Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -23,13 +23,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "k_proj",
-    "q_proj",
     "down_proj",
-    "up_proj",
+    "q_proj",
+    "gate_proj",
     "v_proj",
-    "gate_proj"
+    "o_proj",
+    "k_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:a33fc99f5e08e646d42c0561c56b5821b09250fdd3d15fbcde88b7a475081e12
+oid sha256:df03fd08ab9fdd770afc4be13628ed4a43be6e0fba91329539dedfcc89f6e574
 size 400084608

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:28038fa8b20a5f3e068043e019217f6df5d176864538c272aed89a2656ff32da
+oid sha256:c0417c1fc395346347d0821803c8bab346eaca77ad73b30e9b1a9fdfca09758c
 size 800394282

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95452f226ab8112c5d34713f3d0d2166e6f1079574141c1cfbcd1bfb324ecedb
+oid sha256:9973ae7ee811eca10c6cfe5cb1afa88741500b7e83526e461875e54b5b135838
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5cedf481429a622a9adb82617bc3b10900d833f9f4bf7c67d33be509dd4167a8
+oid sha256:4425e9e3c59645b3de8fe516eef1298cccb7d5b5af65f6072cde0d5b0ef4a6a1
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7583844866a46541e42c9649450c4f6a148085db2b060cfd2b168c9a9503277
+oid sha256:94c2c641558f845ca7cda5cdc282898b1605f0c89e1bdb7b54bfdfc777176377
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab701e676072743a3dd7a6908b21ea8e049e1f2c98b25064c9ac8fa8052a2b00
+oid sha256:1ae48ff9cc95ee1c2516616654792ea74cb1805bb4bf7eca6a07d34c4320d8ff
 size 15024

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.7803696393966675,
+  "best_metric": 0.7827567458152771,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.1572636131315117,
   "eval_steps": 25,
@@ -10,376 +10,376 @@
   "log_history": [
     {
       "epoch": 0.003145272262630234,
-      "grad_norm": 48.137351989746094,
+      "grad_norm": 50.15571594238281,
       "learning_rate": 5e-05,
-      "loss": 23.7645,
+      "loss": 23.649,
       "step": 1
     },
     {
       "epoch": 0.003145272262630234,
-      "eval_loss": 2.990372896194458,
-      "eval_runtime": 70.5527,
-      "eval_samples_per_second": 30.36,
-      "eval_steps_per_second": 3.799,
+      "eval_loss": 2.990436315536499,
+      "eval_runtime": 147.8589,
+      "eval_samples_per_second": 14.487,
+      "eval_steps_per_second": 1.813,
       "step": 1
     },
     {
       "epoch": 0.006290544525260468,
-      "grad_norm": 78.85252380371094,
+      "grad_norm": 84.99065399169922,
       "learning_rate": 0.0001,
-      "loss": 32.7882,
+      "loss": 32.6965,
       "step": 2
     },
     {
       "epoch": 0.009435816787890702,
-      "grad_norm": 93.15648651123047,
+      "grad_norm": 97.3388900756836,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 35.9672,
+      "loss": 36.0529,
       "step": 3
     },
     {
       "epoch": 0.012581089050520937,
-      "grad_norm": 85.15644836425781,
+      "grad_norm": 85.1102066040039,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 31.3601,
+      "loss": 31.1202,
       "step": 4
     },
     {
       "epoch": 0.01572636131315117,
-      "grad_norm": 124.11822509765625,
+      "grad_norm": 128.9368438720703,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 25.2768,
+      "loss": 25.7205,
       "step": 5
     },
     {
       "epoch": 0.018871633575781404,
-      "grad_norm": 131.96466064453125,
+      "grad_norm": 134.2197723388672,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 23.5102,
+      "loss": 23.5406,
       "step": 6
     },
     {
       "epoch": 0.02201690583841164,
-      "grad_norm": 73.48600769042969,
+      "grad_norm": 93.34611511230469,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 19.7645,
+      "loss": 19.9947,
       "step": 7
     },
     {
       "epoch": 0.025162178101041873,
-      "grad_norm": 82.53703308105469,
+      "grad_norm": 87.54219055175781,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 19.3792,
+      "loss": 19.5545,
       "step": 8
     },
     {
       "epoch": 0.028307450363672104,
-      "grad_norm": 45.692649841308594,
+      "grad_norm": 47.01821517944336,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 19.024,
+      "loss": 19.0155,
       "step": 9
     },
     {
       "epoch": 0.03145272262630234,
-      "grad_norm": 55.109004974365234,
+      "grad_norm": 58.72509002685547,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 17.7288,
+      "loss": 18.2442,
       "step": 10
     },
     {
       "epoch": 0.03459799488893257,
-      "grad_norm": 44.44831466674805,
+      "grad_norm": 43.83711242675781,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 19.2609,
+      "loss": 19.0354,
       "step": 11
     },
     {
       "epoch": 0.03774326715156281,
-      "grad_norm": 56.220123291015625,
+      "grad_norm": 44.65055465698242,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 17.8905,
+      "loss": 18.1419,
       "step": 12
     },
     {
       "epoch": 0.04088853941419304,
-      "grad_norm": 57.38528060913086,
+      "grad_norm": 48.8754768371582,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 17.4774,
+      "loss": 17.2647,
       "step": 13
     },
     {
       "epoch": 0.04403381167682328,
-      "grad_norm": 36.43846130371094,
+      "grad_norm": 38.02545928955078,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 12.6264,
+      "loss": 12.8221,
       "step": 14
     },
     {
       "epoch": 0.04717908393945351,
-      "grad_norm": 33.216556549072266,
+      "grad_norm": 34.193153381347656,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 11.5375,
+      "loss": 11.5805,
       "step": 15
     },
     {
       "epoch": 0.050324356202083746,
-      "grad_norm": 35.885189056396484,
+      "grad_norm": 35.46950149536133,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 13.8121,
+      "loss": 13.8534,
       "step": 16
     },
     {
       "epoch": 0.053469628464713974,
-      "grad_norm": 35.649383544921875,
+      "grad_norm": 40.56476974487305,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 11.7226,
+      "loss": 11.8523,
       "step": 17
     },
     {
       "epoch": 0.05661490072734421,
-      "grad_norm": 36.31869888305664,
+      "grad_norm": 39.04943084716797,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 12.9675,
+      "loss": 13.1189,
       "step": 18
     },
     {
       "epoch": 0.05976017298997444,
-      "grad_norm": 39.0863037109375,
+      "grad_norm": 36.440757751464844,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 13.7001,
+      "loss": 13.8588,
       "step": 19
     },
     {
       "epoch": 0.06290544525260468,
-      "grad_norm": 40.13642120361328,
+      "grad_norm": 44.88105010986328,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 11.9583,
+      "loss": 12.2716,
       "step": 20
     },
     {
       "epoch": 0.06605071751523492,
-      "grad_norm": 37.46891784667969,
+      "grad_norm": 30.431800842285156,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 14.8852,
+      "loss": 14.7654,
       "step": 21
     },
     {
       "epoch": 0.06919598977786515,
-      "grad_norm": 40.609046936035156,
+      "grad_norm": 30.384138107299805,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 14.8601,
+      "loss": 14.6801,
       "step": 22
     },
     {
       "epoch": 0.07234126204049537,
-      "grad_norm": 40.82521438598633,
+      "grad_norm": 34.21302795410156,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 16.0274,
+      "loss": 15.708,
       "step": 23
     },
     {
       "epoch": 0.07548653430312562,
-      "grad_norm": 35.38889694213867,
+      "grad_norm": 34.4479866027832,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 14.792,
+      "loss": 14.919,
       "step": 24
     },
     {
       "epoch": 0.07863180656575584,
-      "grad_norm": 31.420101165771484,
+      "grad_norm": 30.595529556274414,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 13.3789,
+      "loss": 13.1339,
       "step": 25
     },
     {
       "epoch": 0.07863180656575584,
-      "eval_loss": 0.822068452835083,
-      "eval_runtime": 115.4818,
-      "eval_samples_per_second": 18.548,
-      "eval_steps_per_second": 2.321,
+      "eval_loss": 0.8246355652809143,
+      "eval_runtime": 148.5117,
+      "eval_samples_per_second": 14.423,
+      "eval_steps_per_second": 1.805,
       "step": 25
     },
     {
       "epoch": 0.08177707882838608,
-      "grad_norm": 27.851903915405273,
+      "grad_norm": 25.552597045898438,
       "learning_rate": 5e-05,
-      "loss": 10.7565,
+      "loss": 10.682,
       "step": 26
     },
     {
       "epoch": 0.08492235109101631,
-      "grad_norm": 33.523040771484375,
+      "grad_norm": 28.349584579467773,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 10.0613,
+      "loss": 10.2223,
       "step": 27
     },
     {
       "epoch": 0.08806762335364655,
-      "grad_norm": 35.41191101074219,
+      "grad_norm": 31.778818130493164,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 11.171,
+      "loss": 11.3269,
       "step": 28
     },
     {
       "epoch": 0.09121289561627678,
-      "grad_norm": 29.849151611328125,
+      "grad_norm": 22.42070198059082,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 10.1654,
+      "loss": 9.7902,
       "step": 29
     },
     {
       "epoch": 0.09435816787890702,
-      "grad_norm": 21.884729385375977,
+      "grad_norm": 19.64124870300293,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 11.1283,
+      "loss": 11.4457,
       "step": 30
     },
     {
       "epoch": 0.09750344014153725,
-      "grad_norm": 33.34331130981445,
+      "grad_norm": 34.70068359375,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 13.2032,
+      "loss": 13.2921,
       "step": 31
     },
     {
       "epoch": 0.10064871240416749,
-      "grad_norm": 31.741554260253906,
+      "grad_norm": 34.29121398925781,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 12.6097,
+      "loss": 13.022,
       "step": 32
     },
     {
       "epoch": 0.10379398466679772,
-      "grad_norm": 39.72273635864258,
+      "grad_norm": 42.92076110839844,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 13.2786,
+      "loss": 13.2991,
       "step": 33
     },
     {
       "epoch": 0.10693925692942795,
-      "grad_norm": 35.299747467041016,
+      "grad_norm": 35.69548416137695,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 13.0968,
+      "loss": 12.9719,
       "step": 34
     },
     {
       "epoch": 0.11008452919205819,
-      "grad_norm": 36.8641242980957,
+      "grad_norm": 37.51350784301758,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 15.7772,
+      "loss": 15.8407,
       "step": 35
     },
     {
       "epoch": 0.11322980145468842,
-      "grad_norm": 33.947181701660156,
+      "grad_norm": 33.40663146972656,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 14.32,
+      "loss": 13.8839,
       "step": 36
     },
     {
       "epoch": 0.11637507371731866,
-      "grad_norm": 36.409114837646484,
+      "grad_norm": 33.837772369384766,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 16.3751,
+      "loss": 16.487,
       "step": 37
     },
     {
       "epoch": 0.11952034597994889,
-      "grad_norm": 18.205045700073242,
+      "grad_norm": 18.629932403564453,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 13.4284,
+      "loss": 13.3913,
       "step": 38
     },
     {
       "epoch": 0.12266561824257913,
-      "grad_norm": 17.913698196411133,
+      "grad_norm": 17.758703231811523,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 9.1217,
+      "loss": 9.1801,
       "step": 39
     },
     {
       "epoch": 0.12581089050520936,
-      "grad_norm": 17.40163230895996,
+      "grad_norm": 17.75163459777832,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 11.089,
+      "loss": 10.9617,
       "step": 40
     },
     {
       "epoch": 0.12895616276783958,
-      "grad_norm": 22.650402069091797,
+      "grad_norm": 27.684919357299805,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 9.2554,
+      "loss": 9.4469,
       "step": 41
     },
     {
       "epoch": 0.13210143503046984,
-      "grad_norm": 24.713539123535156,
+      "grad_norm": 25.891115188598633,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 10.6307,
+      "loss": 10.7454,
       "step": 42
     },
     {
       "epoch": 0.13524670729310007,
-      "grad_norm": 23.468542098999023,
+      "grad_norm": 23.45388412475586,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 12.7203,
+      "loss": 12.6907,
       "step": 43
     },
     {
       "epoch": 0.1383919795557303,
-      "grad_norm": 24.51915168762207,
+      "grad_norm": 21.785078048706055,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 11.6749,
+      "loss": 11.5483,
       "step": 44
     },
     {
       "epoch": 0.14153725181836052,
-      "grad_norm": 22.541257858276367,
+      "grad_norm": 21.240245819091797,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 11.5878,
+      "loss": 11.9439,
       "step": 45
     },
     {
       "epoch": 0.14468252408099075,
-      "grad_norm": 23.1823787689209,
+      "grad_norm": 24.224916458129883,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 13.7743,
+      "loss": 13.9474,
       "step": 46
     },
     {
       "epoch": 0.147827796343621,
-      "grad_norm": 23.573471069335938,
+      "grad_norm": 23.044328689575195,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 13.6634,
+      "loss": 13.5041,
       "step": 47
     },
     {
       "epoch": 0.15097306860625123,
-      "grad_norm": 26.455148696899414,
+      "grad_norm": 24.797870635986328,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 15.6171,
+      "loss": 15.3749,
       "step": 48
     },
     {
       "epoch": 0.15411834086888146,
-      "grad_norm": 27.474468231201172,
+      "grad_norm": 27.531234741210938,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 15.1597,
+      "loss": 15.4553,
       "step": 49
     },
     {
       "epoch": 0.1572636131315117,
-      "grad_norm": 25.294469833374023,
+      "grad_norm": 24.826749801635742,
       "learning_rate": 0.0,
-      "loss": 16.109,
+      "loss": 16.5417,
       "step": 50
     },
     {
       "epoch": 0.1572636131315117,
-      "eval_loss": 0.7803696393966675,
-      "eval_runtime": 111.4341,
-      "eval_samples_per_second": 19.222,
-      "eval_steps_per_second": 2.405,
+      "eval_loss": 0.7827567458152771,
+      "eval_runtime": 148.501,
+      "eval_samples_per_second": 14.424,
+      "eval_steps_per_second": 1.805,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db50dc0bf59cb14564b3dbcd5a6d92202d4676b93f6138c688cdf348df041bd0
+oid sha256:d6a05e61e53a81635f724bab402f3ee6f029810ee111f686410d0b6283774f34
 size 6904