Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -21,12 +21,12 @@
   "revision": null,
   "target_modules": [
     "v_proj",
-    "q_proj",
-    "gate_proj",
-    "k_proj",
     "down_proj",
+    "up_proj",
+    "q_proj",
     "o_proj",
-    "up_proj"
+    "gate_proj",
+    "k_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 sha256:77580874a6cea995a47b3d202895889d9bccee5ec7fe1953fcdade01fedc1c4b
+oid sha256:bafa4cca50aec18233c0193fd3f0088c21986310601de761008a486806d0d9af
 size 503407240

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be688fe6e47744c7892449100e45315aa22bfb0afb28dfd076113c6300e8c501
+oid sha256:c20a9def8a1ffc76af75f59b81e726230c4c8bc8dd6257cd31a14cd3f0df8e35
 size 1007201074

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 1.6343824863433838,
+  "best_metric": 1.636163353919983,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.03311806590495115,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.000662361318099023,
-      "grad_norm": 41.63633728027344,
+      "grad_norm": 39.855995178222656,
       "learning_rate": 5e-05,
       "loss": 27.5824,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.000662361318099023,
       "eval_loss": 2.2696444988250732,
-      "eval_runtime": 431.7537,
-      "eval_samples_per_second": 23.557,
-      "eval_steps_per_second": 2.946,
+      "eval_runtime": 434.2712,
+      "eval_samples_per_second": 23.421,
+      "eval_steps_per_second": 2.929,
       "step": 1
     },
     {
       "epoch": 0.001324722636198046,
-      "grad_norm": 36.75858688354492,
+      "grad_norm": 35.34762191772461,
       "learning_rate": 0.0001,
       "loss": 27.0301,
       "step": 2
     },
     {
       "epoch": 0.001987083954297069,
-      "grad_norm": 42.25046920776367,
+      "grad_norm": 40.128257751464844,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 30.0,
+      "loss": 30.0122,
       "step": 3
     },
     {
       "epoch": 0.002649445272396092,
-      "grad_norm": 36.78057861328125,
+      "grad_norm": 35.508766174316406,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 27.1166,
+      "loss": 27.1283,
       "step": 4
     },
     {
       "epoch": 0.003311806590495115,
-      "grad_norm": 36.5877799987793,
+      "grad_norm": 35.003509521484375,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 30.1777,
+      "loss": 30.1937,
       "step": 5
     },
     {
       "epoch": 0.003974167908594138,
-      "grad_norm": 34.792659759521484,
+      "grad_norm": 33.496620178222656,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 32.1896,
+      "loss": 32.1736,
       "step": 6
     },
     {
       "epoch": 0.004636529226693161,
-      "grad_norm": 28.476728439331055,
+      "grad_norm": 28.025121688842773,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 28.1354,
+      "loss": 28.1705,
       "step": 7
     },
     {
       "epoch": 0.005298890544792184,
-      "grad_norm": 39.693870544433594,
+      "grad_norm": 38.546974182128906,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 31.488,
+      "loss": 31.4793,
       "step": 8
     },
     {
       "epoch": 0.005961251862891207,
-      "grad_norm": 32.581214904785156,
+      "grad_norm": 31.470550537109375,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 29.9008,
+      "loss": 29.978,
       "step": 9
     },
     {
       "epoch": 0.00662361318099023,
-      "grad_norm": 34.23081588745117,
+      "grad_norm": 33.33460235595703,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 29.6987,
+      "loss": 29.724,
       "step": 10
     },
     {
       "epoch": 0.007285974499089253,
-      "grad_norm": 39.92532730102539,
+      "grad_norm": 38.43280792236328,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 33.3046,
+      "loss": 33.3028,
       "step": 11
     },
     {
       "epoch": 0.007948335817188276,
-      "grad_norm": 37.282161712646484,
+      "grad_norm": 35.78132247924805,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 31.1424,
+      "loss": 31.1296,
       "step": 12
     },
     {
       "epoch": 0.0086106971352873,
-      "grad_norm": 36.434452056884766,
+      "grad_norm": 36.88593673706055,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 30.9173,
+      "loss": 30.9662,
       "step": 13
     },
     {
       "epoch": 0.009273058453386322,
-      "grad_norm": 34.44896697998047,
+      "grad_norm": 35.441650390625,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 26.2724,
+      "loss": 26.3844,
       "step": 14
     },
     {
       "epoch": 0.009935419771485345,
-      "grad_norm": 25.074344635009766,
+      "grad_norm": 25.30802345275879,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 26.8365,
+      "loss": 26.8914,
       "step": 15
     },
     {
       "epoch": 0.010597781089584368,
-      "grad_norm": 20.83970069885254,
+      "grad_norm": 20.81322479248047,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 24.8747,
+      "loss": 24.8694,
       "step": 16
     },
     {
       "epoch": 0.011260142407683391,
-      "grad_norm": 20.634695053100586,
+      "grad_norm": 20.53407859802246,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 24.7149,
+      "loss": 24.7313,
       "step": 17
     },
     {
       "epoch": 0.011922503725782414,
-      "grad_norm": 22.36671257019043,
+      "grad_norm": 21.87583351135254,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 25.9939,
+      "loss": 26.0212,
       "step": 18
     },
     {
       "epoch": 0.012584865043881437,
-      "grad_norm": 23.41545867919922,
+      "grad_norm": 22.73112678527832,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 25.4919,
+      "loss": 25.5372,
       "step": 19
     },
     {
       "epoch": 0.01324722636198046,
-      "grad_norm": 23.04401206970215,
+      "grad_norm": 22.921876907348633,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 26.2824,
+      "loss": 26.2909,
       "step": 20
     },
     {
       "epoch": 0.013909587680079483,
-      "grad_norm": 24.668642044067383,
+      "grad_norm": 24.30012321472168,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 27.4812,
+      "loss": 27.5489,
       "step": 21
     },
     {
       "epoch": 0.014571948998178506,
-      "grad_norm": 24.785856246948242,
+      "grad_norm": 24.02613639831543,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 27.3597,
+      "loss": 27.3856,
       "step": 22
     },
     {
       "epoch": 0.01523431031627753,
-      "grad_norm": 28.567190170288086,
+      "grad_norm": 28.35286521911621,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 27.7361,
+      "loss": 27.7234,
       "step": 23
     },
     {
       "epoch": 0.015896671634376552,
-      "grad_norm": 36.09950256347656,
+      "grad_norm": 33.62429428100586,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 29.4177,
+      "loss": 29.4387,
       "step": 24
     },
     {
       "epoch": 0.016559032952475575,
-      "grad_norm": 32.96276092529297,
+      "grad_norm": 34.30913162231445,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 29.8551,
+      "loss": 29.8534,
       "step": 25
     },
     {
       "epoch": 0.016559032952475575,
-      "eval_loss": 1.6892707347869873,
-      "eval_runtime": 435.802,
-      "eval_samples_per_second": 23.339,
-      "eval_steps_per_second": 2.919,
+      "eval_loss": 1.6895121335983276,
+      "eval_runtime": 437.1193,
+      "eval_samples_per_second": 23.268,
+      "eval_steps_per_second": 2.91,
       "step": 25
     },
     {
       "epoch": 0.0172213942705746,
-      "grad_norm": 22.903234481811523,
+      "grad_norm": 21.834144592285156,
       "learning_rate": 5e-05,
-      "loss": 22.4765,
+      "loss": 22.4394,
       "step": 26
     },
     {
       "epoch": 0.01788375558867362,
-      "grad_norm": 19.913135528564453,
+      "grad_norm": 19.145488739013672,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 23.0996,
+      "loss": 23.0711,
       "step": 27
     },
     {
       "epoch": 0.018546116906772644,
-      "grad_norm": 18.672170639038086,
+      "grad_norm": 18.362302780151367,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 23.7646,
+      "loss": 23.7712,
       "step": 28
     },
     {
       "epoch": 0.019208478224871667,
-      "grad_norm": 25.744522094726562,
+      "grad_norm": 26.06848907470703,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 27.9772,
+      "loss": 27.9575,
       "step": 29
     },
     {
       "epoch": 0.01987083954297069,
-      "grad_norm": 21.418855667114258,
+      "grad_norm": 20.97596549987793,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 25.7398,
+      "loss": 25.7123,
       "step": 30
     },
     {
       "epoch": 0.020533200861069714,
-      "grad_norm": 19.990333557128906,
+      "grad_norm": 19.997718811035156,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 24.2821,
+      "loss": 24.3034,
       "step": 31
     },
     {
       "epoch": 0.021195562179168737,
-      "grad_norm": 22.049428939819336,
+      "grad_norm": 22.800750732421875,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 25.765,
+      "loss": 25.7513,
       "step": 32
     },
     {
       "epoch": 0.02185792349726776,
-      "grad_norm": 22.674144744873047,
+      "grad_norm": 22.728124618530273,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 26.8537,
+      "loss": 26.8797,
       "step": 33
     },
     {
       "epoch": 0.022520284815366783,
-      "grad_norm": 22.346158981323242,
+      "grad_norm": 22.23642349243164,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 26.8923,
+      "loss": 26.9258,
       "step": 34
     },
     {
       "epoch": 0.023182646133465806,
-      "grad_norm": 22.35358428955078,
+      "grad_norm": 22.032678604125977,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 25.6358,
+      "loss": 25.6688,
       "step": 35
     },
     {
       "epoch": 0.02384500745156483,
-      "grad_norm": 26.604528427124023,
+      "grad_norm": 26.3785400390625,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 28.5165,
+      "loss": 28.5324,
       "step": 36
     },
     {
       "epoch": 0.02450736876966385,
-      "grad_norm": 25.8643741607666,
+      "grad_norm": 26.36612892150879,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 31.2324,
+      "loss": 31.2677,
       "step": 37
     },
     {
       "epoch": 0.025169730087762875,
-      "grad_norm": 30.428800582885742,
+      "grad_norm": 30.574783325195312,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 27.0481,
+      "loss": 27.0322,
       "step": 38
     },
     {
       "epoch": 0.025832091405861898,
-      "grad_norm": 18.89072608947754,
+      "grad_norm": 18.087467193603516,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 23.8018,
+      "loss": 23.8036,
       "step": 39
     },
     {
       "epoch": 0.02649445272396092,
-      "grad_norm": 24.802570343017578,
+      "grad_norm": 23.013254165649414,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 26.2798,
+      "loss": 26.3198,
       "step": 40
     },
     {
       "epoch": 0.027156814042059944,
-      "grad_norm": 17.563922882080078,
+      "grad_norm": 16.78318977355957,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 22.905,
+      "loss": 22.9086,
       "step": 41
     },
     {
       "epoch": 0.027819175360158967,
-      "grad_norm": 17.73468780517578,
+      "grad_norm": 17.115629196166992,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 25.4188,
+      "loss": 25.436,
       "step": 42
     },
     {
       "epoch": 0.02848153667825799,
-      "grad_norm": 18.167245864868164,
+      "grad_norm": 17.63496208190918,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 25.1034,
+      "loss": 25.0821,
       "step": 43
     },
     {
       "epoch": 0.029143897996357013,
-      "grad_norm": 22.048166275024414,
+      "grad_norm": 21.715238571166992,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 28.0479,
+      "loss": 28.0717,
       "step": 44
     },
     {
       "epoch": 0.029806259314456036,
-      "grad_norm": 17.95345115661621,
+      "grad_norm": 17.863527297973633,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 26.0678,
+      "loss": 26.0547,
       "step": 45
     },
     {
       "epoch": 0.03046862063255506,
-      "grad_norm": 21.010440826416016,
+      "grad_norm": 20.718631744384766,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 25.4916,
+      "loss": 25.5254,
       "step": 46
     },
     {
       "epoch": 0.031130981950654082,
-      "grad_norm": 21.977310180664062,
+      "grad_norm": 21.940893173217773,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 25.0578,
+      "loss": 25.1054,
       "step": 47
     },
     {
       "epoch": 0.031793343268753105,
-      "grad_norm": 23.666597366333008,
+      "grad_norm": 23.446805953979492,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 26.1003,
+      "loss": 26.1938,
       "step": 48
     },
     {
       "epoch": 0.03245570458685213,
-      "grad_norm": 23.877300262451172,
+      "grad_norm": 23.616628646850586,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 27.8739,
+      "loss": 27.9034,
       "step": 49
     },
     {
       "epoch": 0.03311806590495115,
-      "grad_norm": 28.445064544677734,
+      "grad_norm": 28.035184860229492,
       "learning_rate": 0.0,
-      "loss": 28.7613,
+      "loss": 28.7963,
       "step": 50
     },
     {
       "epoch": 0.03311806590495115,
-      "eval_loss": 1.6343824863433838,
-      "eval_runtime": 431.8551,
-      "eval_samples_per_second": 23.552,
-      "eval_steps_per_second": 2.945,
+      "eval_loss": 1.636163353919983,
+      "eval_runtime": 435.9234,
+      "eval_samples_per_second": 23.332,
+      "eval_steps_per_second": 2.918,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab0c519457372ad6a80ee0dfc254474dd57398cb585e6950ec0029bb6e3d976d
+oid sha256:e4ec30c2f040ab99c8c9780866cf93b033c1f7cafeb2492f21c48e1b70ba5d96
 size 6776