Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -21,12 +21,12 @@
   "revision": null,
   "target_modules": [
     "q_proj",
-    "v_proj",
-    "up_proj",
     "gate_proj",
+    "k_proj",
     "down_proj",
     "o_proj",
-    "k_proj"
+    "v_proj",
+    "up_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:a334af04127f5624a904703d6ecadb97c848b90372ed8a7447abecd288cd4255
+oid sha256:77de871694caae9be418270cfe1846fd732ec0169d8d218391fbaec100b30020
 size 144748392

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:837cff29788f064872ebddf030fc8639040dd9765a289ea0d3df0c27224fcf43
+oid sha256:aea083e221f4a6a17827994804d6c0b978faa3e0bf3f67e6fa729d8345397513
 size 289690242

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 3.1869354248046875,
+  "best_metric": 3.1944468021392822,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.6920415224913494,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.01384083044982699,
-      "grad_norm": 0.7818406820297241,
+      "grad_norm": 0.7828583121299744,
       "learning_rate": 5e-05,
       "loss": 6.6306,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.01384083044982699,
       "eval_loss": 6.805291175842285,
-      "eval_runtime": 6.2687,
-      "eval_samples_per_second": 77.687,
-      "eval_steps_per_second": 9.731,
+      "eval_runtime": 6.2044,
+      "eval_samples_per_second": 78.493,
+      "eval_steps_per_second": 9.832,
       "step": 1
     },
     {
       "epoch": 0.02768166089965398,
-      "grad_norm": 0.8963271379470825,
+      "grad_norm": 0.900705099105835,
       "learning_rate": 0.0001,
       "loss": 6.4903,
       "step": 2
     },
     {
       "epoch": 0.04152249134948097,
-      "grad_norm": 0.684287965297699,
+      "grad_norm": 0.6845119595527649,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 6.5287,
+      "loss": 6.5309,
       "step": 3
     },
     {
       "epoch": 0.05536332179930796,
-      "grad_norm": 0.732020914554596,
+      "grad_norm": 0.7316877245903015,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 6.4374,
+      "loss": 6.4378,
       "step": 4
     },
     {
       "epoch": 0.06920415224913495,
-      "grad_norm": 0.7713181972503662,
+      "grad_norm": 0.7706010937690735,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 6.4095,
+      "loss": 6.408,
       "step": 5
     },
     {
       "epoch": 0.08304498269896193,
-      "grad_norm": 0.9216402173042297,
+      "grad_norm": 0.9191716313362122,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 6.4433,
+      "loss": 6.4436,
       "step": 6
     },
     {
       "epoch": 0.09688581314878893,
-      "grad_norm": 0.9989214539527893,
+      "grad_norm": 0.9976472854614258,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 6.3937,
+      "loss": 6.3941,
       "step": 7
     },
     {
       "epoch": 0.11072664359861592,
-      "grad_norm": 1.2477836608886719,
+      "grad_norm": 1.2480906248092651,
       "learning_rate": 9.619397662556435e-05,
       "loss": 6.3564,
       "step": 8
     },
     {
       "epoch": 0.1245674740484429,
-      "grad_norm": 1.5049328804016113,
+      "grad_norm": 1.514481782913208,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 6.3056,
+      "loss": 6.3036,
       "step": 9
     },
     {
       "epoch": 0.1384083044982699,
-      "grad_norm": 1.6154811382293701,
+      "grad_norm": 1.6174757480621338,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 6.1173,
+      "loss": 6.1196,
       "step": 10
     },
     {
       "epoch": 0.1522491349480969,
-      "grad_norm": 1.7500627040863037,
+      "grad_norm": 1.7518984079360962,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 6.1466,
+      "loss": 6.1501,
       "step": 11
     },
     {
       "epoch": 0.16608996539792387,
-      "grad_norm": 1.8479565382003784,
+      "grad_norm": 1.8451406955718994,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 5.8452,
+      "loss": 5.8514,
       "step": 12
     },
     {
       "epoch": 0.17993079584775087,
-      "grad_norm": 1.4685908555984497,
+      "grad_norm": 1.4596737623214722,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 5.8335,
+      "loss": 5.8396,
       "step": 13
     },
     {
       "epoch": 0.19377162629757785,
-      "grad_norm": 1.3053746223449707,
+      "grad_norm": 1.310354471206665,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 5.0541,
+      "loss": 5.0548,
       "step": 14
     },
     {
       "epoch": 0.20761245674740483,
-      "grad_norm": 1.2436093091964722,
+      "grad_norm": 1.2472784519195557,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 4.8929,
+      "loss": 4.8912,
       "step": 15
     },
     {
       "epoch": 0.22145328719723184,
-      "grad_norm": 1.3153407573699951,
+      "grad_norm": 1.3219356536865234,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 4.9075,
+      "loss": 4.9125,
       "step": 16
     },
     {
       "epoch": 0.23529411764705882,
-      "grad_norm": 1.4282431602478027,
+      "grad_norm": 1.4386686086654663,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 4.8483,
+      "loss": 4.8531,
       "step": 17
     },
     {
       "epoch": 0.2491349480968858,
-      "grad_norm": 1.4525681734085083,
+      "grad_norm": 1.4553345441818237,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 4.7716,
+      "loss": 4.776,
       "step": 18
     },
     {
       "epoch": 0.2629757785467128,
-      "grad_norm": 1.364906668663025,
+      "grad_norm": 1.3721455335617065,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 4.4986,
+      "loss": 4.5001,
       "step": 19
     },
     {
       "epoch": 0.2768166089965398,
-      "grad_norm": 1.6167612075805664,
+      "grad_norm": 1.618059754371643,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 4.3998,
+      "loss": 4.401,
       "step": 20
     },
     {
       "epoch": 0.2906574394463668,
-      "grad_norm": 1.840979814529419,
+      "grad_norm": 1.8476649522781372,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 4.1656,
+      "loss": 4.1677,
       "step": 21
     },
     {
       "epoch": 0.3044982698961938,
-      "grad_norm": 1.9929301738739014,
+      "grad_norm": 2.0024266242980957,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 4.3912,
+      "loss": 4.3977,
       "step": 22
     },
     {
       "epoch": 0.31833910034602075,
-      "grad_norm": 1.9383858442306519,
+      "grad_norm": 1.9315515756607056,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 4.2083,
+      "loss": 4.2169,
       "step": 23
     },
     {
       "epoch": 0.33217993079584773,
-      "grad_norm": 1.7524367570877075,
+      "grad_norm": 1.7556025981903076,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 3.9833,
+      "loss": 3.9926,
       "step": 24
     },
     {
       "epoch": 0.3460207612456747,
-      "grad_norm": 1.6341569423675537,
+      "grad_norm": 1.6433321237564087,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 4.2874,
+      "loss": 4.3006,
       "step": 25
     },
     {
       "epoch": 0.3460207612456747,
-      "eval_loss": 3.875951051712036,
-      "eval_runtime": 6.2835,
-      "eval_samples_per_second": 77.505,
-      "eval_steps_per_second": 9.708,
+      "eval_loss": 3.8810713291168213,
+      "eval_runtime": 6.2008,
+      "eval_samples_per_second": 78.538,
+      "eval_steps_per_second": 9.837,
       "step": 25
     },
     {
       "epoch": 0.35986159169550175,
-      "grad_norm": 1.246837854385376,
+      "grad_norm": 1.2370139360427856,
       "learning_rate": 5e-05,
-      "loss": 3.906,
+      "loss": 3.9109,
       "step": 26
     },
     {
       "epoch": 0.3737024221453287,
-      "grad_norm": 1.053470253944397,
+      "grad_norm": 1.0478001832962036,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 3.7484,
+      "loss": 3.7508,
       "step": 27
     },
     {
       "epoch": 0.3875432525951557,
-      "grad_norm": 1.0002684593200684,
+      "grad_norm": 0.9928375482559204,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 3.7783,
+      "loss": 3.7824,
       "step": 28
     },
     {
       "epoch": 0.4013840830449827,
-      "grad_norm": 0.993694007396698,
+      "grad_norm": 0.9907777905464172,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 3.6615,
+      "loss": 3.6651,
       "step": 29
     },
     {
       "epoch": 0.41522491349480967,
-      "grad_norm": 1.0306425094604492,
+      "grad_norm": 1.028157114982605,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 3.5937,
+      "loss": 3.5988,
       "step": 30
     },
     {
       "epoch": 0.4290657439446367,
-      "grad_norm": 1.0861804485321045,
+      "grad_norm": 1.0831353664398193,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 3.4711,
+      "loss": 3.4734,
       "step": 31
     },
     {
       "epoch": 0.4429065743944637,
-      "grad_norm": 1.1918703317642212,
+      "grad_norm": 1.1880797147750854,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 3.402,
+      "loss": 3.4019,
       "step": 32
     },
     {
       "epoch": 0.45674740484429066,
-      "grad_norm": 1.553825855255127,
+      "grad_norm": 1.5488940477371216,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 3.5296,
+      "loss": 3.5365,
       "step": 33
     },
     {
       "epoch": 0.47058823529411764,
-      "grad_norm": 1.6600253582000732,
+      "grad_norm": 1.6649410724639893,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 3.5936,
+      "loss": 3.6033,
       "step": 34
     },
     {
       "epoch": 0.4844290657439446,
-      "grad_norm": 1.5793613195419312,
+      "grad_norm": 1.5827367305755615,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 3.4868,
+      "loss": 3.4936,
       "step": 35
     },
     {
       "epoch": 0.4982698961937716,
-      "grad_norm": 1.5771844387054443,
+      "grad_norm": 1.582950234413147,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 3.3913,
+      "loss": 3.4003,
       "step": 36
     },
     {
       "epoch": 0.5121107266435986,
-      "grad_norm": 1.4913063049316406,
+      "grad_norm": 1.4960726499557495,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 3.3753,
+      "loss": 3.3835,
       "step": 37
     },
     {
       "epoch": 0.5259515570934256,
-      "grad_norm": 1.2248560190200806,
+      "grad_norm": 1.2201118469238281,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 3.5842,
+      "loss": 3.5908,
       "step": 38
     },
     {
       "epoch": 0.5397923875432526,
-      "grad_norm": 1.188950538635254,
+      "grad_norm": 1.18841552734375,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 3.1568,
+      "loss": 3.1628,
       "step": 39
     },
     {
       "epoch": 0.5536332179930796,
-      "grad_norm": 1.0539932250976562,
+      "grad_norm": 1.048747181892395,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 3.1136,
+      "loss": 3.1171,
       "step": 40
     },
     {
       "epoch": 0.5674740484429066,
-      "grad_norm": 0.9032504558563232,
+      "grad_norm": 0.901454508304596,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 3.2446,
+      "loss": 3.2492,
       "step": 41
     },
     {
       "epoch": 0.5813148788927336,
-      "grad_norm": 1.021135926246643,
+      "grad_norm": 1.0152961015701294,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 3.1291,
+      "loss": 3.132,
       "step": 42
     },
     {
       "epoch": 0.5951557093425606,
-      "grad_norm": 0.9351340532302856,
+      "grad_norm": 0.9326971173286438,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 3.0287,
+      "loss": 3.0336,
       "step": 43
     },
     {
       "epoch": 0.6089965397923875,
-      "grad_norm": 0.9577859044075012,
+      "grad_norm": 0.9532666802406311,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 3.1163,
+      "loss": 3.1209,
       "step": 44
     },
     {
       "epoch": 0.6228373702422145,
-      "grad_norm": 1.1382120847702026,
+      "grad_norm": 1.1372768878936768,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 3.2935,
+      "loss": 3.2997,
       "step": 45
     },
     {
       "epoch": 0.6366782006920415,
-      "grad_norm": 1.443152904510498,
+      "grad_norm": 1.4541966915130615,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 3.3713,
+      "loss": 3.3793,
       "step": 46
     },
     {
       "epoch": 0.6505190311418685,
-      "grad_norm": 1.5768526792526245,
+      "grad_norm": 1.5919541120529175,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 3.2518,
+      "loss": 3.261,
       "step": 47
     },
     {
       "epoch": 0.6643598615916955,
-      "grad_norm": 1.6085634231567383,
+      "grad_norm": 1.6075881719589233,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 3.3065,
+      "loss": 3.3155,
       "step": 48
     },
     {
       "epoch": 0.6782006920415224,
-      "grad_norm": 1.483580231666565,
+      "grad_norm": 1.4852365255355835,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 3.089,
+      "loss": 3.0955,
       "step": 49
     },
     {
       "epoch": 0.6920415224913494,
-      "grad_norm": 1.5235376358032227,
+      "grad_norm": 1.521230697631836,
       "learning_rate": 0.0,
-      "loss": 3.5258,
+      "loss": 3.5398,
       "step": 50
     },
     {
       "epoch": 0.6920415224913494,
-      "eval_loss": 3.1869354248046875,
-      "eval_runtime": 6.282,
-      "eval_samples_per_second": 77.523,
-      "eval_steps_per_second": 9.71,
+      "eval_loss": 3.1944468021392822,
+      "eval_runtime": 6.2035,
+      "eval_samples_per_second": 78.504,
+      "eval_steps_per_second": 9.833,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:297365864432a66351de4a28dcab5df4abc8350f5eb8210d38b1e903e733b0b8
+oid sha256:4a76349aa92ef48b300c140adc465b946c6362057acd6daf4a99536f902f6066
 size 6776