Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "o_proj",
     "v_proj",
-    "down_proj",
-    "up_proj",
+    "gate_proj",
     "k_proj",
-    "gate_proj"
+    "up_proj",
+    "down_proj",
+    "q_proj",
+    "o_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 sha256:267f48741f563a59b32da005a28611785fc7d08be02f9cd34083012162535288
+oid sha256:e8ef91d1d9630f9c8ac56a9dd8353d6179e0d98ec4168cba2208b741e4b4cb91
 size 166182480

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a050eb9ed0adbf1ff9fe94d478308128b144ccfadadcb6ef1365fde78bfc354
+oid sha256:0038a122bf140815cf96697d30cc8d7fdca7dd55cb8c5eb98b3faa51e18ae4b4
 size 332574358

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 1.1902629137039185,
+  "best_metric": 1.19004487991333,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 1.2139605462822458,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.024279210925644917,
-      "grad_norm": 1.7722647190093994,
+      "grad_norm": 1.818414330482483,
       "learning_rate": 5e-05,
       "loss": 1.6302,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.024279210925644917,
       "eval_loss": 1.7623697519302368,
-      "eval_runtime": 13.4424,
-      "eval_samples_per_second": 20.681,
-      "eval_steps_per_second": 2.604,
+      "eval_runtime": 13.5986,
+      "eval_samples_per_second": 20.443,
+      "eval_steps_per_second": 2.574,
       "step": 1
     },
     {
       "epoch": 0.048558421851289835,
-      "grad_norm": 1.8735610246658325,
+      "grad_norm": 1.9210597276687622,
       "learning_rate": 0.0001,
       "loss": 1.7639,
       "step": 2
     },
     {
       "epoch": 0.07283763277693475,
-      "grad_norm": 1.242928147315979,
+      "grad_norm": 1.3875622749328613,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 1.6338,
+      "loss": 1.631,
       "step": 3
     },
     {
       "epoch": 0.09711684370257967,
-      "grad_norm": 0.7185656428337097,
+      "grad_norm": 0.7885871529579163,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 1.4464,
+      "loss": 1.4496,
       "step": 4
     },
     {
       "epoch": 0.12139605462822459,
-      "grad_norm": 0.8275676965713501,
+      "grad_norm": 0.8537760376930237,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 1.409,
+      "loss": 1.4107,
       "step": 5
     },
     {
       "epoch": 0.1456752655538695,
-      "grad_norm": 0.7798041105270386,
+      "grad_norm": 0.7437020540237427,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 1.3467,
+      "loss": 1.3459,
       "step": 6
     },
     {
       "epoch": 0.16995447647951442,
-      "grad_norm": 0.7611581683158875,
+      "grad_norm": 0.5616689324378967,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 1.3044,
+      "loss": 1.3039,
       "step": 7
     },
     {
       "epoch": 0.19423368740515934,
-      "grad_norm": 0.5567551255226135,
+      "grad_norm": 0.6682710647583008,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 1.4015,
+      "loss": 1.4006,
       "step": 8
     },
     {
       "epoch": 0.21851289833080426,
-      "grad_norm": 0.580046534538269,
+      "grad_norm": 0.561467707157135,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 1.4482,
+      "loss": 1.4473,
       "step": 9
     },
     {
       "epoch": 0.24279210925644917,
-      "grad_norm": 0.6574867367744446,
+      "grad_norm": 0.6908743381500244,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 1.4606,
+      "loss": 1.4588,
       "step": 10
     },
     {
       "epoch": 0.26707132018209406,
-      "grad_norm": 0.7613857984542847,
+      "grad_norm": 0.7324297428131104,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 1.3532,
+      "loss": 1.3524,
       "step": 11
     },
     {
       "epoch": 0.291350531107739,
-      "grad_norm": 0.5821555852890015,
+      "grad_norm": 0.8180338144302368,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 1.3331,
+      "loss": 1.3334,
       "step": 12
     },
     {
       "epoch": 0.3156297420333839,
-      "grad_norm": 0.4268592894077301,
+      "grad_norm": 0.43275681138038635,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 1.2977,
+      "loss": 1.298,
       "step": 13
     },
     {
       "epoch": 0.33990895295902884,
-      "grad_norm": 0.4200943112373352,
+      "grad_norm": 0.4039974808692932,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 1.2504,
+      "loss": 1.2517,
       "step": 14
     },
     {
       "epoch": 0.36418816388467373,
-      "grad_norm": 0.42564937472343445,
+      "grad_norm": 0.4262292981147766,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 1.2022,
+      "loss": 1.2033,
       "step": 15
     },
     {
       "epoch": 0.3884673748103187,
-      "grad_norm": 0.4572436213493347,
+      "grad_norm": 0.4397771656513214,
       "learning_rate": 8.043807145043604e-05,
       "loss": 1.2242,
       "step": 16
     },
     {
       "epoch": 0.41274658573596357,
-      "grad_norm": 0.49105581641197205,
+      "grad_norm": 0.44248443841934204,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 1.2034,
+      "loss": 1.2025,
       "step": 17
     },
     {
       "epoch": 0.4370257966616085,
-      "grad_norm": 0.4518444836139679,
+      "grad_norm": 0.41254177689552307,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 1.2124,
+      "loss": 1.2112,
       "step": 18
     },
     {
       "epoch": 0.4613050075872534,
-      "grad_norm": 0.3952445983886719,
+      "grad_norm": 0.3902616798877716,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 1.2795,
+      "loss": 1.2797,
       "step": 19
     },
     {
       "epoch": 0.48558421851289835,
-      "grad_norm": 0.4598488509654999,
+      "grad_norm": 0.48718735575675964,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 1.3569,
+      "loss": 1.3553,
       "step": 20
     },
     {
       "epoch": 0.5098634294385432,
-      "grad_norm": 0.4049881398677826,
+      "grad_norm": 0.4111344516277313,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 1.2496,
+      "loss": 1.2506,
       "step": 21
     },
     {
       "epoch": 0.5341426403641881,
-      "grad_norm": 0.4265231788158417,
+      "grad_norm": 0.43272262811660767,
       "learning_rate": 6.294095225512603e-05,
       "loss": 1.3095,
       "step": 22
     },
     {
       "epoch": 0.5584218512898331,
-      "grad_norm": 0.4079309105873108,
+      "grad_norm": 0.4041851758956909,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 1.2289,
+      "loss": 1.2299,
       "step": 23
     },
     {
       "epoch": 0.582701062215478,
-      "grad_norm": 0.33031976222991943,
+      "grad_norm": 0.33193275332450867,
       "learning_rate": 5.6526309611002594e-05,
       "loss": 1.2246,
       "step": 24
     },
     {
       "epoch": 0.6069802731411229,
-      "grad_norm": 0.32245883345603943,
+      "grad_norm": 0.3251614570617676,
       "learning_rate": 5.327015646150716e-05,
       "loss": 1.1406,
       "step": 25
     },
     {
       "epoch": 0.6069802731411229,
-      "eval_loss": 1.2204139232635498,
-      "eval_runtime": 13.4561,
-      "eval_samples_per_second": 20.66,
-      "eval_steps_per_second": 2.601,
+      "eval_loss": 1.2199825048446655,
+      "eval_runtime": 13.5868,
+      "eval_samples_per_second": 20.461,
+      "eval_steps_per_second": 2.576,
       "step": 25
     },
     {
       "epoch": 0.6312594840667678,
-      "grad_norm": 0.29840490221977234,
+      "grad_norm": 0.2942574918270111,
       "learning_rate": 5e-05,
       "loss": 1.1446,
       "step": 26
     },
     {
       "epoch": 0.6555386949924128,
-      "grad_norm": 0.27768757939338684,
+      "grad_norm": 0.2842651903629303,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 1.178,
+      "loss": 1.1783,
       "step": 27
     },
     {
       "epoch": 0.6798179059180577,
-      "grad_norm": 0.2893085777759552,
+      "grad_norm": 0.3190499246120453,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 1.185,
+      "loss": 1.1848,
       "step": 28
     },
     {
       "epoch": 0.7040971168437026,
-      "grad_norm": 0.31861791014671326,
+      "grad_norm": 0.3258350193500519,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 1.2352,
+      "loss": 1.2343,
       "step": 29
     },
     {
       "epoch": 0.7283763277693475,
-      "grad_norm": 0.41295695304870605,
+      "grad_norm": 0.4318709373474121,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 1.3293,
+      "loss": 1.3297,
       "step": 30
     },
     {
       "epoch": 0.7526555386949925,
-      "grad_norm": 0.3978126049041748,
+      "grad_norm": 0.40284281969070435,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 1.2489,
+      "loss": 1.2475,
       "step": 31
     },
     {
       "epoch": 0.7769347496206374,
-      "grad_norm": 0.32944414019584656,
+      "grad_norm": 0.33570441603660583,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 1.2182,
+      "loss": 1.218,
       "step": 32
     },
     {
       "epoch": 0.8012139605462822,
-      "grad_norm": 0.26036539673805237,
+      "grad_norm": 0.26420605182647705,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 1.2381,
+      "loss": 1.2383,
       "step": 33
     },
     {
       "epoch": 0.8254931714719271,
-      "grad_norm": 0.24936436116695404,
+      "grad_norm": 0.24241675436496735,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 1.1885,
+      "loss": 1.1895,
       "step": 34
     },
     {
       "epoch": 0.849772382397572,
-      "grad_norm": 0.2344878613948822,
+      "grad_norm": 0.2417987883090973,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 1.1693,
+      "loss": 1.17,
       "step": 35
     },
     {
       "epoch": 0.874051593323217,
-      "grad_norm": 0.2727144956588745,
+      "grad_norm": 0.2558254599571228,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 1.0871,
+      "loss": 1.088,
       "step": 36
     },
     {
       "epoch": 0.8983308042488619,
-      "grad_norm": 0.25772327184677124,
+      "grad_norm": 0.2571571171283722,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 1.1025,
+      "loss": 1.1032,
       "step": 37
     },
     {
       "epoch": 0.9226100151745068,
-      "grad_norm": 0.27336952090263367,
+      "grad_norm": 0.27578985691070557,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 1.1575,
+      "loss": 1.1577,
       "step": 38
     },
     {
       "epoch": 0.9468892261001517,
-      "grad_norm": 0.295190691947937,
+      "grad_norm": 0.27690500020980835,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 1.2164,
+      "loss": 1.2171,
       "step": 39
     },
     {
       "epoch": 0.9711684370257967,
-      "grad_norm": 0.36160922050476074,
+      "grad_norm": 0.3624616861343384,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 1.2886,
+      "loss": 1.2889,
       "step": 40
     },
     {
       "epoch": 0.9954476479514416,
-      "grad_norm": 0.518794059753418,
+      "grad_norm": 0.5124205350875854,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 1.3845,
+      "loss": 1.3847,
       "step": 41
     },
     {
       "epoch": 1.0197268588770865,
-      "grad_norm": 0.5481667518615723,
+      "grad_norm": 0.5638722777366638,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 2.0889,
+      "loss": 2.0886,
       "step": 42
     },
     {
       "epoch": 1.0440060698027314,
-      "grad_norm": 0.24297133088111877,
+      "grad_norm": 0.24691079556941986,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 1.2121,
+      "loss": 1.212,
       "step": 43
     },
     {
       "epoch": 1.0682852807283763,
-      "grad_norm": 0.3044127821922302,
+      "grad_norm": 0.24191239476203918,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 1.1445,
+      "loss": 1.1454,
       "step": 44
     },
     {
       "epoch": 1.0925644916540211,
-      "grad_norm": 0.22780603170394897,
+      "grad_norm": 0.23386125266551971,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 1.1201,
+      "loss": 1.121,
       "step": 45
     },
     {
       "epoch": 1.1168437025796663,
-      "grad_norm": 0.24752256274223328,
+      "grad_norm": 0.23766495287418365,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 1.1275,
+      "loss": 1.1281,
       "step": 46
     },
     {
       "epoch": 1.1411229135053111,
-      "grad_norm": 0.2455936223268509,
+      "grad_norm": 0.24574440717697144,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 1.1098,
+      "loss": 1.1109,
       "step": 47
     },
     {
       "epoch": 1.165402124430956,
-      "grad_norm": 0.25134918093681335,
+      "grad_norm": 0.2480248361825943,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 1.0926,
+      "loss": 1.0927,
       "step": 48
     },
     {
       "epoch": 1.189681335356601,
-      "grad_norm": 0.2651086449623108,
+      "grad_norm": 0.2597378194332123,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 1.1768,
+      "loss": 1.1772,
       "step": 49
     },
     {
       "epoch": 1.2139605462822458,
-      "grad_norm": 0.2977955639362335,
+      "grad_norm": 0.28662922978401184,
       "learning_rate": 0.0,
       "loss": 1.2105,
       "step": 50
     },
     {
       "epoch": 1.2139605462822458,
-      "eval_loss": 1.1902629137039185,
-      "eval_runtime": 13.4618,
-      "eval_samples_per_second": 20.651,
-      "eval_steps_per_second": 2.6,
+      "eval_loss": 1.19004487991333,
+      "eval_runtime": 13.5738,
+      "eval_samples_per_second": 20.481,
+      "eval_steps_per_second": 2.578,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7f0262936f29db7120a92e75b37b15c657a864512415a0d04c326c07b5bf68a
+oid sha256:f0682ef37f1c43ecbd1dc3b77fe3639e4b844f66014f2d9ceeb47c4d5bd9e5e2
 size 6776