Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "gate_proj",
-    "k_proj",
-    "v_proj",
     "o_proj",
+    "k_proj",
+    "down_proj",
+    "gate_proj",
     "up_proj",
-    "down_proj"
+    "v_proj",
+    "q_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 sha256:8277984d9c60132214b91aa106e3c06d9041f5b092d5cdc6114f694d6f2352d9
+oid sha256:23310b84d6a0c378e814a0af56e6001d95fb6d450458fba02ee243ba4bb54619
 size 147770496

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cbb217f94c6258b296ceb60f825a3a061233cfe269049839c1add5b28abba2cd
+oid sha256:21493c18d2a588c96ae6aeefe9b60db121416e2e0da26c7e0fef49660e1a8348
 size 295765866

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 2.7513415813446045,
+  "best_metric": 2.750397205352783,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.20486555697823303,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.00409731113956466,
-      "grad_norm": 0.3377749025821686,
+      "grad_norm": 0.337307870388031,
       "learning_rate": 5e-05,
       "loss": 2.884,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.00409731113956466,
       "eval_loss": 3.2207844257354736,
-      "eval_runtime": 18.6802,
-      "eval_samples_per_second": 88.007,
-      "eval_steps_per_second": 11.028,
+      "eval_runtime": 18.4935,
+      "eval_samples_per_second": 88.896,
+      "eval_steps_per_second": 11.139,
       "step": 1
     },
     {
       "epoch": 0.00819462227912932,
-      "grad_norm": 0.4230843186378479,
+      "grad_norm": 0.42142367362976074,
       "learning_rate": 0.0001,
       "loss": 2.9061,
       "step": 2
     },
     {
       "epoch": 0.012291933418693982,
-      "grad_norm": 0.486320823431015,
+      "grad_norm": 0.4894143044948578,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 2.8535,
+      "loss": 2.853,
       "step": 3
     },
     {
       "epoch": 0.01638924455825864,
-      "grad_norm": 0.5113270878791809,
+      "grad_norm": 0.509801983833313,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 2.97,
+      "loss": 2.9707,
       "step": 4
     },
     {
       "epoch": 0.020486555697823303,
-      "grad_norm": 0.544889509677887,
+      "grad_norm": 0.5440670251846313,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 2.9132,
+      "loss": 2.9133,
       "step": 5
     },
     {
       "epoch": 0.024583866837387964,
-      "grad_norm": 0.5425451397895813,
+      "grad_norm": 0.5436768531799316,
       "learning_rate": 9.829629131445342e-05,
       "loss": 2.8837,
       "step": 6
     },
     {
       "epoch": 0.028681177976952625,
-      "grad_norm": 0.5697307586669922,
+      "grad_norm": 0.5727759003639221,
       "learning_rate": 9.73465064747553e-05,
       "loss": 2.8397,
       "step": 7
     },
     {
       "epoch": 0.03277848911651728,
-      "grad_norm": 0.5749542713165283,
+      "grad_norm": 0.5788682699203491,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 2.8383,
+      "loss": 2.8384,
       "step": 8
     },
     {
       "epoch": 0.03687580025608195,
-      "grad_norm": 0.6267166137695312,
+      "grad_norm": 0.6321156024932861,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 2.8144,
+      "loss": 2.8146,
       "step": 9
     },
     {
       "epoch": 0.040973111395646605,
-      "grad_norm": 0.633858859539032,
+      "grad_norm": 0.6407620310783386,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 2.8866,
+      "loss": 2.8878,
       "step": 10
     },
     {
       "epoch": 0.04507042253521127,
-      "grad_norm": 0.8049176931381226,
+      "grad_norm": 0.7863659262657166,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 2.758,
+      "loss": 2.7582,
       "step": 11
     },
     {
       "epoch": 0.04916773367477593,
-      "grad_norm": 1.914920449256897,
+      "grad_norm": 1.8568462133407593,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 2.5129,
+      "loss": 2.5068,
       "step": 12
     },
     {
       "epoch": 0.053265044814340586,
-      "grad_norm": 0.977289080619812,
+      "grad_norm": 0.9570862054824829,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 2.8631,
+      "loss": 2.8637,
       "step": 13
     },
     {
       "epoch": 0.05736235595390525,
-      "grad_norm": 0.8639317154884338,
+      "grad_norm": 0.8345481753349304,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 2.8636,
+      "loss": 2.8641,
       "step": 14
     },
     {
       "epoch": 0.06145966709346991,
-      "grad_norm": 0.7026758790016174,
+      "grad_norm": 0.6845197081565857,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 2.8303,
+      "loss": 2.8312,
       "step": 15
     },
     {
       "epoch": 0.06555697823303457,
-      "grad_norm": 0.5873925685882568,
+      "grad_norm": 0.5740463137626648,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 2.8004,
+      "loss": 2.7998,
       "step": 16
     },
     {
       "epoch": 0.06965428937259924,
-      "grad_norm": 0.4253089725971222,
+      "grad_norm": 0.4237423241138458,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 2.7869,
+      "loss": 2.7878,
       "step": 17
     },
     {
       "epoch": 0.0737516005121639,
-      "grad_norm": 0.41267770528793335,
+      "grad_norm": 0.41858452558517456,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 2.745,
+      "loss": 2.7449,
       "step": 18
     },
     {
       "epoch": 0.07784891165172855,
-      "grad_norm": 0.45589208602905273,
+      "grad_norm": 0.44931760430336,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 2.7552,
+      "loss": 2.755,
       "step": 19
     },
     {
       "epoch": 0.08194622279129321,
-      "grad_norm": 0.5008207559585571,
+      "grad_norm": 0.49227553606033325,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 2.74,
+      "loss": 2.7384,
       "step": 20
     },
     {
       "epoch": 0.08604353393085787,
-      "grad_norm": 0.5874332189559937,
+      "grad_norm": 0.5697259902954102,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 2.7908,
+      "loss": 2.7895,
       "step": 21
     },
     {
       "epoch": 0.09014084507042254,
-      "grad_norm": 0.6370252966880798,
+      "grad_norm": 0.620069682598114,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 2.8465,
+      "loss": 2.8453,
       "step": 22
     },
     {
       "epoch": 0.0942381562099872,
-      "grad_norm": 0.7455514073371887,
+      "grad_norm": 0.7299432754516602,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 2.8079,
+      "loss": 2.8072,
       "step": 23
     },
     {
       "epoch": 0.09833546734955186,
-      "grad_norm": 1.0238953828811646,
+      "grad_norm": 1.0050387382507324,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 2.6308,
+      "loss": 2.6318,
       "step": 24
     },
     {
       "epoch": 0.10243277848911651,
-      "grad_norm": 3.58799409866333,
+      "grad_norm": 3.606916904449463,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 2.636,
+      "loss": 2.637,
       "step": 25
     },
     {
       "epoch": 0.10243277848911651,
-      "eval_loss": 2.812803030014038,
-      "eval_runtime": 18.713,
-      "eval_samples_per_second": 87.853,
-      "eval_steps_per_second": 11.008,
+      "eval_loss": 2.8160855770111084,
+      "eval_runtime": 18.5031,
+      "eval_samples_per_second": 88.85,
+      "eval_steps_per_second": 11.133,
       "step": 25
     },
     {
       "epoch": 0.10653008962868117,
-      "grad_norm": 0.2966318428516388,
+      "grad_norm": 0.29132208228111267,
       "learning_rate": 5e-05,
-      "loss": 2.7046,
+      "loss": 2.7031,
       "step": 26
     },
     {
       "epoch": 0.11062740076824584,
-      "grad_norm": 0.3549116849899292,
+      "grad_norm": 0.34843340516090393,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 2.715,
+      "loss": 2.7149,
       "step": 27
     },
     {
       "epoch": 0.1147247119078105,
-      "grad_norm": 0.4115239679813385,
+      "grad_norm": 0.4048047363758087,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 2.8055,
+      "loss": 2.8052,
       "step": 28
     },
     {
       "epoch": 0.11882202304737516,
-      "grad_norm": 0.4180857539176941,
+      "grad_norm": 0.40718603134155273,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 2.7473,
+      "loss": 2.7462,
       "step": 29
     },
     {
       "epoch": 0.12291933418693982,
-      "grad_norm": 0.4537926912307739,
+      "grad_norm": 0.4452973008155823,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 2.7823,
+      "loss": 2.7822,
       "step": 30
     },
     {
       "epoch": 0.12701664532650447,
-      "grad_norm": 0.41609248518943787,
+      "grad_norm": 0.4196675717830658,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 2.7333,
+      "loss": 2.7337,
       "step": 31
     },
     {
       "epoch": 0.13111395646606913,
-      "grad_norm": 0.47336915135383606,
+      "grad_norm": 0.4773973226547241,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 2.6901,
+      "loss": 2.6906,
       "step": 32
     },
     {
       "epoch": 0.1352112676056338,
-      "grad_norm": 0.4530975818634033,
+      "grad_norm": 0.4580395817756653,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 2.8127,
+      "loss": 2.8136,
       "step": 33
     },
     {
       "epoch": 0.13930857874519847,
-      "grad_norm": 0.554924488067627,
+      "grad_norm": 0.5649320483207703,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 2.7751,
+      "loss": 2.7753,
       "step": 34
     },
     {
       "epoch": 0.14340588988476313,
-      "grad_norm": 0.5669254064559937,
+      "grad_norm": 0.5697263479232788,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 2.7738,
+      "loss": 2.7733,
       "step": 35
     },
     {
       "epoch": 0.1475032010243278,
-      "grad_norm": 0.6636052131652832,
+      "grad_norm": 0.6570910811424255,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 2.6836,
+      "loss": 2.6842,
       "step": 36
     },
     {
       "epoch": 0.15160051216389245,
-      "grad_norm": 1.1210123300552368,
+      "grad_norm": 1.1145232915878296,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 2.3253,
+      "loss": 2.326,
       "step": 37
     },
     {
       "epoch": 0.1556978233034571,
-      "grad_norm": 0.4871041178703308,
+      "grad_norm": 0.5004509091377258,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 2.7302,
+      "loss": 2.7298,
       "step": 38
     },
     {
       "epoch": 0.15979513444302176,
-      "grad_norm": 0.32615357637405396,
+      "grad_norm": 0.32440951466560364,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 2.7958,
+      "loss": 2.7961,
       "step": 39
     },
     {
       "epoch": 0.16389244558258642,
-      "grad_norm": 0.36045870184898376,
+      "grad_norm": 0.35644829273223877,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 2.7092,
+      "loss": 2.7085,
       "step": 40
     },
     {
       "epoch": 0.16798975672215108,
-      "grad_norm": 0.35322922468185425,
+      "grad_norm": 0.35068151354789734,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 2.7801,
+      "loss": 2.7808,
       "step": 41
     },
     {
       "epoch": 0.17208706786171574,
-      "grad_norm": 0.3751983344554901,
+      "grad_norm": 0.37324798107147217,
       "learning_rate": 6.698729810778065e-06,
       "loss": 2.8038,
       "step": 42
     },
     {
       "epoch": 0.17618437900128042,
-      "grad_norm": 0.34934067726135254,
+      "grad_norm": 0.3482402265071869,
       "learning_rate": 5.156362923365588e-06,
       "loss": 2.708,
       "step": 43
     },
     {
       "epoch": 0.18028169014084508,
-      "grad_norm": 0.3803287744522095,
+      "grad_norm": 0.3827877640724182,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 2.7296,
+      "loss": 2.7287,
       "step": 44
     },
     {
       "epoch": 0.18437900128040974,
-      "grad_norm": 0.3893742263317108,
+      "grad_norm": 0.39035090804100037,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 2.7378,
+      "loss": 2.7382,
       "step": 45
     },
     {
       "epoch": 0.1884763124199744,
-      "grad_norm": 0.4432777166366577,
+      "grad_norm": 0.44703298807144165,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 2.7923,
+      "loss": 2.7922,
       "step": 46
     },
     {
       "epoch": 0.19257362355953905,
-      "grad_norm": 0.4957706034183502,
+      "grad_norm": 0.5005406737327576,
       "learning_rate": 9.607359798384785e-07,
       "loss": 2.697,
       "step": 47
     },
     {
       "epoch": 0.1966709346991037,
-      "grad_norm": 0.6110988259315491,
+      "grad_norm": 0.6187803149223328,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 2.6986,
+      "loss": 2.698,
       "step": 48
     },
     {
       "epoch": 0.20076824583866837,
-      "grad_norm": 0.8057876229286194,
+      "grad_norm": 0.7934252023696899,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 2.4571,
+      "loss": 2.4563,
       "step": 49
     },
     {
       "epoch": 0.20486555697823303,
-      "grad_norm": 2.9657976627349854,
+      "grad_norm": 2.924260139465332,
       "learning_rate": 0.0,
-      "loss": 2.5226,
+      "loss": 2.5138,
       "step": 50
     },
     {
       "epoch": 0.20486555697823303,
-      "eval_loss": 2.7513415813446045,
-      "eval_runtime": 18.6828,
-      "eval_samples_per_second": 87.996,
-      "eval_steps_per_second": 11.026,
+      "eval_loss": 2.750397205352783,
+      "eval_runtime": 18.5008,
+      "eval_samples_per_second": 88.861,
+      "eval_steps_per_second": 11.135,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c6def4580f714d6ad7e4a090722504acfd79476939235ffabc027470341e135
+oid sha256:9f83fc5afa9c2c0b02c33e188cde50d7502ddda90bc13a39239ce88b24a8e36d
 size 6776