Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,12 +20,12 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "down_proj",
+    "q_proj",
     "gate_proj",
-    "v_proj",
+    "down_proj",
     "k_proj",
-    "q_proj",
     "up_proj",
+    "v_proj",
     "o_proj"
   ],
   "task_type": "CAUSAL_LM",

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:720a6adacc92b01cc7a7e208531ebdbd12ba34c21beaa012b8307f43f61ec542
+oid sha256:7d19c7fd5bc202505f68bcf60150cec0d333afabff305c6eac7f68ed7a62d7d4
 size 119975656

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e84c5640d21f9c27d917fb8063994739fd4bf2368ff299acc4018f383d2401e5
+oid sha256:949404b5aa3e3091bb0fb815d3392f8d4d9d92d114cdbebaadbae16389bf9ced
 size 240145026

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.6820244193077087,
+  "best_metric": 0.6818761825561523,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.1345668629100084,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.002691337258200168,
-      "grad_norm": 0.2526223957538605,
+      "grad_norm": 0.2545667290687561,
       "learning_rate": 5e-05,
       "loss": 0.8201,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.002691337258200168,
       "eval_loss": 0.967555820941925,
-      "eval_runtime": 27.5342,
-      "eval_samples_per_second": 90.941,
-      "eval_steps_per_second": 11.368,
+      "eval_runtime": 27.8167,
+      "eval_samples_per_second": 90.018,
+      "eval_steps_per_second": 11.252,
       "step": 1
     },
     {
       "epoch": 0.005382674516400336,
-      "grad_norm": 0.34248781204223633,
+      "grad_norm": 0.34543758630752563,
       "learning_rate": 0.0001,
       "loss": 0.9103,
       "step": 2
     },
     {
       "epoch": 0.008074011774600504,
-      "grad_norm": 0.3343915343284607,
+      "grad_norm": 0.33696243166923523,
       "learning_rate": 9.989294616193017e-05,
       "loss": 0.933,
       "step": 3
     },
     {
       "epoch": 0.010765349032800672,
-      "grad_norm": 0.32323670387268066,
+      "grad_norm": 0.3235072195529938,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 0.9152,
+      "loss": 0.9148,
       "step": 4
     },
     {
       "epoch": 0.01345668629100084,
-      "grad_norm": 0.2897728383541107,
+      "grad_norm": 0.29262658953666687,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 0.8206,
+      "loss": 0.8203,
       "step": 5
     },
     {
       "epoch": 0.01614802354920101,
-      "grad_norm": 0.25488898158073425,
+      "grad_norm": 0.2570266127586365,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 0.8321,
+      "loss": 0.8314,
       "step": 6
     },
     {
       "epoch": 0.018839360807401177,
-      "grad_norm": 0.22519971430301666,
+      "grad_norm": 0.22506366670131683,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 0.7606,
+      "loss": 0.7602,
       "step": 7
     },
     {
       "epoch": 0.021530698065601345,
-      "grad_norm": 0.2275712788105011,
+      "grad_norm": 0.22845478355884552,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 0.7751,
+      "loss": 0.7756,
       "step": 8
     },
     {
       "epoch": 0.024222035323801513,
-      "grad_norm": 0.3145010769367218,
+      "grad_norm": 0.31451737880706787,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 0.8015,
+      "loss": 0.8014,
       "step": 9
     },
     {
       "epoch": 0.02691337258200168,
-      "grad_norm": 0.30495619773864746,
+      "grad_norm": 0.3066512644290924,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 0.8082,
+      "loss": 0.8079,
       "step": 10
     },
     {
       "epoch": 0.02960470984020185,
-      "grad_norm": 0.2785869538784027,
+      "grad_norm": 0.27897998690605164,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 0.7656,
+      "loss": 0.7658,
       "step": 11
     },
     {
       "epoch": 0.03229604709840202,
-      "grad_norm": 0.23609785735607147,
+      "grad_norm": 0.23739658296108246,
       "learning_rate": 8.966766701456177e-05,
       "loss": 0.7395,
       "step": 12
     },
     {
       "epoch": 0.03498738435660219,
-      "grad_norm": 0.1729651838541031,
+      "grad_norm": 0.1737861931324005,
       "learning_rate": 8.759199037394887e-05,
       "loss": 0.645,
       "step": 13
     },
     {
       "epoch": 0.03767872161480235,
-      "grad_norm": 0.1702946424484253,
+      "grad_norm": 0.17162172496318817,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 0.7152,
+      "loss": 0.7153,
       "step": 14
     },
     {
       "epoch": 0.040370058873002525,
-      "grad_norm": 0.14832372963428497,
+      "grad_norm": 0.15028245747089386,
       "learning_rate": 8.296729075500344e-05,
       "loss": 0.7591,
       "step": 15
     },
     {
       "epoch": 0.04306139613120269,
-      "grad_norm": 0.15146780014038086,
+      "grad_norm": 0.1524060070514679,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 0.7415,
+      "loss": 0.741,
       "step": 16
     },
     {
       "epoch": 0.04575273338940286,
-      "grad_norm": 0.1504327803850174,
+      "grad_norm": 0.1524454951286316,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 0.7213,
+      "loss": 0.7214,
       "step": 17
     },
     {
       "epoch": 0.048444070647603026,
-      "grad_norm": 0.1591215282678604,
+      "grad_norm": 0.16076746582984924,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 0.7222,
+      "loss": 0.7221,
       "step": 18
     },
     {
       "epoch": 0.0511354079058032,
-      "grad_norm": 0.16221410036087036,
+      "grad_norm": 0.16362418234348297,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 0.7569,
+      "loss": 0.7566,
       "step": 19
     },
     {
       "epoch": 0.05382674516400336,
-      "grad_norm": 0.16809888184070587,
+      "grad_norm": 0.16996099054813385,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 0.7021,
+      "loss": 0.7022,
       "step": 20
     },
     {
       "epoch": 0.05651808242220353,
-      "grad_norm": 0.16224785149097443,
+      "grad_norm": 0.16316403448581696,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 0.7158,
+      "loss": 0.7155,
       "step": 21
     },
     {
       "epoch": 0.0592094196804037,
-      "grad_norm": 0.16932708024978638,
+      "grad_norm": 0.16969940066337585,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 0.7172,
+      "loss": 0.7168,
       "step": 22
     },
     {
       "epoch": 0.06190075693860387,
-      "grad_norm": 0.17013271152973175,
+      "grad_norm": 0.17208510637283325,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 0.7397,
+      "loss": 0.739,
       "step": 23
     },
     {
       "epoch": 0.06459209419680403,
-      "grad_norm": 0.17485541105270386,
+      "grad_norm": 0.17505396902561188,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 0.6985,
+      "loss": 0.6977,
       "step": 24
     },
     {
       "epoch": 0.0672834314550042,
-      "grad_norm": 0.18603037297725677,
+      "grad_norm": 0.18648380041122437,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 0.6761,
+      "loss": 0.6757,
       "step": 25
     },
     {
       "epoch": 0.0672834314550042,
-      "eval_loss": 0.693745493888855,
-      "eval_runtime": 27.5134,
-      "eval_samples_per_second": 91.01,
-      "eval_steps_per_second": 11.376,
+      "eval_loss": 0.693597674369812,
+      "eval_runtime": 27.2048,
+      "eval_samples_per_second": 92.042,
+      "eval_steps_per_second": 11.505,
       "step": 25
     },
     {
       "epoch": 0.06997476871320438,
-      "grad_norm": 0.1348254233598709,
+      "grad_norm": 0.13479048013687134,
       "learning_rate": 5e-05,
-      "loss": 0.6984,
+      "loss": 0.6979,
       "step": 26
     },
     {
       "epoch": 0.07266610597140453,
-      "grad_norm": 0.14816786348819733,
+      "grad_norm": 0.14756473898887634,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 0.7012,
+      "loss": 0.7009,
       "step": 27
     },
     {
       "epoch": 0.0753574432296047,
-      "grad_norm": 0.1404827982187271,
+      "grad_norm": 0.14167127013206482,
       "learning_rate": 4.347369038899744e-05,
       "loss": 0.6821,
       "step": 28
     },
     {
       "epoch": 0.07804878048780488,
-      "grad_norm": 0.14742571115493774,
+      "grad_norm": 0.14789500832557678,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 0.7106,
+      "loss": 0.7102,
       "step": 29
     },
     {
       "epoch": 0.08074011774600505,
-      "grad_norm": 0.14235340058803558,
+      "grad_norm": 0.14351996779441833,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 0.7429,
+      "loss": 0.7426,
       "step": 30
     },
     {
       "epoch": 0.08343145500420522,
-      "grad_norm": 0.13567449152469635,
+      "grad_norm": 0.136067196726799,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 0.6724,
+      "loss": 0.6718,
       "step": 31
     },
     {
       "epoch": 0.08612279226240538,
-      "grad_norm": 0.14754807949066162,
+      "grad_norm": 0.14951838552951813,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 0.689,
+      "loss": 0.6889,
       "step": 32
     },
     {
       "epoch": 0.08881412952060555,
-      "grad_norm": 0.14674997329711914,
+      "grad_norm": 0.14795440435409546,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 0.7101,
+      "loss": 0.7103,
       "step": 33
     },
     {
       "epoch": 0.09150546677880572,
-      "grad_norm": 0.14950957894325256,
+      "grad_norm": 0.15034087002277374,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 0.696,
+      "loss": 0.6955,
       "step": 34
     },
     {
       "epoch": 0.0941968040370059,
-      "grad_norm": 0.1410410851240158,
+      "grad_norm": 0.14145706593990326,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 0.6993,
+      "loss": 0.6991,
       "step": 35
     },
     {
       "epoch": 0.09688814129520605,
-      "grad_norm": 0.15637695789337158,
+      "grad_norm": 0.1578945815563202,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 0.7033,
+      "loss": 0.7034,
       "step": 36
     },
     {
       "epoch": 0.09957947855340622,
-      "grad_norm": 0.1605268269777298,
+      "grad_norm": 0.16180913150310516,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 0.6651,
+      "loss": 0.6644,
       "step": 37
     },
     {
       "epoch": 0.1022708158116064,
-      "grad_norm": 0.12332131713628769,
+      "grad_norm": 0.12289156764745712,
       "learning_rate": 1.4644660940672627e-05,
       "loss": 0.6528,
       "step": 38
     },
     {
       "epoch": 0.10496215306980657,
-      "grad_norm": 0.11790407449007034,
+      "grad_norm": 0.11805767565965652,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 0.6801,
+      "loss": 0.68,
       "step": 39
     },
     {
       "epoch": 0.10765349032800672,
-      "grad_norm": 0.1261812448501587,
+      "grad_norm": 0.12661293148994446,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 0.7404,
+      "loss": 0.74,
       "step": 40
     },
     {
       "epoch": 0.1103448275862069,
-      "grad_norm": 0.12297671288251877,
+      "grad_norm": 0.12309448421001434,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 0.7104,
+      "loss": 0.7105,
       "step": 41
     },
     {
       "epoch": 0.11303616484440707,
-      "grad_norm": 0.13225741684436798,
+      "grad_norm": 0.1316811442375183,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 0.7035,
+      "loss": 0.7034,
       "step": 42
     },
     {
       "epoch": 0.11572750210260724,
-      "grad_norm": 0.12430209666490555,
+      "grad_norm": 0.12469199299812317,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 0.7073,
+      "loss": 0.7072,
       "step": 43
     },
     {
       "epoch": 0.1184188393608074,
-      "grad_norm": 0.13539482653141022,
+      "grad_norm": 0.13566303253173828,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 0.7081,
+      "loss": 0.708,
       "step": 44
     },
     {
       "epoch": 0.12111017661900757,
-      "grad_norm": 0.1353532373905182,
+      "grad_norm": 0.13655786216259003,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 0.6732,
+      "loss": 0.6734,
       "step": 45
     },
     {
       "epoch": 0.12380151387720774,
-      "grad_norm": 0.1359177678823471,
+      "grad_norm": 0.13654503226280212,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 0.6587,
+      "loss": 0.6583,
       "step": 46
     },
     {
       "epoch": 0.1264928511354079,
-      "grad_norm": 0.1404392421245575,
+      "grad_norm": 0.14090988039970398,
       "learning_rate": 9.607359798384785e-07,
       "loss": 0.7098,
       "step": 47
     },
     {
       "epoch": 0.12918418839360807,
-      "grad_norm": 0.14978425204753876,
+      "grad_norm": 0.1505599319934845,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 0.7064,
+      "loss": 0.7067,
       "step": 48
     },
     {
       "epoch": 0.13187552565180824,
-      "grad_norm": 0.15383632481098175,
+      "grad_norm": 0.15436117351055145,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 0.6942,
+      "loss": 0.6933,
       "step": 49
     },
     {
       "epoch": 0.1345668629100084,
-      "grad_norm": 0.19792960584163666,
+      "grad_norm": 0.19895680248737335,
       "learning_rate": 0.0,
-      "loss": 0.6501,
+      "loss": 0.6498,
       "step": 50
     },
     {
       "epoch": 0.1345668629100084,
-      "eval_loss": 0.6820244193077087,
-      "eval_runtime": 27.8817,
-      "eval_samples_per_second": 89.808,
-      "eval_steps_per_second": 11.226,
+      "eval_loss": 0.6818761825561523,
+      "eval_runtime": 28.2352,
+      "eval_samples_per_second": 88.684,
+      "eval_steps_per_second": 11.085,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0e354e722c399dce1e3616f5bb46a767ebae1746ae68640eca13cfd6c910d02c
+oid sha256:b11b5c4e0c23f04692ebeb2adbe14349b333df5103e6a6685e5ab295a0c2fe18
 size 6776