.7322179675102234,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 0.7369,
+      "loss": 0.738,
       "step": 11
     },
     {
       "epoch": 0.3516483516483517,
-      "grad_norm": 1.600371241569519,
+      "grad_norm": 1.6381927728652954,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 0.8084,
+      "loss": 0.8121,
       "step": 12
     },
     {
       "epoch": 0.38095238095238093,
-      "grad_norm": 1.577329397201538,
+      "grad_norm": 1.6779375076293945,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 0.7607,
+      "loss": 0.7658,
       "step": 13
     },
     {
       "epoch": 0.41025641025641024,
-      "grad_norm": 1.0018422603607178,
+      "grad_norm": 0.8208694458007812,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 0.8165,
+      "loss": 0.8118,
       "step": 14
     },
     {
       "epoch": 0.43956043956043955,
-      "grad_norm": 0.8424701690673828,
+      "grad_norm": 0.7266029119491577,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 0.8129,
+      "loss": 0.8152,
       "step": 15
     },
     {
       "epoch": 0.46886446886446886,
-      "grad_norm": 0.5086039900779724,
+      "grad_norm": 0.49156033992767334,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 0.8195,
+      "loss": 0.8198,
       "step": 16
     },
     {
       "epoch": 0.4981684981684982,
-      "grad_norm": 0.6842309236526489,
+      "grad_norm": 0.6357100009918213,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 0.8075,
+      "loss": 0.8063,
       "step": 17
     },
     {
       "epoch": 0.5274725274725275,
-      "grad_norm": 0.6705095767974854,
+      "grad_norm": 0.6192078590393066,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 0.7247,
+      "loss": 0.7221,
       "step": 18
     },
     {
       "epoch": 0.5567765567765568,
-      "grad_norm": 0.24981459975242615,
+      "grad_norm": 0.23790700733661652,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 0.741,
+      "loss": 0.7403,
       "step": 19
     },
     {
       "epoch": 0.5860805860805861,
-      "grad_norm": 0.3129333257675171,
+      "grad_norm": 0.3375350534915924,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 0.7393,
+      "loss": 0.7381,
       "step": 20
     },
     {
       "epoch": 0.6153846153846154,
-      "grad_norm": 0.7338513135910034,
+      "grad_norm": 0.7117923498153687,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 0.7904,
+      "loss": 0.7848,
       "step": 21
     },
     {
       "epoch": 0.6446886446886447,
-      "grad_norm": 1.3784418106079102,
+      "grad_norm": 1.3936290740966797,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 0.8909,
+      "loss": 0.8916,
       "step": 22
     },
     {
       "epoch": 0.673992673992674,
-      "grad_norm": 0.8394548296928406,
+      "grad_norm": 0.8301817178726196,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 0.8305,
+      "loss": 0.83,
       "step": 23
     },
     {
       "epoch": 0.7032967032967034,
-      "grad_norm": 0.29354768991470337,
+      "grad_norm": 0.2933414578437805,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 0.8116,
+      "loss": 0.8132,
       "step": 24
     },
     {
       "epoch": 0.7326007326007326,
-      "grad_norm": 0.3640207350254059,
+      "grad_norm": 0.3536495268344879,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 0.8095,
+      "loss": 0.8106,
       "step": 25
     },
     {
       "epoch": 0.7326007326007326,
-      "eval_loss": 0.789313018321991,
-      "eval_runtime": 12.0233,
-      "eval_samples_per_second": 19.13,
-      "eval_steps_per_second": 2.412,
+      "eval_loss": 0.7922340035438538,
+      "eval_runtime": 12.0794,
+      "eval_samples_per_second": 19.041,
+      "eval_steps_per_second": 2.401,
       "step": 25
     },
     {
       "epoch": 0.7619047619047619,
-      "grad_norm": 0.6007457375526428,
+      "grad_norm": 0.63962721824646,
       "learning_rate": 5e-05,
-      "loss": 0.8029,
+      "loss": 0.8015,
       "step": 26
     },
     {
       "epoch": 0.7912087912087912,
-      "grad_norm": 1.0814050436019897,
+      "grad_norm": 1.0908812284469604,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 0.7535,
+      "loss": 0.7551,
       "step": 27
     },
     {
       "epoch": 0.8205128205128205,
-      "grad_norm": 0.8970361351966858,
+      "grad_norm": 0.8861240744590759,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 0.7552,
+      "loss": 0.7574,
       "step": 28
     },
     {
       "epoch": 0.8498168498168498,
-      "grad_norm": 0.5584169030189514,
+      "grad_norm": 0.5566492080688477,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 0.772,
+      "loss": 0.7698,
       "step": 29
     },
     {
       "epoch": 0.8791208791208791,
-      "grad_norm": 0.19214436411857605,
+      "grad_norm": 0.20251761376857758,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 0.7811,
+      "loss": 0.7804,
       "step": 30
     },
     {
       "epoch": 0.9084249084249084,
-      "grad_norm": 0.4855462312698364,
+      "grad_norm": 0.4848445653915405,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 0.8256,
+      "loss": 0.8247,
       "step": 31
     },
     {
       "epoch": 0.9377289377289377,
-      "grad_norm": 0.5026677250862122,
+      "grad_norm": 0.4721270799636841,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 0.8139,
+      "loss": 0.8113,
       "step": 32
     },
     {
       "epoch": 0.967032967032967,
-      "grad_norm": 1.100959062576294,
+      "grad_norm": 1.1203545331954956,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 0.8822,
+      "loss": 0.8849,
       "step": 33
     },
     {
       "epoch": 0.9963369963369964,
-      "grad_norm": 0.678441047668457,
+      "grad_norm": 0.7001569867134094,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 0.8372,
+      "loss": 0.839,
       "step": 34
     },
     {
       "epoch": 1.0256410256410255,
-      "grad_norm": 1.023532748222351,
+      "grad_norm": 1.0713928937911987,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 1.4417,
+      "loss": 1.45,
       "step": 35
     },
     {
       "epoch": 1.054945054945055,
-      "grad_norm": 0.5481376647949219,
+      "grad_norm": 0.5490231513977051,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 0.7156,
+      "loss": 0.7151,
       "step": 36
     },
     {
       "epoch": 1.0842490842490842,
-      "grad_norm": 0.5916444659233093,
+      "grad_norm": 0.5858750343322754,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 0.7388,
+      "loss": 0.7387,
       "step": 37
     },
     {
       "epoch": 1.1135531135531136,
-      "grad_norm": 0.48780718445777893,
+      "grad_norm": 0.4807808995246887,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 0.7178,
+      "loss": 0.7169,
       "step": 38
     },
     {
       "epoch": 1.1428571428571428,
-      "grad_norm": 0.8632380366325378,
+      "grad_norm": 0.865768551826477,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 0.8168,
+      "loss": 0.8138,
       "step": 39
     },
     {
       "epoch": 1.1721611721611722,
-      "grad_norm": 0.6176807284355164,
+      "grad_norm": 0.5984993577003479,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 0.8144,
+      "loss": 0.8128,
       "step": 40
     },
     {
       "epoch": 1.2014652014652014,
-      "grad_norm": 0.3913164436817169,
+      "grad_norm": 0.4118782877922058,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 0.7581,
+      "loss": 0.7551,
       "step": 41
     },
     {
       "epoch": 1.2307692307692308,
-      "grad_norm": 0.5620902180671692,
+      "grad_norm": 0.555441677570343,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 0.8182,
+      "loss": 0.8145,
       "step": 42
     },
     {
       "epoch": 1.26007326007326,
-      "grad_norm": 0.5761415362358093,
+      "grad_norm": 0.5375956296920776,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 0.793,
+      "loss": 0.7918,
       "step": 43
     },
     {
       "epoch": 1.2893772893772895,
-      "grad_norm": 0.623988926410675,
+      "grad_norm": 0.5693370699882507,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 0.7259,
+      "loss": 0.7256,
       "step": 44
     },
     {
       "epoch": 1.3186813186813187,
-      "grad_norm": 0.6083937287330627,
+      "grad_norm": 0.5636836290359497,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 0.7406,
+      "loss": 0.7449,
       "step": 45
     },
     {
       "epoch": 1.347985347985348,
-      "grad_norm": 0.6067966818809509,
+      "grad_norm": 0.5880256295204163,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 0.7433,
+      "loss": 0.742,
       "step": 46
     },
     {
       "epoch": 1.3772893772893773,
-      "grad_norm": 0.5738537311553955,
+      "grad_norm": 0.5251706838607788,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 0.7611,
+      "loss": 0.763,
       "step": 47
     },
     {
       "epoch": 1.4065934065934065,
-      "grad_norm": 0.5093715190887451,
+      "grad_norm": 0.49943527579307556,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 0.7872,
+      "loss": 0.7867,
       "step": 48
     },
     {
       "epoch": 1.435897435897436,
-      "grad_norm": 0.4598458707332611,
+      "grad_norm": 0.4566574692726135,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 0.804,
+      "loss": 0.8033,
       "step": 49
     },
     {
       "epoch": 1.4652014652014653,
-      "grad_norm": 0.6957669258117676,
+      "grad_norm": 0.6736223101615906,
       "learning_rate": 0.0,
-      "loss": 0.8004,
+      "loss": 0.798,
       "step": 50
     },
     {
       "epoch": 1.4652014652014653,
-      "eval_loss": 0.7666683197021484,
-      "eval_runtime": 12.0189,
-      "eval_samples_per_second": 19.137,
-      "eval_steps_per_second": 2.413,
+      "eval_loss": 0.7679101824760437,
+      "eval_runtime": 12.0726,
+      "eval_samples_per_second": 19.051,
+      "eval_steps_per_second": 2.402,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:768050f293879227102c7c91bc296e82e7ed39164763de483fc202082d0b9cf5
+oid sha256:071494bbe95931a08631bf47f2dba38f70b866d5bb370d5380baf7279eff516c
 size 6776