Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -21,12 +21,12 @@
   "revision": null,
   "target_modules": [
     "down_proj",
+    "q_proj",
     "gate_proj",
-    "v_proj",
     "o_proj",
+    "v_proj",
     "k_proj",
-    "up_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:a7215b2f8ded34489a67898c1e1b418f87cc3ca49f5368abf0e618e5a285d3ed
+oid sha256:0eb0d518b0f3e943c2c7e7afa09244667006303e38d58df7effc23a1825d550e
 size 147770496

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb3c426669537bca2b254e98bd192f9b340a6cdec4de124bc8230f7b000a15c4
+oid sha256:3c4eae20a7feba8dace6168b47792ab2c8908a281a64b1b4e9b67db4aebd2dde
 size 295765866

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:07b54192036e87adbcb0a518b1909dfe69188fb4d07e4ad0b470cc764c463932
-size 15024
+oid sha256:e8d88c30a3440d91db013bdd0fad3930d3cf4b7e92ba083dfddff12b6471dee1
+size 14960

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 1.0068762302398682,
+  "best_metric": 1.007637858390808,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.02781157656874674,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.0005562315313749349,
-      "grad_norm": 0.16819290816783905,
+      "grad_norm": 0.17074637115001678,
       "learning_rate": 5e-05,
       "loss": 0.4995,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.0005562315313749349,
       "eval_loss": 1.4900273084640503,
-      "eval_runtime": 138.3344,
-      "eval_samples_per_second": 87.556,
-      "eval_steps_per_second": 10.944,
+      "eval_runtime": 139.2807,
+      "eval_samples_per_second": 86.961,
+      "eval_steps_per_second": 10.87,
       "step": 1
     },
     {
       "epoch": 0.0011124630627498697,
-      "grad_norm": 0.2780616879463196,
+      "grad_norm": 0.2847374379634857,
       "learning_rate": 0.0001,
       "loss": 0.7094,
       "step": 2
     },
     {
       "epoch": 0.0016686945941248046,
-      "grad_norm": 0.2760772109031677,
+      "grad_norm": 0.28109991550445557,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 0.7877,
+      "loss": 0.7874,
       "step": 3
     },
     {
       "epoch": 0.0022249261254997394,
-      "grad_norm": 0.1785762757062912,
+      "grad_norm": 0.17991852760314941,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 0.6781,
+      "loss": 0.6778,
       "step": 4
     },
     {
       "epoch": 0.0027811576568746743,
-      "grad_norm": 0.20672507584095,
+      "grad_norm": 0.21055077016353607,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 0.749,
+      "loss": 0.7491,
       "step": 5
     },
     {
       "epoch": 0.003337389188249609,
-      "grad_norm": 0.2551461458206177,
+      "grad_norm": 0.25590717792510986,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 0.7543,
+      "loss": 0.7538,
       "step": 6
     },
     {
       "epoch": 0.0038936207196245435,
-      "grad_norm": 0.302836149930954,
+      "grad_norm": 0.3053005337715149,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 1.0068,
+      "loss": 1.007,
       "step": 7
     },
     {
       "epoch": 0.004449852250999479,
-      "grad_norm": 0.4083991050720215,
+      "grad_norm": 0.41728004813194275,
       "learning_rate": 9.619397662556435e-05,
       "loss": 1.2281,
       "step": 8
     },
     {
       "epoch": 0.005006083782374413,
-      "grad_norm": 0.6307817101478577,
+      "grad_norm": 0.638646125793457,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 1.3475,
+      "loss": 1.3474,
       "step": 9
     },
     {
       "epoch": 0.0055623153137493485,
-      "grad_norm": 0.8528248071670532,
+      "grad_norm": 0.847240686416626,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 1.3916,
+      "loss": 1.3924,
       "step": 10
     },
     {
       "epoch": 0.006118546845124283,
-      "grad_norm": 0.7784711718559265,
+      "grad_norm": 0.7838229537010193,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 1.5344,
+      "loss": 1.5348,
       "step": 11
     },
     {
       "epoch": 0.006674778376499218,
-      "grad_norm": 1.3868682384490967,
+      "grad_norm": 1.4161337614059448,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 1.7784,
+      "loss": 1.7805,
       "step": 12
     },
     {
       "epoch": 0.007231009907874153,
-      "grad_norm": 0.2851199209690094,
+      "grad_norm": 0.2901410162448883,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 0.3871,
+      "loss": 0.3884,
       "step": 13
     },
     {
       "epoch": 0.007787241439249087,
-      "grad_norm": 0.49446403980255127,
+      "grad_norm": 0.5015016794204712,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 0.6567,
+      "loss": 0.658,
       "step": 14
     },
     {
       "epoch": 0.008343472970624021,
-      "grad_norm": 0.5644774436950684,
+      "grad_norm": 0.5744218826293945,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 0.7766,
+      "loss": 0.7786,
       "step": 15
     },
     {
       "epoch": 0.008899704501998958,
-      "grad_norm": 0.5088273882865906,
+      "grad_norm": 0.5170615315437317,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 0.7204,
+      "loss": 0.7221,
       "step": 16
     },
     {
       "epoch": 0.009455936033373892,
-      "grad_norm": 0.40882161259651184,
+      "grad_norm": 0.41735491156578064,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 0.7042,
+      "loss": 0.7054,
       "step": 17
     },
     {
       "epoch": 0.010012167564748826,
-      "grad_norm": 0.35806888341903687,
+      "grad_norm": 0.3628758490085602,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 0.6965,
+      "loss": 0.6976,
       "step": 18
     },
     {
       "epoch": 0.010568399096123761,
-      "grad_norm": 0.2751835882663727,
+      "grad_norm": 0.2777642011642456,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 0.7627,
+      "loss": 0.7637,
       "step": 19
     },
     {
       "epoch": 0.011124630627498697,
-      "grad_norm": 0.28427332639694214,
+      "grad_norm": 0.28582563996315,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 0.9482,
+      "loss": 0.949,
       "step": 20
     },
     {
       "epoch": 0.011680862158873631,
-      "grad_norm": 0.3408927917480469,
+      "grad_norm": 0.34367185831069946,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 1.1685,
+      "loss": 1.17,
       "step": 21
     },
     {
       "epoch": 0.012237093690248566,
-      "grad_norm": 0.4195518493652344,
+      "grad_norm": 0.41310620307922363,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 1.2933,
+      "loss": 1.2937,
       "step": 22
     },
     {
       "epoch": 0.0127933252216235,
-      "grad_norm": 0.6826456189155579,
+      "grad_norm": 0.6642657518386841,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 1.4564,
+      "loss": 1.4569,
       "step": 23
     },
     {
       "epoch": 0.013349556752998436,
-      "grad_norm": 0.7509661912918091,
+      "grad_norm": 0.7483640909194946,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 1.3699,
+      "loss": 1.3725,
       "step": 24
     },
     {
       "epoch": 0.01390578828437337,
-      "grad_norm": 1.598557949066162,
+      "grad_norm": 1.5988129377365112,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 1.9502,
+      "loss": 1.9567,
       "step": 25
     },
     {
       "epoch": 0.01390578828437337,
-      "eval_loss": 1.0297775268554688,
-      "eval_runtime": 138.3612,
-      "eval_samples_per_second": 87.539,
-      "eval_steps_per_second": 10.942,
+      "eval_loss": 1.0317566394805908,
+      "eval_runtime": 139.3327,
+      "eval_samples_per_second": 86.929,
+      "eval_steps_per_second": 10.866,
       "step": 25
     },
     {
       "epoch": 0.014462019815748305,
-      "grad_norm": 0.09323353320360184,
+      "grad_norm": 0.0968230590224266,
       "learning_rate": 5e-05,
-      "loss": 0.4263,
+      "loss": 0.4264,
       "step": 26
     },
     {
       "epoch": 0.01501825134712324,
-      "grad_norm": 0.13217025995254517,
+      "grad_norm": 0.1362060010433197,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 0.6665,
+      "loss": 0.6668,
       "step": 27
     },
     {
       "epoch": 0.015574482878498174,
-      "grad_norm": 0.14283215999603271,
+      "grad_norm": 0.14593543112277985,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 0.6363,
+      "loss": 0.6366,
       "step": 28
     },
     {
       "epoch": 0.01613071440987311,
-      "grad_norm": 0.1312437802553177,
+      "grad_norm": 0.13117048144340515,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 0.6328,
+      "loss": 0.6331,
       "step": 29
     },
     {
       "epoch": 0.016686945941248043,
-      "grad_norm": 0.14886406064033508,
+      "grad_norm": 0.1489531695842743,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 0.7318,
+      "loss": 0.7321,
       "step": 30
     },
     {
       "epoch": 0.01724317747262298,
-      "grad_norm": 0.16798894107341766,
+      "grad_norm": 0.16366565227508545,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 0.7748,
+      "loss": 0.7745,
       "step": 31
     },
     {
       "epoch": 0.017799409003997915,
-      "grad_norm": 0.21576356887817383,
+      "grad_norm": 0.2094602733850479,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 1.0073,
+      "loss": 1.0068,
       "step": 32
     },
     {
       "epoch": 0.01835564053537285,
-      "grad_norm": 0.28977954387664795,
+      "grad_norm": 0.2810556888580322,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 1.1981,
+      "loss": 1.198,
       "step": 33
     },
     {
       "epoch": 0.018911872066747784,
-      "grad_norm": 0.3524511754512787,
+      "grad_norm": 0.3413487672805786,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 1.2503,
+      "loss": 1.25,
       "step": 34
     },
     {
       "epoch": 0.01946810359812272,
-      "grad_norm": 0.39892593026161194,
+      "grad_norm": 0.3932642340660095,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 1.3395,
+      "loss": 1.3403,
       "step": 35
     },
     {
       "epoch": 0.020024335129497653,
-      "grad_norm": 0.5505003929138184,
+      "grad_norm": 0.5464429259300232,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 1.3837,
+      "loss": 1.3872,
       "step": 36
     },
     {
       "epoch": 0.020580566660872587,
-      "grad_norm": 0.938264012336731,
+      "grad_norm": 0.9510241150856018,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 1.5507,
+      "loss": 1.5583,
       "step": 37
     },
     {
       "epoch": 0.021136798192247522,
-      "grad_norm": 0.10745574533939362,
+      "grad_norm": 0.14339905977249146,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 0.429,
+      "loss": 0.4293,
       "step": 38
     },
     {
       "epoch": 0.021693029723622456,
-      "grad_norm": 0.1093042865395546,
+      "grad_norm": 0.1106351837515831,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 0.5853,
+      "loss": 0.5854,
       "step": 39
     },
     {
       "epoch": 0.022249261254997394,
-      "grad_norm": 0.13401463627815247,
+      "grad_norm": 0.13558265566825867,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 0.7035,
+      "loss": 0.7036,
       "step": 40
     },
     {
       "epoch": 0.02280549278637233,
-      "grad_norm": 0.13401392102241516,
+      "grad_norm": 0.134271502494812,
       "learning_rate": 8.426519384872733e-06,
       "loss": 0.6431,
       "step": 41
     },
     {
       "epoch": 0.023361724317747263,
-      "grad_norm": 0.13436633348464966,
+      "grad_norm": 0.13426171243190765,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 0.6846,
+      "loss": 0.6847,
       "step": 42
     },
     {
       "epoch": 0.023917955849122197,
-      "grad_norm": 0.15161360800266266,
+      "grad_norm": 0.15028472244739532,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 0.8573,
+      "loss": 0.857,
       "step": 43
     },
     {
       "epoch": 0.024474187380497132,
-      "grad_norm": 0.17184071242809296,
+      "grad_norm": 0.17103053629398346,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 0.8592,
+      "loss": 0.8595,
       "step": 44
     },
     {
       "epoch": 0.025030418911872066,
-      "grad_norm": 0.23989973962306976,
+      "grad_norm": 0.2375383824110031,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 1.1216,
+      "loss": 1.1208,
       "step": 45
     },
     {
       "epoch": 0.025586650443247,
-      "grad_norm": 0.26583200693130493,
+      "grad_norm": 0.26565784215927124,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 1.131,
+      "loss": 1.1306,
       "step": 46
     },
     {
       "epoch": 0.026142881974621935,
-      "grad_norm": 0.3194172978401184,
+      "grad_norm": 0.32174623012542725,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 1.2225,
+      "loss": 1.2226,
       "step": 47
     },
     {
       "epoch": 0.026699113505996873,
-      "grad_norm": 0.4354291558265686,
+      "grad_norm": 0.42432740330696106,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 1.3967,
+      "loss": 1.3962,
       "step": 48
     },
     {
       "epoch": 0.027255345037371807,
-      "grad_norm": 0.5957903861999512,
+      "grad_norm": 0.5926147699356079,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 1.3031,
+      "loss": 1.3057,
       "step": 49
     },
     {
       "epoch": 0.02781157656874674,
-      "grad_norm": 1.0612412691116333,
+      "grad_norm": 1.0448579788208008,
       "learning_rate": 0.0,
-      "loss": 1.7977,
+      "loss": 1.7992,
       "step": 50
     },
     {
       "epoch": 0.02781157656874674,
-      "eval_loss": 1.0068762302398682,
-      "eval_runtime": 138.3091,
-      "eval_samples_per_second": 87.572,
-      "eval_steps_per_second": 10.946,
+      "eval_loss": 1.007637858390808,
+      "eval_runtime": 139.376,
+      "eval_samples_per_second": 86.902,
+      "eval_steps_per_second": 10.863,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8dc80fc03fca4d388d2ccc424302d7f624eaf96f11b738febd29270741043805
+oid sha256:d0840b8b7a3a6832313ff3a0b7e27880da3547b309bddebf8df0dcf0c152b9ea
 size 6776