Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "down_proj",
-    "up_proj",
-    "v_proj",
-    "q_proj",
+    "gate_proj",
     "k_proj",
-    "gate_proj"
+    "q_proj",
+    "v_proj",
+    "down_proj",
+    "o_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 sha256:5a4fd1a1204f820541a8f3b2731e5a5ef37999a03a9c7d0a27dd5dd588cafb95
+oid sha256:4fc4f0532a8a124f8f6da731cb0bfb3fa5bc024dae5474dea2861a74dc0c47ec
 size 97728

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f4f48320bbf5c5a6823402dff77050fb6330d7bb7d8c7d072e7d14d57a37ae5
+oid sha256:b68885e2c9409123496c2f74f3ef7b60cd2fdb6598afd471ed84f995a985f3e8
 size 212298

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 10.363082885742188,
+  "best_metric": 10.363097190856934,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.11341083073433512,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.0022682166146867026,
-      "grad_norm": 0.13453303277492523,
+      "grad_norm": 0.14479899406433105,
       "learning_rate": 5e-05,
       "loss": 10.3842,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.0022682166146867026,
       "eval_loss": 10.386616706848145,
-      "eval_runtime": 3.0147,
-      "eval_samples_per_second": 985.168,
-      "eval_steps_per_second": 123.395,
+      "eval_runtime": 3.2554,
+      "eval_samples_per_second": 912.334,
+      "eval_steps_per_second": 114.272,
       "step": 1
     },
     {
       "epoch": 0.004536433229373405,
-      "grad_norm": 0.1353277713060379,
+      "grad_norm": 0.14807619154453278,
       "learning_rate": 0.0001,
       "loss": 10.3876,
       "step": 2
     },
     {
       "epoch": 0.006804649844060108,
-      "grad_norm": 0.14166195690631866,
+      "grad_norm": 0.15308985114097595,
       "learning_rate": 9.989294616193017e-05,
       "loss": 10.3875,
       "step": 3
     },
     {
       "epoch": 0.00907286645874681,
-      "grad_norm": 0.14265072345733643,
+      "grad_norm": 0.1542433798313141,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 10.3854,
+      "loss": 10.3853,
       "step": 4
     },
     {
       "epoch": 0.011341083073433513,
-      "grad_norm": 0.1489330232143402,
+      "grad_norm": 0.161626935005188,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 10.3863,
+      "loss": 10.3862,
       "step": 5
     },
     {
       "epoch": 0.013609299688120215,
-      "grad_norm": 0.14977039396762848,
+      "grad_norm": 0.16123023629188538,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 10.3842,
+      "loss": 10.3841,
       "step": 6
     },
     {
       "epoch": 0.01587751630280692,
-      "grad_norm": 0.15355272591114044,
+      "grad_norm": 0.1623682677745819,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 10.3833,
+      "loss": 10.3832,
       "step": 7
     },
     {
       "epoch": 0.01814573291749362,
-      "grad_norm": 0.1444183886051178,
+      "grad_norm": 0.15166254341602325,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 10.3862,
+      "loss": 10.3861,
       "step": 8
     },
     {
       "epoch": 0.020413949532180325,
-      "grad_norm": 0.15106390416622162,
+      "grad_norm": 0.16110487282276154,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 10.3849,
+      "loss": 10.3847,
       "step": 9
     },
     {
       "epoch": 0.022682166146867026,
-      "grad_norm": 0.154390349984169,
+      "grad_norm": 0.15980345010757446,
       "learning_rate": 9.330127018922194e-05,
       "loss": 10.3838,
       "step": 10
     },
     {
       "epoch": 0.02495038276155373,
-      "grad_norm": 0.14189912378787994,
+      "grad_norm": 0.146297425031662,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 10.3832,
+      "loss": 10.3831,
       "step": 11
     },
     {
       "epoch": 0.02721859937624043,
-      "grad_norm": 0.16048261523246765,
+      "grad_norm": 0.16222330927848816,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 10.3848,
+      "loss": 10.3847,
       "step": 12
     },
     {
       "epoch": 0.029486815990927135,
-      "grad_norm": 0.18116344511508942,
+      "grad_norm": 0.1891157180070877,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 10.3773,
+      "loss": 10.377,
       "step": 13
     },
     {
       "epoch": 0.03175503260561384,
-      "grad_norm": 0.1787891834974289,
+      "grad_norm": 0.19309911131858826,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 10.3796,
+      "loss": 10.3792,
       "step": 14
     },
     {
       "epoch": 0.03402324922030054,
-      "grad_norm": 0.19100786745548248,
+      "grad_norm": 0.202869713306427,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 10.3789,
+      "loss": 10.3785,
       "step": 15
     },
     {
       "epoch": 0.03629146583498724,
-      "grad_norm": 0.19920316338539124,
+      "grad_norm": 0.2117149531841278,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 10.3765,
+      "loss": 10.3761,
       "step": 16
     },
     {
       "epoch": 0.03855968244967394,
-      "grad_norm": 0.19364210963249207,
+      "grad_norm": 0.20285949110984802,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 10.3789,
+      "loss": 10.3787,
       "step": 17
     },
     {
       "epoch": 0.04082789906436065,
-      "grad_norm": 0.1934957504272461,
+      "grad_norm": 0.202358677983284,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 10.3758,
+      "loss": 10.3756,
       "step": 18
     },
     {
       "epoch": 0.04309611567904735,
-      "grad_norm": 0.19975155591964722,
+      "grad_norm": 0.20665286481380463,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 10.3772,
+      "loss": 10.3771,
       "step": 19
     },
     {
       "epoch": 0.04536433229373405,
-      "grad_norm": 0.20424290001392365,
+      "grad_norm": 0.21152782440185547,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 10.3748,
+      "loss": 10.3745,
       "step": 20
     },
     {
       "epoch": 0.04763254890842075,
-      "grad_norm": 0.21936988830566406,
+      "grad_norm": 0.22421160340309143,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 10.3752,
+      "loss": 10.3751,
       "step": 21
     },
     {
       "epoch": 0.04990076552310746,
-      "grad_norm": 0.19495096802711487,
+      "grad_norm": 0.19612239301204681,
       "learning_rate": 6.294095225512603e-05,
       "loss": 10.3742,
       "step": 22
     },
     {
       "epoch": 0.05216898213779416,
-      "grad_norm": 0.1962108314037323,
+      "grad_norm": 0.19813410937786102,
       "learning_rate": 5.9754516100806423e-05,
       "loss": 10.3749,
       "step": 23
     },
     {
       "epoch": 0.05443719875248086,
-      "grad_norm": 0.21276497840881348,
+      "grad_norm": 0.20711098611354828,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 10.3741,
+      "loss": 10.3744,
       "step": 24
     },
     {
       "epoch": 0.05670541536716756,
-      "grad_norm": 0.22711583971977234,
+      "grad_norm": 0.2224423587322235,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 10.3694,
+      "loss": 10.3696,
       "step": 25
     },
     {
       "epoch": 0.05670541536716756,
-      "eval_loss": 10.370012283325195,
-      "eval_runtime": 3.012,
-      "eval_samples_per_second": 986.057,
-      "eval_steps_per_second": 123.506,
+      "eval_loss": 10.369842529296875,
+      "eval_runtime": 3.2494,
+      "eval_samples_per_second": 914.016,
+      "eval_steps_per_second": 114.483,
       "step": 25
     },
     {
       "epoch": 0.05897363198185427,
-      "grad_norm": 0.2256181389093399,
+      "grad_norm": 0.23957474529743195,
       "learning_rate": 5e-05,
-      "loss": 10.3691,
+      "loss": 10.3686,
       "step": 26
     },
     {
       "epoch": 0.06124184859654097,
-      "grad_norm": 0.2604275941848755,
+      "grad_norm": 0.27102959156036377,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 10.3696,
+      "loss": 10.3692,
       "step": 27
     },
     {
       "epoch": 0.06351006521122768,
-      "grad_norm": 0.25977396965026855,
+      "grad_norm": 0.27026283740997314,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 10.3701,
+      "loss": 10.3697,
       "step": 28
     },
     {
       "epoch": 0.06577828182591437,
-      "grad_norm": 0.24867641925811768,
+      "grad_norm": 0.25880345702171326,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 10.3697,
+      "loss": 10.3693,
       "step": 29
     },
     {
       "epoch": 0.06804649844060108,
-      "grad_norm": 0.25991758704185486,
+      "grad_norm": 0.2676348090171814,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 10.3685,
+      "loss": 10.3682,
       "step": 30
     },
     {
       "epoch": 0.07031471505528777,
-      "grad_norm": 0.2578147351741791,
+      "grad_norm": 0.267744779586792,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 10.3677,
+      "loss": 10.3674,
       "step": 31
     },
     {
       "epoch": 0.07258293166997448,
-      "grad_norm": 0.26339924335479736,
+      "grad_norm": 0.26661357283592224,
       "learning_rate": 3.086582838174551e-05,
       "loss": 10.3657,
       "step": 32
     },
     {
       "epoch": 0.07485114828466119,
-      "grad_norm": 0.26088207960128784,
+      "grad_norm": 0.26832523941993713,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 10.3669,
+      "loss": 10.3667,
       "step": 33
     },
     {
       "epoch": 0.07711936489934788,
-      "grad_norm": 0.2547827959060669,
+      "grad_norm": 0.25640127062797546,
       "learning_rate": 2.500000000000001e-05,
       "loss": 10.3682,
       "step": 34
     },
     {
       "epoch": 0.07938758151403459,
-      "grad_norm": 0.24972853064537048,
+      "grad_norm": 0.24771539866924286,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 10.3678,
+      "loss": 10.3681,
       "step": 35
     },
     {
       "epoch": 0.0816557981287213,
-      "grad_norm": 0.24014292657375336,
+      "grad_norm": 0.23294077813625336,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 10.3681,
+      "loss": 10.3686,
       "step": 36
     },
     {
       "epoch": 0.08392401474340799,
-      "grad_norm": 0.25370946526527405,
+      "grad_norm": 0.23941195011138916,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 10.368,
+      "loss": 10.3687,
       "step": 37
     },
     {
       "epoch": 0.0861922313580947,
-      "grad_norm": 0.2755444645881653,
+      "grad_norm": 0.28074121475219727,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 10.365,
+      "loss": 10.3648,
       "step": 38
     },
     {
       "epoch": 0.0884604479727814,
-      "grad_norm": 0.28044813871383667,
+      "grad_norm": 0.28860318660736084,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 10.3633,
+      "loss": 10.363,
       "step": 39
     },
     {
       "epoch": 0.0907286645874681,
-      "grad_norm": 0.2877492308616638,
+      "grad_norm": 0.29766014218330383,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 10.3634,
+      "loss": 10.363,
       "step": 40
     },
     {
       "epoch": 0.09299688120215481,
-      "grad_norm": 0.28548380732536316,
+      "grad_norm": 0.28895074129104614,
       "learning_rate": 8.426519384872733e-06,
       "loss": 10.3642,
       "step": 41
     },
     {
       "epoch": 0.0952650978168415,
-      "grad_norm": 0.287057101726532,
+      "grad_norm": 0.29562824964523315,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 10.3644,
+      "loss": 10.3641,
       "step": 42
     },
     {
       "epoch": 0.09753331443152821,
-      "grad_norm": 0.27017641067504883,
+      "grad_norm": 0.27963921427726746,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 10.3632,
+      "loss": 10.3629,
       "step": 43
     },
     {
       "epoch": 0.09980153104621492,
-      "grad_norm": 0.28016337752342224,
+      "grad_norm": 0.2855812609195709,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 10.3635,
+      "loss": 10.3633,
       "step": 44
     },
     {
       "epoch": 0.10206974766090161,
-      "grad_norm": 0.2789553701877594,
+      "grad_norm": 0.2819939851760864,
       "learning_rate": 2.653493525244721e-06,
       "loss": 10.3652,
       "step": 45
     },
     {
       "epoch": 0.10433796427558832,
-      "grad_norm": 0.26287078857421875,
+      "grad_norm": 0.2683885097503662,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 10.3658,
+      "loss": 10.3656,
       "step": 46
     },
     {
       "epoch": 0.10660618089027502,
-      "grad_norm": 0.2548418939113617,
+      "grad_norm": 0.25791868567466736,
       "learning_rate": 9.607359798384785e-07,
       "loss": 10.3657,
       "step": 47
     },
     {
       "epoch": 0.10887439750496172,
-      "grad_norm": 0.2463953047990799,
+      "grad_norm": 0.24798472225666046,
       "learning_rate": 4.277569313094809e-07,
       "loss": 10.3657,
       "step": 48
     },
     {
       "epoch": 0.11114261411964843,
-      "grad_norm": 0.24887269735336304,
+      "grad_norm": 0.23849883675575256,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 10.3673,
+      "loss": 10.368,
       "step": 49
     },
     {
       "epoch": 0.11341083073433512,
-      "grad_norm": 0.28098398447036743,
+      "grad_norm": 0.26522383093833923,
       "learning_rate": 0.0,
-      "loss": 10.3651,
+      "loss": 10.366,
       "step": 50
     },
     {
       "epoch": 0.11341083073433512,
-      "eval_loss": 10.363082885742188,
-      "eval_runtime": 3.0067,
-      "eval_samples_per_second": 987.799,
-      "eval_steps_per_second": 123.724,
+      "eval_loss": 10.363097190856934,
+      "eval_runtime": 3.246,
+      "eval_samples_per_second": 914.961,
+      "eval_steps_per_second": 114.601,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7438df3680029d1b3ebd8cda75f93f924b03e7e0482eee806cb9cb7640a2209c
+oid sha256:7a177b60e693dc813e8f3df7a2d42c94f1117dfde37b89347c189b46762472d4
 size 6776