Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -22,8 +22,8 @@
   "target_modules": [
     "down_proj",
     "gate_up_proj",
-    "o_proj",
-    "qkv_proj"
+    "qkv_proj",
+    "o_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 sha256:357d35084e68617f7cdee2d4336d7568d5ee18d30b2ca4dc01ef3db5063be1f4
+oid sha256:8ed7b305ab3315805dc62ee383db2d5d26bab2d6017ca984db4fc653b76047d8
 size 201361312

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:42dcfdb86818d4bd0894b7263487ba73283ec84b82a03b82409dedefab94fc21
+oid sha256:573e5eb42c130cfa8ef6037fd67575ee5d06c08c123fccf4c80e14ead882fe59
 size 402868986

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eea4de1cc4842f65adeacc3d4a3f7494157fc96b8a830e7266bb471df5d24409
+oid sha256:8f8d6f03c1bdb0f44b9f34f5369d5e1a20791a69ffd583bc8cf7557d960d87ff
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d0e96d75f16313745b98055c980041ef43dd747de64cc92e2b020ac77f80c519
+oid sha256:b5a826e7137fb3972b193e4caabf48b3aff6f37cfb5f57cbecee4e0f5199ef73
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b7e3da4d46cf1158634756d333b0fbe8991960ef54757ae7dff7193156407c7
+oid sha256:3bfecddcf1149a451d1213e8bd664095be2554798b0c2c92be391b3cf6682be5
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5c05a533f4305c3ac7234753323d943feafef4ddb3d62bec8eecb4b254a7cfb
+oid sha256:6efaa2e89eee3b432f2d171c150a885e390c6d55242c2c03bf1beba45bba54ac
 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

@@ -1,203 +1,386 @@
 {
-  "best_metric": 0.06404294073581696,
-  "best_model_checkpoint": "miner_id_24/checkpoint-25",
-  "epoch": 0.2107481559536354,
+  "best_metric": 0.05814410373568535,
+  "best_model_checkpoint": "miner_id_24/checkpoint-50",
+  "epoch": 0.4214963119072708,
   "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": 0.008429926238145416,
-      "grad_norm": 1.3341302871704102,
+      "grad_norm": 1.3518892526626587,
       "learning_rate": 5e-05,
-      "loss": 1.7493,
+      "loss": 1.7474,
       "step": 1
     },
     {
       "epoch": 0.008429926238145416,
-      "eval_loss": 0.21355804800987244,
-      "eval_runtime": 128.7109,
-      "eval_samples_per_second": 6.215,
-      "eval_steps_per_second": 0.777,
+      "eval_loss": 0.21345072984695435,
+      "eval_runtime": 68.629,
+      "eval_samples_per_second": 11.657,
+      "eval_steps_per_second": 1.457,
       "step": 1
     },
     {
       "epoch": 0.01685985247629083,
-      "grad_norm": 1.6189820766448975,
+      "grad_norm": 1.6836062669754028,
       "learning_rate": 0.0001,
-      "loss": 1.5138,
+      "loss": 1.5145,
       "step": 2
     },
     {
       "epoch": 0.02528977871443625,
-      "grad_norm": 1.9272456169128418,
+      "grad_norm": 1.508198618888855,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 1.3181,
+      "loss": 1.3162,
       "step": 3
     },
     {
       "epoch": 0.03371970495258166,
-      "grad_norm": 1.0431249141693115,
+      "grad_norm": 1.0490714311599731,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 1.8563,
+      "loss": 1.8503,
       "step": 4
     },
     {
       "epoch": 0.04214963119072708,
-      "grad_norm": 1.2577431201934814,
+      "grad_norm": 0.97480309009552,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 1.3348,
+      "loss": 1.3278,
       "step": 5
     },
     {
       "epoch": 0.0505795574288725,
-      "grad_norm": 0.8566964864730835,
+      "grad_norm": 0.7965978384017944,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 1.3272,
+      "loss": 1.3214,
       "step": 6
     },
     {
       "epoch": 0.059009483667017915,
-      "grad_norm": 0.9277756214141846,
+      "grad_norm": 1.3504085540771484,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 1.4395,
+      "loss": 1.4296,
       "step": 7
     },
     {
       "epoch": 0.06743940990516333,
-      "grad_norm": 1.50494384765625,
+      "grad_norm": 3.461765766143799,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 1.9858,
+      "loss": 1.9774,
       "step": 8
     },
     {
       "epoch": 0.07586933614330875,
-      "grad_norm": 6.605533599853516,
+      "grad_norm": 7.153843879699707,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 5.6619,
+      "loss": 5.7443,
       "step": 9
     },
     {
       "epoch": 0.08429926238145416,
-      "grad_norm": 3.0556588172912598,
+      "grad_norm": 3.3547306060791016,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 0.9706,
+      "loss": 1.0362,
       "step": 10
     },
     {
       "epoch": 0.09272918861959958,
-      "grad_norm": 4.112440586090088,
+      "grad_norm": 5.1782450675964355,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 0.9054,
+      "loss": 0.9484,
       "step": 11
     },
     {
       "epoch": 0.101159114857745,
-      "grad_norm": 2.5288426876068115,
+      "grad_norm": 2.743224859237671,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 0.44,
+      "loss": 0.498,
       "step": 12
     },
     {
       "epoch": 0.1095890410958904,
-      "grad_norm": 2.0018320083618164,
+      "grad_norm": 2.1662654876708984,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 0.9965,
+      "loss": 1.0388,
       "step": 13
     },
     {
       "epoch": 0.11801896733403583,
-      "grad_norm": 1.453355073928833,
+      "grad_norm": 1.461312174797058,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 0.6349,
+      "loss": 0.6449,
       "step": 14
     },
     {
       "epoch": 0.12644889357218125,
-      "grad_norm": 1.1725133657455444,
+      "grad_norm": 1.1520819664001465,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 0.9425,
+      "loss": 0.9421,
       "step": 15
     },
     {
       "epoch": 0.13487881981032665,
-      "grad_norm": 0.9537891745567322,
+      "grad_norm": 0.8277627229690552,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 0.9102,
+      "loss": 0.9109,
       "step": 16
     },
     {
       "epoch": 0.14330874604847207,
-      "grad_norm": 1.0789560079574585,
+      "grad_norm": 1.0022519826889038,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 1.0972,
+      "loss": 1.0969,
       "step": 17
     },
     {
       "epoch": 0.1517386722866175,
-      "grad_norm": 2.126995801925659,
+      "grad_norm": 0.969890832901001,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 1.0853,
+      "loss": 1.0802,
       "step": 18
     },
     {
       "epoch": 0.1601685985247629,
-      "grad_norm": 1.467456340789795,
+      "grad_norm": 1.401827096939087,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 1.6946,
+      "loss": 1.6649,
       "step": 19
     },
     {
       "epoch": 0.16859852476290832,
-      "grad_norm": 1.6159006357192993,
+      "grad_norm": 1.4785505533218384,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 1.0233,
+      "loss": 1.0116,
       "step": 20
     },
     {
       "epoch": 0.17702845100105374,
-      "grad_norm": 3.9295225143432617,
+      "grad_norm": 3.7158868312835693,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 2.2685,
+      "loss": 2.2352,
       "step": 21
     },
     {
       "epoch": 0.18545837723919917,
-      "grad_norm": 5.1888909339904785,
+      "grad_norm": 4.5050578117370605,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 1.9563,
+      "loss": 1.9199,
       "step": 22
     },
     {
       "epoch": 0.19388830347734456,
-      "grad_norm": 0.7670079469680786,
+      "grad_norm": 0.5490809679031372,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 0.096,
+      "loss": 0.0866,
       "step": 23
     },
     {
       "epoch": 0.20231822971549,
-      "grad_norm": 0.6988094449043274,
+      "grad_norm": 0.7847179770469666,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 0.0889,
+      "loss": 0.0969,
       "step": 24
     },
     {
       "epoch": 0.2107481559536354,
-      "grad_norm": 0.9281933307647705,
+      "grad_norm": 1.2605834007263184,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 0.1051,
+      "loss": 0.1191,
       "step": 25
     },
     {
       "epoch": 0.2107481559536354,
-      "eval_loss": 0.06404294073581696,
-      "eval_runtime": 121.7973,
-      "eval_samples_per_second": 6.568,
-      "eval_steps_per_second": 0.821,
+      "eval_loss": 0.06485110521316528,
+      "eval_runtime": 68.626,
+      "eval_samples_per_second": 11.657,
+      "eval_steps_per_second": 1.457,
       "step": 25
+    },
+    {
+      "epoch": 0.2191780821917808,
+      "grad_norm": 0.5278792381286621,
+      "learning_rate": 5e-05,
+      "loss": 1.0779,
+      "step": 26
+    },
+    {
+      "epoch": 0.22760800842992623,
+      "grad_norm": 1.5710746049880981,
+      "learning_rate": 4.6729843538492847e-05,
+      "loss": 1.006,
+      "step": 27
+    },
+    {
+      "epoch": 0.23603793466807166,
+      "grad_norm": 0.8472424745559692,
+      "learning_rate": 4.347369038899744e-05,
+      "loss": 0.7343,
+      "step": 28
+    },
+    {
+      "epoch": 0.24446786090621708,
+      "grad_norm": 0.46274369955062866,
+      "learning_rate": 4.0245483899193595e-05,
+      "loss": 0.4039,
+      "step": 29
+    },
+    {
+      "epoch": 0.2528977871443625,
+      "grad_norm": 0.47283774614334106,
+      "learning_rate": 3.705904774487396e-05,
+      "loss": 0.8722,
+      "step": 30
+    },
+    {
+      "epoch": 0.2613277133825079,
+      "grad_norm": 0.5509626865386963,
+      "learning_rate": 3.392802673484193e-05,
+      "loss": 0.9503,
+      "step": 31
+    },
+    {
+      "epoch": 0.2697576396206533,
+      "grad_norm": 0.9330966472625732,
+      "learning_rate": 3.086582838174551e-05,
+      "loss": 1.839,
+      "step": 32
+    },
+    {
+      "epoch": 0.2781875658587987,
+      "grad_norm": 1.0985690355300903,
+      "learning_rate": 2.7885565489049946e-05,
+      "loss": 1.4955,
+      "step": 33
+    },
+    {
+      "epoch": 0.28661749209694415,
+      "grad_norm": 2.928253173828125,
+      "learning_rate": 2.500000000000001e-05,
+      "loss": 2.2568,
+      "step": 34
+    },
+    {
+      "epoch": 0.2950474183350896,
+      "grad_norm": 0.8522372841835022,
+      "learning_rate": 2.2221488349019903e-05,
+      "loss": 0.2512,
+      "step": 35
+    },
+    {
+      "epoch": 0.303477344573235,
+      "grad_norm": 0.28212442994117737,
+      "learning_rate": 1.9561928549563968e-05,
+      "loss": 0.0256,
+      "step": 36
+    },
+    {
+      "epoch": 0.3119072708113804,
+      "grad_norm": 1.2757818698883057,
+      "learning_rate": 1.703270924499656e-05,
+      "loss": 0.0565,
+      "step": 37
+    },
+    {
+      "epoch": 0.3203371970495258,
+      "grad_norm": 0.6521183252334595,
+      "learning_rate": 1.4644660940672627e-05,
+      "loss": 0.6645,
+      "step": 38
+    },
+    {
+      "epoch": 0.3287671232876712,
+      "grad_norm": 0.30466631054878235,
+      "learning_rate": 1.2408009626051137e-05,
+      "loss": 0.4422,
+      "step": 39
+    },
+    {
+      "epoch": 0.33719704952581664,
+      "grad_norm": 1.3516244888305664,
+      "learning_rate": 1.0332332985438248e-05,
+      "loss": 0.6052,
+      "step": 40
+    },
+    {
+      "epoch": 0.34562697576396206,
+      "grad_norm": 0.36761319637298584,
+      "learning_rate": 8.426519384872733e-06,
+      "loss": 0.5726,
+      "step": 41
+    },
+    {
+      "epoch": 0.3540569020021075,
+      "grad_norm": 0.38046741485595703,
+      "learning_rate": 6.698729810778065e-06,
+      "loss": 0.531,
+      "step": 42
+    },
+    {
+      "epoch": 0.3624868282402529,
+      "grad_norm": 0.47448211908340454,
+      "learning_rate": 5.156362923365588e-06,
+      "loss": 0.9822,
+      "step": 43
+    },
+    {
+      "epoch": 0.37091675447839834,
+      "grad_norm": 1.0301836729049683,
+      "learning_rate": 3.8060233744356633e-06,
+      "loss": 1.6145,
+      "step": 44
+    },
+    {
+      "epoch": 0.3793466807165437,
+      "grad_norm": 0.6347737908363342,
+      "learning_rate": 2.653493525244721e-06,
+      "loss": 0.6615,
+      "step": 45
+    },
+    {
+      "epoch": 0.38777660695468913,
+      "grad_norm": 1.736060380935669,
+      "learning_rate": 1.70370868554659e-06,
+      "loss": 1.9725,
+      "step": 46
+    },
+    {
+      "epoch": 0.39620653319283455,
+      "grad_norm": 2.7162489891052246,
+      "learning_rate": 9.607359798384785e-07,
+      "loss": 1.4738,
+      "step": 47
+    },
+    {
+      "epoch": 0.40463645943098,
+      "grad_norm": 0.6035640239715576,
+      "learning_rate": 4.277569313094809e-07,
+      "loss": 0.0327,
+      "step": 48
+    },
+    {
+      "epoch": 0.4130663856691254,
+      "grad_norm": 0.2597901523113251,
+      "learning_rate": 1.0705383806982606e-07,
+      "loss": 0.0292,
+      "step": 49
+    },
+    {
+      "epoch": 0.4214963119072708,
+      "grad_norm": 1.3744761943817139,
+      "learning_rate": 0.0,
+      "loss": 0.0479,
+      "step": 50
+    },
+    {
+      "epoch": 0.4214963119072708,
+      "eval_loss": 0.05814410373568535,
+      "eval_runtime": 68.0985,
+      "eval_samples_per_second": 11.748,
+      "eval_steps_per_second": 1.468,
+      "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": 2.5331937443630285e+17,
+  "total_flos": 5.086786570116137e+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 sha256:8399d9ccd5298114cbc3d90c6beef8c39db2b62b38db86fce7c63e656b9f1612
+oid sha256:d19f2eb19899cd285a1c786aa566276856c41c6665698ec6b31d3befdba6e632
 size 6776