Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -21,12 +21,12 @@
   "revision": null,
   "target_modules": [
     "up_proj",
-    "q_proj",
-    "o_proj",
     "v_proj",
+    "o_proj",
+    "down_proj",
+    "q_proj",
     "k_proj",
-    "gate_proj",
-    "down_proj"
+    "gate_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:c9c9f5fcea190ce3edfa354c1b1207dfa263fe771f75c07167f7bb081d2502a3
+oid sha256:bf2ded8e59d9dc6a945e9d305f14e24f190d04cd7b65f6cbcfb045b227ef09c6
 size 39131224

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4fa01b52ea17d63ea6268698c0545ed096c26a04fc1cf7cceff5ff46092aac7f
+oid sha256:d7d52c981f0e413bfce471073fa3414689464702cbea9415e75528bf1acfc57e
 size 78504446

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 2.4178695678710938,
+  "best_metric": 2.4438157081604004,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.4550625711035267,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.009101251422070534,
-      "grad_norm": 2.218447208404541,
+      "grad_norm": 2.1346871852874756,
       "learning_rate": 5e-05,
       "loss": 4.9719,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.009101251422070534,
       "eval_loss": 4.942025184631348,
-      "eval_runtime": 3.9581,
-      "eval_samples_per_second": 187.213,
-      "eval_steps_per_second": 23.496,
+      "eval_runtime": 4.0731,
+      "eval_samples_per_second": 181.927,
+      "eval_steps_per_second": 22.833,
       "step": 1
     },
     {
       "epoch": 0.01820250284414107,
-      "grad_norm": 1.7499760389328003,
+      "grad_norm": 1.6914374828338623,
       "learning_rate": 0.0001,
       "loss": 4.9787,
       "step": 2
     },
     {
       "epoch": 0.027303754266211604,
-      "grad_norm": 1.5187320709228516,
+      "grad_norm": 1.4702175855636597,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 4.8311,
+      "loss": 4.8334,
       "step": 3
     },
     {
       "epoch": 0.03640500568828214,
-      "grad_norm": 1.738635778427124,
+      "grad_norm": 1.6867676973342896,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 5.0714,
+      "loss": 5.0761,
       "step": 4
     },
     {
       "epoch": 0.04550625711035267,
-      "grad_norm": 1.8651036024093628,
+      "grad_norm": 1.8130220174789429,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 5.0124,
+      "loss": 5.0172,
       "step": 5
     },
     {
       "epoch": 0.05460750853242321,
-      "grad_norm": 1.4074238538742065,
+      "grad_norm": 1.3689041137695312,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 4.4807,
+      "loss": 4.4869,
       "step": 6
     },
     {
       "epoch": 0.06370875995449374,
-      "grad_norm": 2.007289171218872,
+      "grad_norm": 1.9557944536209106,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 4.7966,
+      "loss": 4.8043,
       "step": 7
     },
     {
       "epoch": 0.07281001137656427,
-      "grad_norm": 1.4123353958129883,
+      "grad_norm": 1.3595128059387207,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 4.7974,
+      "loss": 4.8043,
       "step": 8
     },
     {
       "epoch": 0.08191126279863481,
-      "grad_norm": 1.8059446811676025,
+      "grad_norm": 1.7356663942337036,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 4.9765,
+      "loss": 4.988,
       "step": 9
     },
     {
       "epoch": 0.09101251422070535,
-      "grad_norm": 1.2806165218353271,
+      "grad_norm": 1.2412830591201782,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 4.3223,
+      "loss": 4.3269,
       "step": 10
     },
     {
       "epoch": 0.10011376564277588,
-      "grad_norm": 1.3192808628082275,
+      "grad_norm": 1.268452763557434,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 4.4645,
+      "loss": 4.4747,
       "step": 11
     },
     {
       "epoch": 0.10921501706484642,
-      "grad_norm": 1.3087587356567383,
+      "grad_norm": 1.2650614976882935,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 4.0836,
+      "loss": 4.0926,
       "step": 12
     },
     {
       "epoch": 0.11831626848691695,
-      "grad_norm": 1.2792857885360718,
+      "grad_norm": 1.2461438179016113,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 3.8406,
+      "loss": 3.8517,
       "step": 13
     },
     {
       "epoch": 0.12741751990898748,
-      "grad_norm": 1.6954691410064697,
+      "grad_norm": 1.6487263441085815,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 3.6963,
+      "loss": 3.712,
       "step": 14
     },
     {
       "epoch": 0.13651877133105803,
-      "grad_norm": 1.4813385009765625,
+      "grad_norm": 1.447868824005127,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 3.5827,
+      "loss": 3.5978,
       "step": 15
     },
     {
       "epoch": 0.14562002275312855,
-      "grad_norm": 1.468715786933899,
+      "grad_norm": 1.4297068119049072,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 3.5321,
+      "loss": 3.5478,
       "step": 16
     },
     {
       "epoch": 0.1547212741751991,
-      "grad_norm": 1.7881653308868408,
+      "grad_norm": 1.739940881729126,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 3.6404,
+      "loss": 3.6575,
       "step": 17
     },
     {
       "epoch": 0.16382252559726962,
-      "grad_norm": 1.579484224319458,
+      "grad_norm": 1.542245864868164,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 3.4393,
+      "loss": 3.4597,
       "step": 18
     },
     {
       "epoch": 0.17292377701934017,
-      "grad_norm": 1.7545942068099976,
+      "grad_norm": 1.7134660482406616,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 3.3732,
+      "loss": 3.3924,
       "step": 19
     },
     {
       "epoch": 0.1820250284414107,
-      "grad_norm": 1.4987428188323975,
+      "grad_norm": 1.4687405824661255,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 3.491,
+      "loss": 3.5119,
       "step": 20
     },
     {
       "epoch": 0.19112627986348124,
-      "grad_norm": 1.7849924564361572,
+      "grad_norm": 1.7465482950210571,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 3.8579,
+      "loss": 3.8799,
       "step": 21
     },
     {
       "epoch": 0.20022753128555176,
-      "grad_norm": 1.5346680879592896,
+      "grad_norm": 1.506808876991272,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 3.631,
+      "loss": 3.6514,
       "step": 22
     },
     {
       "epoch": 0.20932878270762229,
-      "grad_norm": 1.3760654926300049,
+      "grad_norm": 1.3481961488723755,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 3.7062,
+      "loss": 3.7276,
       "step": 23
     },
     {
       "epoch": 0.21843003412969283,
-      "grad_norm": 1.4505860805511475,
+      "grad_norm": 1.4149906635284424,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 3.5337,
+      "loss": 3.5602,
       "step": 24
     },
     {
       "epoch": 0.22753128555176336,
-      "grad_norm": 1.7248953580856323,
+      "grad_norm": 1.665215015411377,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 3.144,
+      "loss": 3.1668,
       "step": 25
     },
     {
       "epoch": 0.22753128555176336,
-      "eval_loss": 3.0551671981811523,
-      "eval_runtime": 3.9448,
-      "eval_samples_per_second": 187.844,
-      "eval_steps_per_second": 23.576,
+      "eval_loss": 3.0762650966644287,
+      "eval_runtime": 4.0834,
+      "eval_samples_per_second": 181.466,
+      "eval_steps_per_second": 22.775,
       "step": 25
     },
     {
       "epoch": 0.2366325369738339,
-      "grad_norm": 2.223836898803711,
+      "grad_norm": 2.168104410171509,
       "learning_rate": 5e-05,
-      "loss": 2.6062,
+      "loss": 2.6265,
       "step": 26
     },
     {
       "epoch": 0.24573378839590443,
-      "grad_norm": 2.0380048751831055,
+      "grad_norm": 1.9618370532989502,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 2.6892,
+      "loss": 2.7075,
       "step": 27
     },
     {
       "epoch": 0.25483503981797495,
-      "grad_norm": 1.9326778650283813,
+      "grad_norm": 1.8602323532104492,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 2.6803,
+      "loss": 2.6952,
       "step": 28
     },
     {
       "epoch": 0.26393629124004553,
-      "grad_norm": 1.8209830522537231,
+      "grad_norm": 1.7716730833053589,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 2.6056,
+      "loss": 2.6196,
       "step": 29
     },
     {
       "epoch": 0.27303754266211605,
-      "grad_norm": 2.001737117767334,
+      "grad_norm": 1.9440735578536987,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 2.5775,
+      "loss": 2.5962,
       "step": 30
     },
     {
       "epoch": 0.2821387940841866,
-      "grad_norm": 1.8633928298950195,
+      "grad_norm": 1.8115489482879639,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 2.6003,
+      "loss": 2.6244,
       "step": 31
     },
     {
       "epoch": 0.2912400455062571,
-      "grad_norm": 1.9178550243377686,
+      "grad_norm": 1.9076594114303589,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 2.4198,
+      "loss": 2.4369,
       "step": 32
     },
     {
       "epoch": 0.3003412969283277,
-      "grad_norm": 1.5524181127548218,
+      "grad_norm": 1.523880124092102,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 2.8017,
+      "loss": 2.8228,
       "step": 33
     },
     {
       "epoch": 0.3094425483503982,
-      "grad_norm": 2.2952160835266113,
+      "grad_norm": 2.241621971130371,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 3.1581,
+      "loss": 3.1889,
       "step": 34
     },
     {
       "epoch": 0.3185437997724687,
-      "grad_norm": 1.8156814575195312,
+      "grad_norm": 1.773310661315918,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 2.8874,
+      "loss": 2.9108,
       "step": 35
     },
     {
       "epoch": 0.32764505119453924,
-      "grad_norm": 1.6784733533859253,
+      "grad_norm": 1.6375255584716797,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 3.0705,
+      "loss": 3.106,
       "step": 36
     },
     {
       "epoch": 0.33674630261660976,
-      "grad_norm": 1.7788101434707642,
+      "grad_norm": 1.7255483865737915,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 2.7986,
+      "loss": 2.8329,
       "step": 37
     },
     {
       "epoch": 0.34584755403868034,
-      "grad_norm": 1.6396690607070923,
+      "grad_norm": 1.6252745389938354,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 2.2875,
+      "loss": 2.3072,
       "step": 38
     },
     {
       "epoch": 0.35494880546075086,
-      "grad_norm": 2.173144578933716,
+      "grad_norm": 2.1704254150390625,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 2.1296,
+      "loss": 2.1483,
       "step": 39
     },
     {
       "epoch": 0.3640500568828214,
-      "grad_norm": 1.8107587099075317,
+      "grad_norm": 1.8003798723220825,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 2.1515,
+      "loss": 2.1688,
       "step": 40
     },
     {
       "epoch": 0.3731513083048919,
-      "grad_norm": 1.812587022781372,
+      "grad_norm": 1.7969120740890503,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 2.1216,
+      "loss": 2.1399,
       "step": 41
     },
     {
       "epoch": 0.3822525597269625,
-      "grad_norm": 1.9823962450027466,
+      "grad_norm": 1.9582467079162598,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 2.1248,
+      "loss": 2.1466,
       "step": 42
     },
     {
       "epoch": 0.391353811149033,
-      "grad_norm": 1.938646674156189,
+      "grad_norm": 1.8976678848266602,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 2.1327,
+      "loss": 2.1565,
       "step": 43
     },
     {
       "epoch": 0.4004550625711035,
-      "grad_norm": 1.9895631074905396,
+      "grad_norm": 1.9421591758728027,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 2.0066,
+      "loss": 2.0217,
       "step": 44
     },
     {
       "epoch": 0.40955631399317405,
-      "grad_norm": 1.6511071920394897,
+      "grad_norm": 1.625119924545288,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 2.3224,
+      "loss": 2.344,
       "step": 45
     },
     {
       "epoch": 0.41865756541524457,
-      "grad_norm": 2.236948013305664,
+      "grad_norm": 2.1905601024627686,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 2.8656,
+      "loss": 2.901,
       "step": 46
     },
     {
       "epoch": 0.42775881683731515,
-      "grad_norm": 2.0217387676239014,
+      "grad_norm": 1.9888536930084229,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 2.6909,
+      "loss": 2.7267,
       "step": 47
     },
     {
       "epoch": 0.43686006825938567,
-      "grad_norm": 1.6731841564178467,
+      "grad_norm": 1.635216474533081,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 2.8569,
+      "loss": 2.8926,
       "step": 48
     },
     {
       "epoch": 0.4459613196814562,
-      "grad_norm": 1.7817379236221313,
+      "grad_norm": 1.7304415702819824,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 2.9391,
+      "loss": 2.9722,
       "step": 49
     },
     {
       "epoch": 0.4550625711035267,
-      "grad_norm": 2.3926618099212646,
+      "grad_norm": 2.324007749557495,
       "learning_rate": 0.0,
-      "loss": 2.5736,
+      "loss": 2.6059,
       "step": 50
     },
     {
       "epoch": 0.4550625711035267,
-      "eval_loss": 2.4178695678710938,
-      "eval_runtime": 3.9323,
-      "eval_samples_per_second": 188.439,
-      "eval_steps_per_second": 23.65,
+      "eval_loss": 2.4438157081604004,
+      "eval_runtime": 4.0594,
+      "eval_samples_per_second": 182.539,
+      "eval_steps_per_second": 22.91,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f1847c084eed9b0fa107ac868a39a27dca587d90916c6d92b3c3a3faeadaf97
+oid sha256:d5c0bebfbdf04b7958b1fae63dbe06a88e4db1f5fe8c714531900fb37bc3fc22
 size 6776