Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "q_proj",
+    "k_proj",
+    "up_proj",
     "down_proj",
-    "v_proj",
     "o_proj",
-    "up_proj",
-    "q_proj",
     "gate_proj",
-    "k_proj"
+    "v_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 sha256:053f00e208bb4935692221b132ec91dc2c991bb521e6202622701d52d411f4d3
+oid sha256:6c6f6d311ca62365ea4617d04edfaf28f0f3049f9e56e2fb031e907c5c9de6ef
 size 335604696

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c4517cdbed62a7ea761881afc98251bfdf147fff90f9eed24884b067fd2777fc
+oid sha256:0950dc218f013d1bd2b3d7b666856263068ac8ef94064f77ce77a23715bc5cad
 size 671466706

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 1.6284352540969849,
+  "best_metric": 1.6282744407653809,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.01835662329914412,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.00036713246598288243,
-      "grad_norm": 0.31078195571899414,
+      "grad_norm": 0.2898488938808441,
       "learning_rate": 5e-05,
       "loss": 1.3837,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.00036713246598288243,
       "eval_loss": 1.8131353855133057,
-      "eval_runtime": 550.3091,
-      "eval_samples_per_second": 33.345,
-      "eval_steps_per_second": 4.169,
+      "eval_runtime": 549.3405,
+      "eval_samples_per_second": 33.404,
+      "eval_steps_per_second": 4.176,
       "step": 1
     },
     {
       "epoch": 0.0007342649319657649,
-      "grad_norm": 0.2641288936138153,
+      "grad_norm": 0.25190261006355286,
       "learning_rate": 0.0001,
       "loss": 1.3587,
       "step": 2
     },
     {
       "epoch": 0.0011013973979486474,
-      "grad_norm": 0.27233171463012695,
+      "grad_norm": 0.26423922181129456,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 1.48,
+      "loss": 1.4802,
       "step": 3
     },
     {
       "epoch": 0.0014685298639315297,
-      "grad_norm": 0.25501444935798645,
+      "grad_norm": 0.24811314046382904,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 1.549,
+      "loss": 1.5491,
       "step": 4
     },
     {
       "epoch": 0.0018356623299144123,
-      "grad_norm": 0.3122928738594055,
+      "grad_norm": 0.30088868737220764,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 1.4885,
+      "loss": 1.4884,
       "step": 5
     },
     {
       "epoch": 0.002202794795897295,
-      "grad_norm": 0.318195641040802,
+      "grad_norm": 0.30977341532707214,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 1.6574,
+      "loss": 1.6573,
       "step": 6
     },
     {
       "epoch": 0.0025699272618801773,
-      "grad_norm": 0.3191247284412384,
+      "grad_norm": 0.31186559796333313,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 1.7707,
+      "loss": 1.771,
       "step": 7
     },
     {
       "epoch": 0.0029370597278630594,
-      "grad_norm": 0.3157280385494232,
+      "grad_norm": 0.3099216818809509,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 1.7948,
+      "loss": 1.7946,
       "step": 8
     },
     {
       "epoch": 0.003304192193845942,
-      "grad_norm": 0.3386346399784088,
+      "grad_norm": 0.33014586567878723,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 1.9268,
+      "loss": 1.9267,
       "step": 9
     },
     {
       "epoch": 0.0036713246598288245,
-      "grad_norm": 0.33855950832366943,
+      "grad_norm": 0.3368901014328003,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 1.8802,
+      "loss": 1.8801,
       "step": 10
     },
     {
       "epoch": 0.004038457125811707,
-      "grad_norm": 0.37005379796028137,
+      "grad_norm": 0.3644021153450012,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 1.8973,
+      "loss": 1.8974,
       "step": 11
     },
     {
       "epoch": 0.00440558959179459,
-      "grad_norm": 0.6089059114456177,
+      "grad_norm": 0.5990589261054993,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 2.0231,
+      "loss": 2.0226,
       "step": 12
     },
     {
       "epoch": 0.004772722057777472,
-      "grad_norm": 0.6466706991195679,
+      "grad_norm": 0.6255826354026794,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 1.4021,
+      "loss": 1.4007,
       "step": 13
     },
     {
       "epoch": 0.005139854523760355,
-      "grad_norm": 0.5018526911735535,
+      "grad_norm": 0.48832088708877563,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 1.3515,
+      "loss": 1.3507,
       "step": 14
     },
     {
       "epoch": 0.005506986989743236,
-      "grad_norm": 0.2542576491832733,
+      "grad_norm": 0.24915535748004913,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 1.3624,
+      "loss": 1.362,
       "step": 15
     },
     {
       "epoch": 0.005874119455726119,
-      "grad_norm": 0.23958976566791534,
+      "grad_norm": 0.23602904379367828,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 1.4347,
+      "loss": 1.4343,
       "step": 16
     },
     {
       "epoch": 0.006241251921709001,
-      "grad_norm": 0.2992570102214813,
+      "grad_norm": 0.29532483220100403,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 1.5112,
+      "loss": 1.5104,
       "step": 17
     },
     {
       "epoch": 0.006608384387691884,
-      "grad_norm": 0.3050135374069214,
+      "grad_norm": 0.2998216450214386,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 1.5898,
+      "loss": 1.5892,
       "step": 18
     },
     {
       "epoch": 0.0069755168536747665,
-      "grad_norm": 0.316407710313797,
+      "grad_norm": 0.3126564919948578,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 1.6934,
+      "loss": 1.6929,
       "step": 19
     },
     {
       "epoch": 0.007342649319657649,
-      "grad_norm": 0.26744428277015686,
+      "grad_norm": 0.26496151089668274,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 1.7074,
+      "loss": 1.7072,
       "step": 20
     },
     {
       "epoch": 0.0077097817856405315,
-      "grad_norm": 0.2763262391090393,
+      "grad_norm": 0.27758315205574036,
       "learning_rate": 6.607197326515808e-05,
       "loss": 1.8256,
       "step": 21
     },
     {
       "epoch": 0.008076914251623414,
-      "grad_norm": 0.26780959963798523,
+      "grad_norm": 0.2669735550880432,
       "learning_rate": 6.294095225512603e-05,
       "loss": 1.7701,
       "step": 22
     },
     {
       "epoch": 0.008444046717606296,
-      "grad_norm": 0.30958154797554016,
+      "grad_norm": 0.30628520250320435,
       "learning_rate": 5.9754516100806423e-05,
       "loss": 1.9034,
       "step": 23
     },
     {
       "epoch": 0.00881117918358918,
-      "grad_norm": 0.38749268651008606,
+      "grad_norm": 0.38713809847831726,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 1.9776,
+      "loss": 1.977,
       "step": 24
     },
     {
       "epoch": 0.00917831164957206,
-      "grad_norm": 0.6935418844223022,
+      "grad_norm": 0.690243661403656,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 2.0543,
+      "loss": 2.0544,
       "step": 25
     },
     {
       "epoch": 0.00917831164957206,
-      "eval_loss": 1.6496174335479736,
-      "eval_runtime": 550.8006,
-      "eval_samples_per_second": 33.315,
-      "eval_steps_per_second": 4.165,
+      "eval_loss": 1.6493293046951294,
+      "eval_runtime": 550.3826,
+      "eval_samples_per_second": 33.34,
+      "eval_steps_per_second": 4.168,
       "step": 25
     },
     {
       "epoch": 0.009545444115554944,
-      "grad_norm": 0.4805835485458374,
+      "grad_norm": 0.4602947235107422,
       "learning_rate": 5e-05,
       "loss": 1.3293,
       "step": 26
     },
     {
       "epoch": 0.009912576581537826,
-      "grad_norm": 0.43551477789878845,
+      "grad_norm": 0.41046762466430664,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 1.3279,
+      "loss": 1.3277,
       "step": 27
     },
     {
       "epoch": 0.01027970904752071,
-      "grad_norm": 0.41012075543403625,
+      "grad_norm": 0.38931915163993835,
       "learning_rate": 4.347369038899744e-05,
       "loss": 1.3236,
       "step": 28
     },
     {
       "epoch": 0.010646841513503591,
-      "grad_norm": 0.29394716024398804,
+      "grad_norm": 0.2806582450866699,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 1.4154,
+      "loss": 1.415,
       "step": 29
     },
     {
       "epoch": 0.011013973979486473,
-      "grad_norm": 0.21994492411613464,
+      "grad_norm": 0.21503187716007233,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 1.4519,
+      "loss": 1.4517,
       "step": 30
     },
     {
       "epoch": 0.011381106445469356,
-      "grad_norm": 0.22759422659873962,
+      "grad_norm": 0.22820733487606049,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 1.5519,
+      "loss": 1.5522,
       "step": 31
     },
     {
       "epoch": 0.011748238911452238,
-      "grad_norm": 0.22850295901298523,
+      "grad_norm": 0.22769905626773834,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 1.6661,
+      "loss": 1.666,
       "step": 32
     },
     {
       "epoch": 0.012115371377435121,
-      "grad_norm": 0.27733510732650757,
+      "grad_norm": 0.2772124111652374,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 1.749,
+      "loss": 1.7495,
       "step": 33
     },
     {
       "epoch": 0.012482503843418003,
-      "grad_norm": 0.3003651797771454,
+      "grad_norm": 0.29833248257637024,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 1.8519,
+      "loss": 1.8522,
       "step": 34
     },
     {
       "epoch": 0.012849636309400886,
-      "grad_norm": 0.33368366956710815,
+      "grad_norm": 0.32912251353263855,
       "learning_rate": 2.2221488349019903e-05,
       "loss": 1.9284,
       "step": 35
     },
     {
       "epoch": 0.013216768775383768,
-      "grad_norm": 0.3779643476009369,
+      "grad_norm": 0.37289896607398987,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 1.8587,
+      "loss": 1.859,
       "step": 36
     },
     {
       "epoch": 0.013583901241366651,
-      "grad_norm": 0.45034879446029663,
+      "grad_norm": 0.44199585914611816,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 2.0736,
+      "loss": 2.0734,
       "step": 37
     },
     {
       "epoch": 0.013951033707349533,
-      "grad_norm": 0.23827709257602692,
+      "grad_norm": 0.24129906296730042,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 1.3326,
+      "loss": 1.3332,
       "step": 38
     },
     {
       "epoch": 0.014318166173332416,
-      "grad_norm": 0.19718056917190552,
+      "grad_norm": 0.19583018124103546,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 1.2701,
+      "loss": 1.2705,
       "step": 39
     },
     {
       "epoch": 0.014685298639315298,
-      "grad_norm": 0.1556357443332672,
+      "grad_norm": 0.15503276884555817,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 1.2748,
+      "loss": 1.2746,
       "step": 40
     },
     {
       "epoch": 0.01505243110529818,
-      "grad_norm": 0.15502218902111053,
+      "grad_norm": 0.15353931486606598,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 1.408,
+      "loss": 1.4078,
       "step": 41
     },
     {
       "epoch": 0.015419563571281063,
-      "grad_norm": 0.16751334071159363,
+      "grad_norm": 0.16584447026252747,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 1.4046,
+      "loss": 1.4045,
       "step": 42
     },
     {
       "epoch": 0.015786696037263945,
-      "grad_norm": 0.174475759267807,
+      "grad_norm": 0.17310383915901184,
       "learning_rate": 5.156362923365588e-06,
       "loss": 1.5088,
       "step": 43
     },
     {
       "epoch": 0.016153828503246828,
-      "grad_norm": 0.1867865025997162,
+      "grad_norm": 0.18577170372009277,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 1.6376,
+      "loss": 1.6375,
       "step": 44
     },
     {
       "epoch": 0.01652096096922971,
-      "grad_norm": 0.19639185070991516,
+      "grad_norm": 0.19366207718849182,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 1.5946,
+      "loss": 1.5945,
       "step": 45
     },
     {
       "epoch": 0.01688809343521259,
-      "grad_norm": 0.22939303517341614,
+      "grad_norm": 0.227140873670578,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 1.7621,
+      "loss": 1.7622,
       "step": 46
     },
     {
       "epoch": 0.017255225901195475,
-      "grad_norm": 0.26366662979125977,
+      "grad_norm": 0.2605684995651245,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 1.9381,
+      "loss": 1.9384,
       "step": 47
     },
     {
       "epoch": 0.01762235836717836,
-      "grad_norm": 0.2914908230304718,
+      "grad_norm": 0.28882354497909546,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 1.9049,
+      "loss": 1.9044,
       "step": 48
     },
     {
       "epoch": 0.017989490833161238,
-      "grad_norm": 0.3384196162223816,
+      "grad_norm": 0.33389711380004883,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 1.966,
+      "loss": 1.9658,
       "step": 49
     },
     {
       "epoch": 0.01835662329914412,
-      "grad_norm": 0.6291220188140869,
+      "grad_norm": 0.6271953582763672,
       "learning_rate": 0.0,
-      "loss": 2.0628,
+      "loss": 2.0631,
       "step": 50
     },
     {
       "epoch": 0.01835662329914412,
-      "eval_loss": 1.6284352540969849,
-      "eval_runtime": 550.6091,
-      "eval_samples_per_second": 33.327,
-      "eval_steps_per_second": 4.166,
+      "eval_loss": 1.6282744407653809,
+      "eval_runtime": 551.4838,
+      "eval_samples_per_second": 33.274,
+      "eval_steps_per_second": 4.16,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ade93211eed489b83b485a4719a06ffc515c7f15da156c1122a548f282a85e95
+oid sha256:d8926273349e9723e2cc347ed7fcdd321c18137a8eb161c01155b353634b1038
 size 6776