Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
+    "q_proj",
     "gate_proj",
-    "down_proj",
-    "k_proj",
-    "v_proj",
     "up_proj",
-    "q_proj"
+    "k_proj",
+    "down_proj",
+    "o_proj",
+    "v_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:ef65185383fe2309629b81577e55d84a21d7b4d3bfd83a6a44f3a2b50c35da1c
+oid sha256:7d30c1b5464902c493f3d8655d9da49d534dcd5539dbb0cc9c940301fd3a4897
 size 4525640

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31ecebd15430909226b95eecc8f3a1c0149603cc6916560885fdec84947b6f8f
+oid sha256:d1d2d4081e2a2b6d016b9f6c30469b47e2cfa36731273cb05fa0e5e296f3718e
 size 9067978

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.7643183469772339,
+  "best_metric": 0.7737964987754822,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.9070294784580499,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.018140589569160998,
-      "grad_norm": 5.590025424957275,
+      "grad_norm": 5.461395263671875,
       "learning_rate": 5e-05,
       "loss": 1.7304,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.018140589569160998,
       "eval_loss": 3.188836097717285,
-      "eval_runtime": 0.4221,
-      "eval_samples_per_second": 881.252,
-      "eval_steps_per_second": 111.341,
+      "eval_runtime": 0.3979,
+      "eval_samples_per_second": 935.012,
+      "eval_steps_per_second": 118.133,
       "step": 1
     },
     {
       "epoch": 0.036281179138321996,
-      "grad_norm": 5.492339611053467,
+      "grad_norm": 5.328118324279785,
       "learning_rate": 0.0001,
       "loss": 2.1221,
       "step": 2
     },
     {
       "epoch": 0.05442176870748299,
-      "grad_norm": 5.249869346618652,
+      "grad_norm": 5.080837249755859,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 2.315,
+      "loss": 2.3173,
       "step": 3
     },
     {
       "epoch": 0.07256235827664399,
-      "grad_norm": 5.1705241203308105,
+      "grad_norm": 4.980224609375,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 2.3758,
+      "loss": 2.3835,
       "step": 4
     },
     {
       "epoch": 0.09070294784580499,
-      "grad_norm": 4.776398658752441,
+      "grad_norm": 4.697572231292725,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 2.2978,
+      "loss": 2.3059,
       "step": 5
     },
     {
       "epoch": 0.10884353741496598,
-      "grad_norm": 4.269885540008545,
+      "grad_norm": 4.239828109741211,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 2.2096,
+      "loss": 2.2238,
       "step": 6
     },
     {
       "epoch": 0.12698412698412698,
-      "grad_norm": 3.7767159938812256,
+      "grad_norm": 3.7476613521575928,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 2.2817,
+      "loss": 2.2876,
       "step": 7
     },
     {
       "epoch": 0.14512471655328799,
-      "grad_norm": 4.116052150726318,
+      "grad_norm": 4.063739776611328,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 2.4657,
+      "loss": 2.4718,
       "step": 8
     },
     {
       "epoch": 0.16326530612244897,
-      "grad_norm": 4.2486443519592285,
+      "grad_norm": 4.103112697601318,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 2.4442,
+      "loss": 2.4509,
       "step": 9
     },
     {
       "epoch": 0.18140589569160998,
-      "grad_norm": 4.38127326965332,
+      "grad_norm": 4.1713714599609375,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 2.4121,
+      "loss": 2.4224,
       "step": 10
     },
     {
       "epoch": 0.19954648526077098,
-      "grad_norm": 4.731548309326172,
+      "grad_norm": 4.444705963134766,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 2.3342,
+      "loss": 2.3554,
       "step": 11
     },
     {
       "epoch": 0.21768707482993196,
-      "grad_norm": 5.8904500007629395,
+      "grad_norm": 5.537754535675049,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 2.6067,
+      "loss": 2.634,
       "step": 12
     },
     {
       "epoch": 0.23582766439909297,
-      "grad_norm": 2.465359926223755,
+      "grad_norm": 2.4987051486968994,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 1.1479,
+      "loss": 1.1579,
       "step": 13
     },
     {
       "epoch": 0.25396825396825395,
-      "grad_norm": 2.161435127258301,
+      "grad_norm": 2.1515774726867676,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 0.8831,
+      "loss": 0.8945,
       "step": 14
     },
     {
       "epoch": 0.272108843537415,
-      "grad_norm": 2.666551351547241,
+      "grad_norm": 2.5923502445220947,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 0.9729,
+      "loss": 0.9945,
       "step": 15
     },
     {
       "epoch": 0.29024943310657597,
-      "grad_norm": 3.4165947437286377,
+      "grad_norm": 3.341392993927002,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 1.0965,
+      "loss": 1.1265,
       "step": 16
     },
     {
       "epoch": 0.30839002267573695,
-      "grad_norm": 3.7290210723876953,
+      "grad_norm": 3.689511775970459,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 1.0886,
+      "loss": 1.111,
       "step": 17
     },
     {
       "epoch": 0.32653061224489793,
-      "grad_norm": 3.5556693077087402,
+      "grad_norm": 3.7931575775146484,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 0.9905,
+      "loss": 1.0156,
       "step": 18
     },
     {
       "epoch": 0.34467120181405897,
-      "grad_norm": 3.1256349086761475,
+      "grad_norm": 3.3206164836883545,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 0.9842,
+      "loss": 1.006,
       "step": 19
     },
     {
       "epoch": 0.36281179138321995,
-      "grad_norm": 4.922985553741455,
+      "grad_norm": 3.3767266273498535,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 0.992,
+      "loss": 0.9708,
       "step": 20
     },
     {
       "epoch": 0.38095238095238093,
-      "grad_norm": 9.018806457519531,
+      "grad_norm": 6.424919128417969,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 1.0751,
+      "loss": 1.0268,
       "step": 21
     },
     {
       "epoch": 0.39909297052154197,
-      "grad_norm": 10.828701972961426,
+      "grad_norm": 8.727537155151367,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 1.0425,
+      "loss": 1.0198,
       "step": 22
     },
     {
       "epoch": 0.41723356009070295,
-      "grad_norm": 13.514071464538574,
+      "grad_norm": 11.852781295776367,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 1.1556,
+      "loss": 1.1303,
       "step": 23
     },
     {
       "epoch": 0.43537414965986393,
-      "grad_norm": 10.113996505737305,
+      "grad_norm": 9.829825401306152,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 1.2207,
+      "loss": 1.2111,
       "step": 24
     },
     {
       "epoch": 0.45351473922902497,
-      "grad_norm": 10.830758094787598,
+      "grad_norm": 10.78152847290039,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 1.4634,
+      "loss": 1.4272,
       "step": 25
     },
     {
       "epoch": 0.45351473922902497,
-      "eval_loss": 0.9609749913215637,
-      "eval_runtime": 0.42,
-      "eval_samples_per_second": 885.643,
-      "eval_steps_per_second": 111.896,
+      "eval_loss": 0.9794092774391174,
+      "eval_runtime": 0.3954,
+      "eval_samples_per_second": 940.816,
+      "eval_steps_per_second": 118.867,
       "step": 25
     },
     {
       "epoch": 0.47165532879818595,
-      "grad_norm": 10.232973098754883,
+      "grad_norm": 12.8195161819458,
       "learning_rate": 5e-05,
-      "loss": 0.7028,
+      "loss": 0.7586,
       "step": 26
     },
     {
       "epoch": 0.4897959183673469,
-      "grad_norm": 8.298254013061523,
+      "grad_norm": 10.42916202545166,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 0.721,
+      "loss": 0.7685,
       "step": 27
     },
     {
       "epoch": 0.5079365079365079,
-      "grad_norm": 3.948960304260254,
+      "grad_norm": 5.808238983154297,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 0.6604,
+      "loss": 0.6889,
       "step": 28
     },
     {
       "epoch": 0.5260770975056689,
-      "grad_norm": 2.161618947982788,
+      "grad_norm": 3.0616390705108643,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 0.6653,
+      "loss": 0.6829,
       "step": 29
     },
     {
       "epoch": 0.54421768707483,
-      "grad_norm": 2.082620859146118,
+      "grad_norm": 2.7042250633239746,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 0.6469,
+      "loss": 0.664,
       "step": 30
     },
     {
       "epoch": 0.562358276643991,
-      "grad_norm": 2.8667328357696533,
+      "grad_norm": 2.346519708633423,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 0.672,
+      "loss": 0.6699,
       "step": 31
     },
     {
       "epoch": 0.5804988662131519,
-      "grad_norm": 2.9454550743103027,
+      "grad_norm": 2.5303056240081787,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 0.6669,
+      "loss": 0.6689,
       "step": 32
     },
     {
       "epoch": 0.5986394557823129,
-      "grad_norm": 3.692382574081421,
+      "grad_norm": 3.385298013687134,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 0.7938,
+      "loss": 0.7969,
       "step": 33
     },
     {
       "epoch": 0.6167800453514739,
-      "grad_norm": 3.46189546585083,
+      "grad_norm": 3.2012507915496826,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 0.7485,
+      "loss": 0.7519,
       "step": 34
     },
     {
       "epoch": 0.6349206349206349,
-      "grad_norm": 4.580383777618408,
+      "grad_norm": 4.390873432159424,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 0.8067,
+      "loss": 0.8079,
       "step": 35
     },
     {
       "epoch": 0.6530612244897959,
-      "grad_norm": 4.773802757263184,
+      "grad_norm": 4.700263977050781,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 0.8824,
+      "loss": 0.8718,
       "step": 36
     },
     {
       "epoch": 0.671201814058957,
-      "grad_norm": 5.636185646057129,
+      "grad_norm": 5.664077281951904,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 0.8769,
+      "loss": 0.8799,
       "step": 37
     },
     {
       "epoch": 0.6893424036281179,
-      "grad_norm": 2.001661777496338,
+      "grad_norm": 2.0212113857269287,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 0.5979,
+      "loss": 0.5985,
       "step": 38
     },
     {
       "epoch": 0.7074829931972789,
-      "grad_norm": 1.8174525499343872,
+      "grad_norm": 1.7313159704208374,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 0.4685,
+      "loss": 0.4655,
       "step": 39
     },
     {
       "epoch": 0.7256235827664399,
-      "grad_norm": 1.8532778024673462,
+      "grad_norm": 1.8450844287872314,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 0.5186,
+      "loss": 0.523,
       "step": 40
     },
     {
       "epoch": 0.7437641723356009,
-      "grad_norm": 1.7148576974868774,
+      "grad_norm": 1.7504814863204956,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 0.5006,
+      "loss": 0.5081,
       "step": 41
     },
     {
       "epoch": 0.7619047619047619,
-      "grad_norm": 1.8537143468856812,
+      "grad_norm": 1.9477037191390991,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 0.5336,
+      "loss": 0.54,
       "step": 42
     },
     {
       "epoch": 0.780045351473923,
-      "grad_norm": 1.7065343856811523,
+      "grad_norm": 1.7497814893722534,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 0.5468,
+      "loss": 0.5434,
       "step": 43
     },
     {
       "epoch": 0.7981859410430839,
-      "grad_norm": 1.9868086576461792,
+      "grad_norm": 2.106696128845215,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 0.5738,
+      "loss": 0.5857,
       "step": 44
     },
     {
       "epoch": 0.8163265306122449,
-      "grad_norm": 2.079329252243042,
+      "grad_norm": 2.2159831523895264,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 0.5892,
+      "loss": 0.6028,
       "step": 45
     },
     {
       "epoch": 0.8344671201814059,
-      "grad_norm": 3.187840461730957,
+      "grad_norm": 3.368765115737915,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 0.7307,
+      "loss": 0.7448,
       "step": 46
     },
     {
       "epoch": 0.8526077097505669,
-      "grad_norm": 3.260232925415039,
+      "grad_norm": 3.3153789043426514,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 0.7631,
+      "loss": 0.7732,
       "step": 47
     },
     {
       "epoch": 0.8707482993197279,
-      "grad_norm": 2.960374355316162,
+      "grad_norm": 3.0871963500976562,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 0.7864,
+      "loss": 0.7924,
       "step": 48
     },
     {
       "epoch": 0.8888888888888888,
-      "grad_norm": 2.97607159614563,
+      "grad_norm": 3.090071201324463,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 0.795,
+      "loss": 0.7918,
       "step": 49
     },
     {
       "epoch": 0.9070294784580499,
-      "grad_norm": 6.814345836639404,
+      "grad_norm": 7.242076873779297,
       "learning_rate": 0.0,
-      "loss": 1.0176,
+      "loss": 1.0303,
       "step": 50
     },
     {
       "epoch": 0.9070294784580499,
-      "eval_loss": 0.7643183469772339,
-      "eval_runtime": 0.4208,
-      "eval_samples_per_second": 884.085,
-      "eval_steps_per_second": 111.699,
+      "eval_loss": 0.7737964987754822,
+      "eval_runtime": 0.3952,
+      "eval_samples_per_second": 941.242,
+      "eval_steps_per_second": 118.92,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:932a887e9960f63304814e79fc371c9113523e2b8f75e479ccc4b679fd509ca0
+oid sha256:e89af8d3641ca1874bca8d0938c7f9fa1bc22a0eb154a553b38786f59428706d
 size 6776