Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -21,12 +21,12 @@
   "revision": null,
   "target_modules": [
     "k_proj",
-    "o_proj",
     "gate_proj",
     "q_proj",
-    "down_proj",
+    "o_proj",
     "up_proj",
-    "v_proj"
+    "v_proj",
+    "down_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 sha256:9699528bcbcee0ebee391faa3689828df5a8597aeb3eb6a224bf0ed81edf48c8
+oid sha256:32ea58307de55053fa70874bed7cc545719689fffef23d9a2520ebcebcc0f5bc
 size 335604696

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f9fe5804005d577da085ee06d329a4af0bc888760afaa240d4f203fcfeda958
+oid sha256:9fe61fbd7fa36c15397fecc9a1b482a7a1c3b36b433776323734e7443ea60924
 size 671466706

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33115d97fe774ab8acbd87e9845558eb899b733693eb49c2d9ae44cdb6b9c733
+oid sha256:08c2287b306f30ab999928becb3329c30ed74e5ea5e0043e2a0366f7f7ba48f7
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cdb5aef3f2d46e52af1709e5a7a4d1eae229f71757a2e5f41212b0c2e2741ecf
+oid sha256:ebc8b6e5ceea733d87822226c7a804932c6c860e94e991e254ebcbbc4a3a22e4
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca52b360cf6e3bbe821ce930c11a8a1ca297ca3f43778de4a6e2a868528ec876
+oid sha256:8a557eebe1c8f1f0508364847d35c06f8f830c39c8532f59b2c90312b678fe4c
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90e8c906057b6be69d75b03a85804748931b6ad61680b8a060fa1d2c93541745
+oid sha256:ef2e91f749206948bdff181f04fe26a5c1135b3d1b893b22f06d3a11ff5dab98
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f37b2aa490ccb1598b01e14cda36e9081f7ce646deab4d3c2d03de0d2169a755
+oid sha256:b1df0528620c07325b8faa7567e59b0c1e86a1f1ee6af1245a69c6c0463fe4e2
 size 1064

last-checkpoint/trainer_state.json

@@ -1,16 +1,16 @@
 {
-  "best_metric": 0.12282446771860123,
-  "best_model_checkpoint": "miner_id_24/checkpoint-25",
-  "epoch": 0.35810205908683973,
+  "best_metric": 0.09809810668230057,
+  "best_model_checkpoint": "miner_id_24/checkpoint-50",
+  "epoch": 0.7162041181736795,
   "eval_steps": 25,
-  "global_step": 25,
+  "global_step": 50,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.01432408236347359,
-      "grad_norm": 63.055362701416016,
+      "grad_norm": 62.79829788208008,
       "learning_rate": 5e-05,
       "loss": 1.103,
       "step": 1
@@ -18,186 +18,369 @@
     {
       "epoch": 0.01432408236347359,
       "eval_loss": 1.0215070247650146,
-      "eval_runtime": 14.3994,
-      "eval_samples_per_second": 32.64,
-      "eval_steps_per_second": 4.097,
+      "eval_runtime": 14.5171,
+      "eval_samples_per_second": 32.376,
+      "eval_steps_per_second": 4.064,
       "step": 1
     },
     {
       "epoch": 0.02864816472694718,
-      "grad_norm": 74.32926177978516,
+      "grad_norm": 73.86506652832031,
       "learning_rate": 0.0001,
       "loss": 0.9493,
       "step": 2
     },
     {
       "epoch": 0.04297224709042077,
-      "grad_norm": 72.27410888671875,
+      "grad_norm": 71.2076187133789,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 0.5497,
+      "loss": 0.5483,
       "step": 3
     },
     {
       "epoch": 0.05729632945389436,
-      "grad_norm": 46.700836181640625,
+      "grad_norm": 45.03946304321289,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 0.1979,
+      "loss": 0.1948,
       "step": 4
     },
     {
       "epoch": 0.07162041181736795,
-      "grad_norm": 19.62844467163086,
+      "grad_norm": 19.018739700317383,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 0.1001,
+      "loss": 0.1005,
       "step": 5
     },
     {
       "epoch": 0.08594449418084155,
-      "grad_norm": 19.621530532836914,
+      "grad_norm": 13.106770515441895,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 0.0659,
+      "loss": 0.0625,
       "step": 6
     },
     {
       "epoch": 0.10026857654431513,
-      "grad_norm": 16.500459671020508,
+      "grad_norm": 14.373490333557129,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 0.1086,
+      "loss": 0.1051,
       "step": 7
     },
     {
       "epoch": 0.11459265890778872,
-      "grad_norm": 16.219865798950195,
+      "grad_norm": 12.184380531311035,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 0.0392,
+      "loss": 0.0421,
       "step": 8
     },
     {
       "epoch": 0.12891674127126232,
-      "grad_norm": 27.937068939208984,
+      "grad_norm": 29.28894805908203,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 0.0779,
+      "loss": 0.0724,
       "step": 9
     },
     {
       "epoch": 0.1432408236347359,
-      "grad_norm": 12.639532089233398,
+      "grad_norm": 12.572104454040527,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 0.1651,
+      "loss": 0.1643,
       "step": 10
     },
     {
       "epoch": 0.1575649059982095,
-      "grad_norm": 45.64315414428711,
+      "grad_norm": 47.56278610229492,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 0.7781,
+      "loss": 0.7734,
       "step": 11
     },
     {
       "epoch": 0.1718889883616831,
-      "grad_norm": 51.64421463012695,
+      "grad_norm": 50.31513595581055,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 0.3408,
+      "loss": 0.3372,
       "step": 12
     },
     {
       "epoch": 0.18621307072515667,
-      "grad_norm": 55.34473419189453,
+      "grad_norm": 54.783206939697266,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 1.0015,
+      "loss": 0.9998,
       "step": 13
     },
     {
       "epoch": 0.20053715308863027,
-      "grad_norm": 66.8047103881836,
+      "grad_norm": 53.139102935791016,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 0.3822,
+      "loss": 0.3767,
       "step": 14
     },
     {
       "epoch": 0.21486123545210384,
-      "grad_norm": 57.349098205566406,
+      "grad_norm": 53.0949821472168,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 0.1784,
+      "loss": 0.1364,
       "step": 15
     },
     {
       "epoch": 0.22918531781557744,
-      "grad_norm": 23.43671226501465,
+      "grad_norm": 27.34348487854004,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 0.089,
+      "loss": 0.1032,
       "step": 16
     },
     {
       "epoch": 0.24350940017905104,
-      "grad_norm": 6.633354187011719,
+      "grad_norm": 10.79715633392334,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 0.029,
+      "loss": 0.0374,
       "step": 17
     },
     {
       "epoch": 0.25783348254252464,
-      "grad_norm": 6.5723557472229,
+      "grad_norm": 8.136823654174805,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 0.0422,
+      "loss": 0.0486,
       "step": 18
     },
     {
       "epoch": 0.2721575649059982,
-      "grad_norm": 4.889287948608398,
+      "grad_norm": 4.938852787017822,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 0.0295,
+      "loss": 0.0318,
       "step": 19
     },
     {
       "epoch": 0.2864816472694718,
-      "grad_norm": 8.272890090942383,
+      "grad_norm": 7.370121479034424,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 0.0129,
+      "loss": 0.0133,
       "step": 20
     },
     {
       "epoch": 0.3008057296329454,
-      "grad_norm": 3.8905489444732666,
+      "grad_norm": 4.348172664642334,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 0.0569,
+      "loss": 0.0585,
       "step": 21
     },
     {
       "epoch": 0.315129811996419,
-      "grad_norm": 4.120612621307373,
+      "grad_norm": 3.3651235103607178,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 0.0305,
+      "loss": 0.0307,
       "step": 22
     },
     {
       "epoch": 0.3294538943598926,
-      "grad_norm": 22.227991104125977,
+      "grad_norm": 24.419286727905273,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 0.1857,
+      "loss": 0.2009,
       "step": 23
     },
     {
       "epoch": 0.3437779767233662,
-      "grad_norm": 56.35881423950195,
+      "grad_norm": 56.320003509521484,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 0.3667,
+      "loss": 0.3678,
       "step": 24
     },
     {
       "epoch": 0.35810205908683973,
-      "grad_norm": 46.99864196777344,
+      "grad_norm": 58.8711051940918,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 0.1759,
+      "loss": 0.1835,
       "step": 25
     },
     {
       "epoch": 0.35810205908683973,
-      "eval_loss": 0.12282446771860123,
-      "eval_runtime": 14.6624,
-      "eval_samples_per_second": 32.055,
-      "eval_steps_per_second": 4.024,
+      "eval_loss": 0.12341080605983734,
+      "eval_runtime": 14.7821,
+      "eval_samples_per_second": 31.795,
+      "eval_steps_per_second": 3.991,
       "step": 25
+    },
+    {
+      "epoch": 0.37242614145031333,
+      "grad_norm": 56.06815719604492,
+      "learning_rate": 5e-05,
+      "loss": 0.6099,
+      "step": 26
+    },
+    {
+      "epoch": 0.38675022381378693,
+      "grad_norm": 15.629636764526367,
+      "learning_rate": 4.6729843538492847e-05,
+      "loss": 0.0645,
+      "step": 27
+    },
+    {
+      "epoch": 0.40107430617726053,
+      "grad_norm": 3.4101009368896484,
+      "learning_rate": 4.347369038899744e-05,
+      "loss": 0.0134,
+      "step": 28
+    },
+    {
+      "epoch": 0.41539838854073413,
+      "grad_norm": 4.199280738830566,
+      "learning_rate": 4.0245483899193595e-05,
+      "loss": 0.0218,
+      "step": 29
+    },
+    {
+      "epoch": 0.4297224709042077,
+      "grad_norm": 1.0034520626068115,
+      "learning_rate": 3.705904774487396e-05,
+      "loss": 0.0022,
+      "step": 30
+    },
+    {
+      "epoch": 0.4440465532676813,
+      "grad_norm": 1.799588680267334,
+      "learning_rate": 3.392802673484193e-05,
+      "loss": 0.01,
+      "step": 31
+    },
+    {
+      "epoch": 0.4583706356311549,
+      "grad_norm": 0.40965548157691956,
+      "learning_rate": 3.086582838174551e-05,
+      "loss": 0.0008,
+      "step": 32
+    },
+    {
+      "epoch": 0.4726947179946285,
+      "grad_norm": 0.3261648714542389,
+      "learning_rate": 2.7885565489049946e-05,
+      "loss": 0.0006,
+      "step": 33
+    },
+    {
+      "epoch": 0.4870188003581021,
+      "grad_norm": 3.464586019515991,
+      "learning_rate": 2.500000000000001e-05,
+      "loss": 0.0177,
+      "step": 34
+    },
+    {
+      "epoch": 0.5013428827215757,
+      "grad_norm": 7.937085151672363,
+      "learning_rate": 2.2221488349019903e-05,
+      "loss": 0.0466,
+      "step": 35
+    },
+    {
+      "epoch": 0.5156669650850493,
+      "grad_norm": 31.531373977661133,
+      "learning_rate": 1.9561928549563968e-05,
+      "loss": 0.2353,
+      "step": 36
+    },
+    {
+      "epoch": 0.5299910474485229,
+      "grad_norm": 22.552165985107422,
+      "learning_rate": 1.703270924499656e-05,
+      "loss": 0.1084,
+      "step": 37
+    },
+    {
+      "epoch": 0.5443151298119964,
+      "grad_norm": 21.721837997436523,
+      "learning_rate": 1.4644660940672627e-05,
+      "loss": 0.6744,
+      "step": 38
+    },
+    {
+      "epoch": 0.55863921217547,
+      "grad_norm": 28.745018005371094,
+      "learning_rate": 1.2408009626051137e-05,
+      "loss": 0.1014,
+      "step": 39
+    },
+    {
+      "epoch": 0.5729632945389436,
+      "grad_norm": 2.650120258331299,
+      "learning_rate": 1.0332332985438248e-05,
+      "loss": 0.0172,
+      "step": 40
+    },
+    {
+      "epoch": 0.5872873769024172,
+      "grad_norm": 1.9996299743652344,
+      "learning_rate": 8.426519384872733e-06,
+      "loss": 0.0164,
+      "step": 41
+    },
+    {
+      "epoch": 0.6016114592658908,
+      "grad_norm": 0.9967368841171265,
+      "learning_rate": 6.698729810778065e-06,
+      "loss": 0.0031,
+      "step": 42
+    },
+    {
+      "epoch": 0.6159355416293644,
+      "grad_norm": 1.6311664581298828,
+      "learning_rate": 5.156362923365588e-06,
+      "loss": 0.0142,
+      "step": 43
+    },
+    {
+      "epoch": 0.630259623992838,
+      "grad_norm": 0.6483197212219238,
+      "learning_rate": 3.8060233744356633e-06,
+      "loss": 0.0019,
+      "step": 44
+    },
+    {
+      "epoch": 0.6445837063563116,
+      "grad_norm": 1.0308743715286255,
+      "learning_rate": 2.653493525244721e-06,
+      "loss": 0.0114,
+      "step": 45
+    },
+    {
+      "epoch": 0.6589077887197852,
+      "grad_norm": 1.8736237287521362,
+      "learning_rate": 1.70370868554659e-06,
+      "loss": 0.0212,
+      "step": 46
+    },
+    {
+      "epoch": 0.6732318710832588,
+      "grad_norm": 1.430781602859497,
+      "learning_rate": 9.607359798384785e-07,
+      "loss": 0.0069,
+      "step": 47
+    },
+    {
+      "epoch": 0.6875559534467324,
+      "grad_norm": 13.981078147888184,
+      "learning_rate": 4.277569313094809e-07,
+      "loss": 0.0907,
+      "step": 48
+    },
+    {
+      "epoch": 0.7018800358102059,
+      "grad_norm": 21.427534103393555,
+      "learning_rate": 1.0705383806982606e-07,
+      "loss": 0.1823,
+      "step": 49
+    },
+    {
+      "epoch": 0.7162041181736795,
+      "grad_norm": 17.836923599243164,
+      "learning_rate": 0.0,
+      "loss": 0.1237,
+      "step": 50
+    },
+    {
+      "epoch": 0.7162041181736795,
+      "eval_loss": 0.09809810668230057,
+      "eval_runtime": 14.7974,
+      "eval_samples_per_second": 31.762,
+      "eval_steps_per_second": 3.987,
+      "step": 50
     }
   ],
   "logging_steps": 1,
@@ -221,12 +404,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 2.8426299460131226e+17,
+  "total_flos": 5.688797699537633e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd9b7c798852b4ea78872275e113bb8f2ff03ff3197007bb28c49e8e7cdb9d7d
+oid sha256:50c156758b7161ecf7152ecb6ee0bd178ac38c2c00fe8d346bc6f3885c57edf2
 size 6776