Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "k_proj",
+    "q_proj",
+    "o_proj",
+    "up_proj",
     "down_proj",
-    "v_proj",
     "gate_proj",
-    "up_proj",
-    "q_proj",
-    "k_proj",
-    "o_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:e17665ce5a09cfa29e3d3dcb84f65d1d7dacfd5b1884c75a226827ecf58dc0b8
+oid sha256:cb9c433e67d7d67735ab4c08b384b700ad7bbb2ef098bc9db8b5f55f05f73f4a
 size 239536272

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5111106d6eced45466e27e3af0a74820e49a678ebe149dda5234146fabd83f72
+oid sha256:8eba68e7a1fa59cac36df6b8ba0efca02f9c853b0a7a8f33c89e21efd55e65ad
 size 479362682

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.6561319231987,
+  "best_metric": 0.6542271971702576,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.823045267489712,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.01646090534979424,
-      "grad_norm": 2.0745296478271484,
+      "grad_norm": 2.1076626777648926,
       "learning_rate": 5e-05,
       "loss": 2.8513,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.01646090534979424,
       "eval_loss": 3.8586878776550293,
-      "eval_runtime": 9.15,
-      "eval_samples_per_second": 44.809,
-      "eval_steps_per_second": 5.683,
+      "eval_runtime": 9.2637,
+      "eval_samples_per_second": 44.259,
+      "eval_steps_per_second": 5.613,
       "step": 1
     },
     {
       "epoch": 0.03292181069958848,
-      "grad_norm": 4.245978832244873,
+      "grad_norm": 4.27289342880249,
       "learning_rate": 0.0001,
       "loss": 3.7586,
       "step": 2
     },
     {
       "epoch": 0.04938271604938271,
-      "grad_norm": 3.185753345489502,
+      "grad_norm": 3.2222888469696045,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 2.3779,
+      "loss": 2.3785,
       "step": 3
     },
     {
       "epoch": 0.06584362139917696,
-      "grad_norm": 1.9672001600265503,
+      "grad_norm": 1.957718014717102,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 2.0297,
+      "loss": 2.0282,
       "step": 4
     },
     {
       "epoch": 0.0823045267489712,
-      "grad_norm": 1.712628960609436,
+      "grad_norm": 1.7186223268508911,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 1.8397,
+      "loss": 1.8344,
       "step": 5
     },
     {
       "epoch": 0.09876543209876543,
-      "grad_norm": 1.813711166381836,
+      "grad_norm": 1.804044485092163,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 1.8286,
+      "loss": 1.8234,
       "step": 6
     },
     {
       "epoch": 0.11522633744855967,
-      "grad_norm": 1.2593228816986084,
+      "grad_norm": 1.2378959655761719,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 1.4568,
+      "loss": 1.4513,
       "step": 7
     },
     {
       "epoch": 0.13168724279835392,
-      "grad_norm": 1.2200363874435425,
+      "grad_norm": 1.2065438032150269,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 1.3131,
+      "loss": 1.3096,
       "step": 8
     },
     {
       "epoch": 0.14814814814814814,
-      "grad_norm": 1.1022920608520508,
+      "grad_norm": 1.109233021736145,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 1.2503,
+      "loss": 1.2455,
       "step": 9
     },
     {
       "epoch": 0.1646090534979424,
-      "grad_norm": 1.273776650428772,
+      "grad_norm": 1.291716456413269,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 1.1055,
+      "loss": 1.0978,
       "step": 10
     },
     {
       "epoch": 0.18106995884773663,
-      "grad_norm": 1.3083699941635132,
+      "grad_norm": 1.3139619827270508,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 1.0066,
+      "loss": 0.9933,
       "step": 11
     },
     {
       "epoch": 0.19753086419753085,
-      "grad_norm": 1.1210218667984009,
+      "grad_norm": 1.085813283920288,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 0.8673,
+      "loss": 0.8552,
       "step": 12
     },
     {
       "epoch": 0.2139917695473251,
-      "grad_norm": 0.7338252663612366,
+      "grad_norm": 0.7361060380935669,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 0.9367,
+      "loss": 0.9337,
       "step": 13
     },
     {
       "epoch": 0.23045267489711935,
-      "grad_norm": 2.1798887252807617,
+      "grad_norm": 2.214524984359741,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 1.6209,
+      "loss": 1.6196,
       "step": 14
     },
     {
       "epoch": 0.24691358024691357,
-      "grad_norm": 1.767669677734375,
+      "grad_norm": 1.789655327796936,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 0.7957,
+      "loss": 0.7971,
       "step": 15
     },
     {
       "epoch": 0.26337448559670784,
-      "grad_norm": 1.1363743543624878,
+      "grad_norm": 1.104480504989624,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 0.7195,
+      "loss": 0.7159,
       "step": 16
     },
     {
       "epoch": 0.27983539094650206,
-      "grad_norm": 0.9281293749809265,
+      "grad_norm": 0.9207172989845276,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 0.7021,
+      "loss": 0.7,
       "step": 17
     },
     {
       "epoch": 0.2962962962962963,
-      "grad_norm": 0.8513988852500916,
+      "grad_norm": 0.8482317328453064,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 0.6473,
+      "loss": 0.6445,
       "step": 18
     },
     {
       "epoch": 0.31275720164609055,
-      "grad_norm": 0.8034453988075256,
+      "grad_norm": 0.8116461634635925,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 0.6974,
+      "loss": 0.6966,
       "step": 19
     },
     {
       "epoch": 0.3292181069958848,
-      "grad_norm": 0.7160072922706604,
+      "grad_norm": 0.7192047238349915,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 0.5821,
+      "loss": 0.5805,
       "step": 20
     },
     {
       "epoch": 0.345679012345679,
-      "grad_norm": 0.5699191093444824,
+      "grad_norm": 0.5720171332359314,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 0.7074,
+      "loss": 0.706,
       "step": 21
     },
     {
       "epoch": 0.36213991769547327,
-      "grad_norm": 0.5850489735603333,
+      "grad_norm": 0.5808382630348206,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 0.7256,
+      "loss": 0.7216,
       "step": 22
     },
     {
       "epoch": 0.3786008230452675,
-      "grad_norm": 0.5858528017997742,
+      "grad_norm": 0.5952896475791931,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 0.6395,
+      "loss": 0.6375,
       "step": 23
     },
     {
       "epoch": 0.3950617283950617,
-      "grad_norm": 0.5442656874656677,
+      "grad_norm": 0.5605278015136719,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 0.7342,
+      "loss": 0.7344,
       "step": 24
     },
     {
       "epoch": 0.411522633744856,
-      "grad_norm": 0.49498945474624634,
+      "grad_norm": 0.49808284640312195,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 0.5981,
+      "loss": 0.5986,
       "step": 25
     },
     {
       "epoch": 0.411522633744856,
-      "eval_loss": 0.7050751447677612,
-      "eval_runtime": 9.1683,
-      "eval_samples_per_second": 44.719,
-      "eval_steps_per_second": 5.672,
+      "eval_loss": 0.7032530307769775,
+      "eval_runtime": 9.2958,
+      "eval_samples_per_second": 44.106,
+      "eval_steps_per_second": 5.594,
       "step": 25
     },
     {
       "epoch": 0.4279835390946502,
-      "grad_norm": 1.625022053718567,
+      "grad_norm": 1.5510308742523193,
       "learning_rate": 5e-05,
-      "loss": 1.3808,
+      "loss": 1.3782,
       "step": 26
     },
     {
       "epoch": 0.4444444444444444,
-      "grad_norm": 0.6726037263870239,
+      "grad_norm": 0.7058679461479187,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 0.6507,
+      "loss": 0.6491,
       "step": 27
     },
     {
       "epoch": 0.4609053497942387,
-      "grad_norm": 0.6659382581710815,
+      "grad_norm": 0.6497540473937988,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 0.5273,
+      "loss": 0.524,
       "step": 28
     },
     {
       "epoch": 0.4773662551440329,
-      "grad_norm": 0.698225736618042,
+      "grad_norm": 0.6953015923500061,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 0.5542,
+      "loss": 0.552,
       "step": 29
     },
     {
       "epoch": 0.49382716049382713,
-      "grad_norm": 0.6371154189109802,
+      "grad_norm": 0.6362207531929016,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 0.6459,
+      "loss": 0.645,
       "step": 30
     },
     {
       "epoch": 0.5102880658436214,
-      "grad_norm": 0.5590847730636597,
+      "grad_norm": 0.5652956962585449,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 0.685,
+      "loss": 0.6847,
       "step": 31
     },
     {
       "epoch": 0.5267489711934157,
-      "grad_norm": 0.4336808919906616,
+      "grad_norm": 0.43394219875335693,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 0.5531,
+      "loss": 0.5517,
       "step": 32
     },
     {
       "epoch": 0.5432098765432098,
-      "grad_norm": 0.5399011969566345,
+      "grad_norm": 0.5500791668891907,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 0.6532,
+      "loss": 0.6542,
       "step": 33
     },
     {
       "epoch": 0.5596707818930041,
-      "grad_norm": 0.4386877417564392,
+      "grad_norm": 0.43191617727279663,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 0.6143,
+      "loss": 0.6129,
       "step": 34
     },
     {
       "epoch": 0.5761316872427984,
-      "grad_norm": 0.5017386078834534,
+      "grad_norm": 0.5142382383346558,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 0.6102,
+      "loss": 0.6085,
       "step": 35
     },
     {
       "epoch": 0.5925925925925926,
-      "grad_norm": 0.6162391304969788,
+      "grad_norm": 0.6265186667442322,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 0.6991,
+      "loss": 0.6996,
       "step": 36
     },
     {
       "epoch": 0.6090534979423868,
-      "grad_norm": 0.6711835265159607,
+      "grad_norm": 0.679259717464447,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 0.693,
+      "loss": 0.6922,
       "step": 37
     },
     {
       "epoch": 0.6255144032921811,
-      "grad_norm": 0.5756227970123291,
+      "grad_norm": 0.5726998448371887,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 0.8572,
+      "loss": 0.8567,
       "step": 38
     },
     {
       "epoch": 0.6419753086419753,
-      "grad_norm": 2.781473398208618,
+      "grad_norm": 2.6537625789642334,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 1.7294,
+      "loss": 1.7269,
       "step": 39
     },
     {
       "epoch": 0.6584362139917695,
-      "grad_norm": 0.6352210640907288,
+      "grad_norm": 0.6081885099411011,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 0.5651,
+      "loss": 0.5644,
       "step": 40
     },
     {
       "epoch": 0.6748971193415638,
-      "grad_norm": 0.47287803888320923,
+      "grad_norm": 0.4832661747932434,
       "learning_rate": 8.426519384872733e-06,
       "loss": 0.5217,
       "step": 41
     },
     {
       "epoch": 0.691358024691358,
-      "grad_norm": 0.5060291886329651,
+      "grad_norm": 0.49536004662513733,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 0.5924,
+      "loss": 0.5929,
       "step": 42
     },
     {
       "epoch": 0.7078189300411523,
-      "grad_norm": 0.434349000453949,
+      "grad_norm": 0.4383414089679718,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 0.529,
+      "loss": 0.5276,
       "step": 43
     },
     {
       "epoch": 0.7242798353909465,
-      "grad_norm": 0.5044702291488647,
+      "grad_norm": 0.5090070962905884,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 0.5424,
+      "loss": 0.5399,
       "step": 44
     },
     {
       "epoch": 0.7407407407407407,
-      "grad_norm": 0.42993804812431335,
+      "grad_norm": 0.4303985834121704,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 0.6676,
+      "loss": 0.6663,
       "step": 45
     },
     {
       "epoch": 0.757201646090535,
-      "grad_norm": 0.4085606634616852,
+      "grad_norm": 0.41048020124435425,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 0.5848,
+      "loss": 0.5851,
       "step": 46
     },
     {
       "epoch": 0.7736625514403292,
-      "grad_norm": 0.4246390163898468,
+      "grad_norm": 0.4269702732563019,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 0.6776,
+      "loss": 0.6774,
       "step": 47
     },
     {
       "epoch": 0.7901234567901234,
-      "grad_norm": 0.35061928629875183,
+      "grad_norm": 0.3528209924697876,
       "learning_rate": 4.277569313094809e-07,
       "loss": 0.5528,
       "step": 48
     },
     {
       "epoch": 0.8065843621399177,
-      "grad_norm": 0.4323487877845764,
+      "grad_norm": 0.4349346458911896,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 0.6973,
+      "loss": 0.6985,
       "step": 49
     },
     {
       "epoch": 0.823045267489712,
-      "grad_norm": 0.4282306730747223,
+      "grad_norm": 0.40909990668296814,
       "learning_rate": 0.0,
-      "loss": 0.5867,
+      "loss": 0.5858,
       "step": 50
     },
     {
       "epoch": 0.823045267489712,
-      "eval_loss": 0.6561319231987,
-      "eval_runtime": 9.1655,
-      "eval_samples_per_second": 44.733,
-      "eval_steps_per_second": 5.673,
+      "eval_loss": 0.6542271971702576,
+      "eval_runtime": 9.2891,
+      "eval_samples_per_second": 44.138,
+      "eval_steps_per_second": 5.598,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:85e8dfa1540ca3fe1a9279429500b8935e685568e77962b57855c979b3e8ce15
+oid sha256:94f8966a7f5610ce3154bf3af6bc87a6aa3dca284fbee83e288364d669bb4725
 size 6776