Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -21,12 +21,12 @@
   "revision": null,
   "target_modules": [
     "o_proj",
-    "v_proj",
     "down_proj",
-    "k_proj",
-    "up_proj",
     "gate_proj",
-    "q_proj"
+    "k_proj",
+    "v_proj",
+    "q_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:a22f9375dd7f4f0ada7b41a4e0c78592fc31ad9f3bacce41636af8643d7395a5
+oid sha256:a704533d698c7ba0a63169225e51fe598844be04e0004f38d5f187079038e55d
 size 147770496

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:785bb480d8e758e7ca847ed3cf975951a5215aa290ce424fb810109f53a378f5
+oid sha256:3adb6464f9e197a258140da815c1d67020babe4d3d20818c1917c77a42d9070c
 size 295765866

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.4023851752281189,
+  "best_metric": 0.40252843499183655,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 2.435820895522388,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.04776119402985075,
-      "grad_norm": 0.29309335350990295,
+      "grad_norm": 0.29185420274734497,
       "learning_rate": 5e-05,
       "loss": 0.551,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.04776119402985075,
       "eval_loss": 0.5565004348754883,
-      "eval_runtime": 1.6321,
-      "eval_samples_per_second": 86.392,
-      "eval_steps_per_second": 11.029,
+      "eval_runtime": 1.6581,
+      "eval_samples_per_second": 85.038,
+      "eval_steps_per_second": 10.856,
       "step": 1
     },
     {
       "epoch": 0.0955223880597015,
-      "grad_norm": 0.34882017970085144,
+      "grad_norm": 0.34828293323516846,
       "learning_rate": 0.0001,
       "loss": 0.5825,
       "step": 2
     },
     {
       "epoch": 0.14328358208955225,
-      "grad_norm": 0.33288702368736267,
+      "grad_norm": 0.33185914158821106,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 0.5791,
+      "loss": 0.5789,
       "step": 3
     },
     {
       "epoch": 0.191044776119403,
-      "grad_norm": 0.36883237957954407,
+      "grad_norm": 0.3632985055446625,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 0.538,
+      "loss": 0.5382,
       "step": 4
     },
     {
       "epoch": 0.23880597014925373,
-      "grad_norm": 0.3539741337299347,
+      "grad_norm": 0.3530183434486389,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 0.5281,
+      "loss": 0.5279,
       "step": 5
     },
     {
       "epoch": 0.2865671641791045,
-      "grad_norm": 0.23581592738628387,
+      "grad_norm": 0.23330174386501312,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 0.5197,
+      "loss": 0.5199,
       "step": 6
     },
     {
       "epoch": 0.33432835820895523,
-      "grad_norm": 0.28109219670295715,
+      "grad_norm": 0.278849333524704,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 0.5066,
+      "loss": 0.5065,
       "step": 7
     },
     {
       "epoch": 0.382089552238806,
-      "grad_norm": 0.29403433203697205,
+      "grad_norm": 0.2901425361633301,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 0.4929,
+      "loss": 0.493,
       "step": 8
     },
     {
       "epoch": 0.4298507462686567,
-      "grad_norm": 0.2939996123313904,
+      "grad_norm": 0.29416143894195557,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 0.5313,
+      "loss": 0.5316,
       "step": 9
     },
     {
       "epoch": 0.47761194029850745,
-      "grad_norm": 0.363861083984375,
+      "grad_norm": 0.3660103976726532,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 0.4514,
+      "loss": 0.4523,
       "step": 10
     },
     {
       "epoch": 0.5253731343283582,
-      "grad_norm": 0.19598151743412018,
+      "grad_norm": 0.19246715307235718,
       "learning_rate": 9.157348061512727e-05,
       "loss": 0.4927,
       "step": 11
     },
     {
       "epoch": 0.573134328358209,
-      "grad_norm": 0.20769856870174408,
+      "grad_norm": 0.20531879365444183,
       "learning_rate": 8.966766701456177e-05,
       "loss": 0.5076,
       "step": 12
     },
     {
       "epoch": 0.6208955223880597,
-      "grad_norm": 0.2053491175174713,
+      "grad_norm": 0.20391716063022614,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 0.4653,
+      "loss": 0.4657,
       "step": 13
     },
     {
       "epoch": 0.6686567164179105,
-      "grad_norm": 0.22446739673614502,
+      "grad_norm": 0.22268980741500854,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 0.4791,
+      "loss": 0.4792,
       "step": 14
     },
     {
       "epoch": 0.7164179104477612,
-      "grad_norm": 0.2886030972003937,
+      "grad_norm": 0.2864244878292084,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 0.4275,
+      "loss": 0.4274,
       "step": 15
     },
     {
       "epoch": 0.764179104477612,
-      "grad_norm": 0.1693258434534073,
+      "grad_norm": 0.16762878000736237,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 0.4891,
+      "loss": 0.4894,
       "step": 16
     },
     {
       "epoch": 0.8119402985074626,
-      "grad_norm": 0.18894262611865997,
+      "grad_norm": 0.18803684413433075,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 0.4471,
+      "loss": 0.4475,
       "step": 17
     },
     {
       "epoch": 0.8597014925373134,
-      "grad_norm": 0.2001345455646515,
+      "grad_norm": 0.19906236231327057,
       "learning_rate": 7.500000000000001e-05,
       "loss": 0.4717,
       "step": 18
     },
     {
       "epoch": 0.9074626865671642,
-      "grad_norm": 0.25183868408203125,
+      "grad_norm": 0.24964551627635956,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 0.458,
+      "loss": 0.4579,
       "step": 19
     },
     {
       "epoch": 0.9552238805970149,
-      "grad_norm": 0.29334449768066406,
+      "grad_norm": 0.29042649269104004,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 0.3965,
+      "loss": 0.3962,
       "step": 20
     },
     {
       "epoch": 1.026865671641791,
-      "grad_norm": 0.28625965118408203,
+      "grad_norm": 0.28468796610832214,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 0.796,
+      "loss": 0.7955,
       "step": 21
     },
     {
       "epoch": 1.0746268656716418,
-      "grad_norm": 0.16455918550491333,
+      "grad_norm": 0.16371072828769684,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 0.4404,
+      "loss": 0.4403,
       "step": 22
     },
     {
       "epoch": 1.1223880597014926,
-      "grad_norm": 0.1942681223154068,
+      "grad_norm": 0.19278578460216522,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 0.4026,
+      "loss": 0.4031,
       "step": 23
     },
     {
       "epoch": 1.1701492537313434,
-      "grad_norm": 0.21410231292247772,
+      "grad_norm": 0.21322554349899292,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 0.413,
+      "loss": 0.4135,
       "step": 24
     },
     {
       "epoch": 1.217910447761194,
-      "grad_norm": 0.25955915451049805,
+      "grad_norm": 0.2575645446777344,
       "learning_rate": 5.327015646150716e-05,
       "loss": 0.4206,
       "step": 25
     },
     {
       "epoch": 1.217910447761194,
-      "eval_loss": 0.4251263737678528,
-      "eval_runtime": 1.6332,
-      "eval_samples_per_second": 86.336,
-      "eval_steps_per_second": 11.022,
+      "eval_loss": 0.42539140582084656,
+      "eval_runtime": 1.6618,
+      "eval_samples_per_second": 84.848,
+      "eval_steps_per_second": 10.832,
       "step": 25
     },
     {
       "epoch": 1.2656716417910447,
-      "grad_norm": 0.18197745084762573,
+      "grad_norm": 0.18160618841648102,
       "learning_rate": 5e-05,
-      "loss": 0.378,
+      "loss": 0.3783,
       "step": 26
     },
     {
       "epoch": 1.3134328358208955,
-      "grad_norm": 0.1775834709405899,
+      "grad_norm": 0.17580193281173706,
       "learning_rate": 4.6729843538492847e-05,
       "loss": 0.4182,
       "step": 27
     },
     {
       "epoch": 1.3611940298507463,
-      "grad_norm": 0.19821526110172272,
+      "grad_norm": 0.19652894139289856,
       "learning_rate": 4.347369038899744e-05,
       "loss": 0.4157,
       "step": 28
     },
     {
       "epoch": 1.408955223880597,
-      "grad_norm": 0.21566687524318695,
+      "grad_norm": 0.21493984758853912,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 0.3821,
+      "loss": 0.3818,
       "step": 29
     },
     {
       "epoch": 1.4567164179104477,
-      "grad_norm": 0.2709551453590393,
+      "grad_norm": 0.2695300579071045,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 0.3742,
+      "loss": 0.3741,
       "step": 30
     },
     {
       "epoch": 1.5044776119402985,
-      "grad_norm": 0.1894427090883255,
+      "grad_norm": 0.18835151195526123,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 0.372,
+      "loss": 0.3718,
       "step": 31
     },
     {
       "epoch": 1.5522388059701493,
-      "grad_norm": 0.19189490377902985,
+      "grad_norm": 0.19078905880451202,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 0.4192,
+      "loss": 0.4194,
       "step": 32
     },
     {
       "epoch": 1.6,
-      "grad_norm": 0.22904697060585022,
+      "grad_norm": 0.22648519277572632,
       "learning_rate": 2.7885565489049946e-05,
       "loss": 0.4083,
       "step": 33
     },
     {
       "epoch": 1.6477611940298509,
-      "grad_norm": 0.24419617652893066,
+      "grad_norm": 0.2402975708246231,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 0.4,
+      "loss": 0.3999,
       "step": 34
     },
     {
       "epoch": 1.6955223880597015,
-      "grad_norm": 0.250863641500473,
+      "grad_norm": 0.2498691976070404,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 0.3735,
+      "loss": 0.3738,
       "step": 35
     },
     {
       "epoch": 1.7432835820895523,
-      "grad_norm": 0.22108954191207886,
+      "grad_norm": 0.22093889117240906,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 0.3983,
+      "loss": 0.3985,
       "step": 36
     },
     {
       "epoch": 1.7910447761194028,
-      "grad_norm": 0.1873919814825058,
+      "grad_norm": 0.18763570487499237,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 0.3949,
+      "loss": 0.3952,
       "step": 37
     },
     {
       "epoch": 1.8388059701492536,
-      "grad_norm": 0.2162359058856964,
+      "grad_norm": 0.21417222917079926,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 0.4166,
+      "loss": 0.4162,
       "step": 38
     },
     {
       "epoch": 1.8865671641791044,
-      "grad_norm": 0.2337595373392105,
+      "grad_norm": 0.2319900542497635,
       "learning_rate": 1.2408009626051137e-05,
       "loss": 0.377,
       "step": 39
     },
     {
       "epoch": 1.9343283582089552,
-      "grad_norm": 0.2757619619369507,
+      "grad_norm": 0.2740539014339447,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 0.3678,
+      "loss": 0.3683,
       "step": 40
     },
     {
       "epoch": 2.005970149253731,
-      "grad_norm": 0.30736440420150757,
+      "grad_norm": 0.3065386414527893,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 0.5346,
+      "loss": 0.534,
       "step": 41
     },
     {
       "epoch": 2.053731343283582,
-      "grad_norm": 0.17773054540157318,
+      "grad_norm": 0.17674775421619415,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 0.3897,
+      "loss": 0.3898,
       "step": 42
     },
     {
       "epoch": 2.1014925373134328,
-      "grad_norm": 0.18972079455852509,
+      "grad_norm": 0.18936415016651154,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 0.3795,
+      "loss": 0.3792,
       "step": 43
     },
     {
       "epoch": 2.1492537313432836,
-      "grad_norm": 0.21727779507637024,
+      "grad_norm": 0.21581114828586578,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 0.3802,
+      "loss": 0.3798,
       "step": 44
     },
     {
       "epoch": 2.1970149253731344,
-      "grad_norm": 0.25702959299087524,
+      "grad_norm": 0.25616875290870667,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 0.3841,
+      "loss": 0.3838,
       "step": 45
     },
     {
       "epoch": 2.244776119402985,
-      "grad_norm": 0.26603198051452637,
+      "grad_norm": 0.2650687098503113,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 0.3291,
+      "loss": 0.3289,
       "step": 46
     },
     {
       "epoch": 2.292537313432836,
-      "grad_norm": 0.1831669956445694,
+      "grad_norm": 0.18314772844314575,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 0.4069,
+      "loss": 0.4072,
       "step": 47
     },
     {
       "epoch": 2.3402985074626868,
-      "grad_norm": 0.20122045278549194,
+      "grad_norm": 0.19991053640842438,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 0.3851,
+      "loss": 0.3849,
       "step": 48
     },
     {
       "epoch": 2.388059701492537,
-      "grad_norm": 0.2129780352115631,
+      "grad_norm": 0.21161042153835297,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 0.3787,
+      "loss": 0.3785,
       "step": 49
     },
     {
       "epoch": 2.435820895522388,
-      "grad_norm": 0.2397255003452301,
+      "grad_norm": 0.23710204660892487,
       "learning_rate": 0.0,
-      "loss": 0.3508,
+      "loss": 0.3513,
       "step": 50
     },
     {
       "epoch": 2.435820895522388,
-      "eval_loss": 0.4023851752281189,
-      "eval_runtime": 1.632,
-      "eval_samples_per_second": 86.399,
-      "eval_steps_per_second": 11.03,
+      "eval_loss": 0.40252843499183655,
+      "eval_runtime": 1.6626,
+      "eval_samples_per_second": 84.805,
+      "eval_steps_per_second": 10.826,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c3dea4593a7a820a778fc84ba6e04d4471ebc30e6df1994d208d89710cfaf65d
+oid sha256:e94c1e826225abe88b190761b2645c6636cea2d7bee422d5c89d0033daed0be6
 size 6776