Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,9 +20,9 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "down_proj",
     "qkv_proj",
     "o_proj",
-    "down_proj",
     "gate_up_proj"
   ],
   "task_type": "CAUSAL_LM",

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ef7aa1ad3133f7c084d92b42411b3352f03041e55803e0729e6e094fe1999ee6
+oid sha256:57220ba9a17fa6e4dba34f0bc82eabe7b1dba570ca47ce468ea5aa1bf4fd0748
 size 201361312

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ebc796025b869ce96594882df58f996e9964d53b03548ee20ace609fea30ab99
+oid sha256:64b5843437d5b65c7b6e205e61ab53ff2d7abfa4d98031aa7639591a9142a426
 size 402868986

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:40b137c9e96cb8c5c4c0532a3fdd50fd1de41cf1d218126475eb72f4fb4969ea
+oid sha256:dd294e960b9f86cfd37b663bfbc08bb7f3b77a0463fcb0a75b57ddffc39a55d5
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9fe5ec914316b142cb01c52581917a9046194b3a5c200d109a1f2164f3e1b00b
+oid sha256:8b69092f4065b6b055cc305390951cb5821a6dd807eafbdffbff807abd2d104b
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57bac477237fa0331b01a701ba0a1fcb13fe910b76955c97b061d11fa2b9e2b8
+oid sha256:14c66f1dab56b37b4c5b2cee28a56f07c718ae725ac7ef6aad7463c71591c9b9
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6ae013deb2c91d8feb944ad3f0545aff3b22598faffe3411680df996f8788d5a
+oid sha256:aab836f25af2e8c622ba5993e289989969ad90290c59be22254ae0ee3df0f6a1
 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": 0.23452582955360413,
-  "best_model_checkpoint": "miner_id_24/checkpoint-25",
-  "epoch": 0.01297016861219196,
+  "best_metric": 0.20182333886623383,
+  "best_model_checkpoint": "miner_id_24/checkpoint-50",
+  "epoch": 0.02594033722438392,
   "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": 0.0005188067444876783,
-      "grad_norm": 7.088029384613037,
+      "grad_norm": 7.257709980010986,
       "learning_rate": 5e-05,
       "loss": 0.8421,
       "step": 1
@@ -18,186 +18,369 @@
     {
       "epoch": 0.0005188067444876783,
       "eval_loss": 0.9019366502761841,
-      "eval_runtime": 439.5345,
-      "eval_samples_per_second": 29.543,
-      "eval_steps_per_second": 3.695,
+      "eval_runtime": 437.3032,
+      "eval_samples_per_second": 29.693,
+      "eval_steps_per_second": 3.714,
       "step": 1
     },
     {
       "epoch": 0.0010376134889753567,
-      "grad_norm": 9.270331382751465,
+      "grad_norm": 9.514312744140625,
       "learning_rate": 0.0001,
       "loss": 0.7793,
       "step": 2
     },
     {
       "epoch": 0.0015564202334630351,
-      "grad_norm": 13.402799606323242,
+      "grad_norm": 13.657303810119629,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 0.8548,
+      "loss": 0.8538,
       "step": 3
     },
     {
       "epoch": 0.0020752269779507134,
-      "grad_norm": 13.478734016418457,
+      "grad_norm": 13.498005867004395,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 0.7784,
+      "loss": 0.7789,
       "step": 4
     },
     {
       "epoch": 0.0025940337224383916,
-      "grad_norm": 15.679061889648438,
+      "grad_norm": 15.755859375,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 0.596,
+      "loss": 0.6017,
       "step": 5
     },
     {
       "epoch": 0.0031128404669260703,
-      "grad_norm": 13.199936866760254,
+      "grad_norm": 13.434879302978516,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 0.4246,
+      "loss": 0.4315,
       "step": 6
     },
     {
       "epoch": 0.0036316472114137485,
-      "grad_norm": 8.329848289489746,
+      "grad_norm": 8.647991180419922,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 0.2895,
+      "loss": 0.2934,
       "step": 7
     },
     {
       "epoch": 0.004150453955901427,
-      "grad_norm": 5.857521057128906,
+      "grad_norm": 5.960851192474365,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 0.2731,
+      "loss": 0.2742,
       "step": 8
     },
     {
       "epoch": 0.004669260700389105,
-      "grad_norm": 5.562647342681885,
+      "grad_norm": 5.692524433135986,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 0.2526,
+      "loss": 0.254,
       "step": 9
     },
     {
       "epoch": 0.005188067444876783,
-      "grad_norm": 6.4964423179626465,
+      "grad_norm": 6.573745250701904,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 0.2667,
+      "loss": 0.266,
       "step": 10
     },
     {
       "epoch": 0.005706874189364461,
-      "grad_norm": 7.804101467132568,
+      "grad_norm": 7.913269996643066,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 0.2699,
+      "loss": 0.2703,
       "step": 11
     },
     {
       "epoch": 0.0062256809338521405,
-      "grad_norm": 7.244912147521973,
+      "grad_norm": 7.290318489074707,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 0.2022,
+      "loss": 0.202,
       "step": 12
     },
     {
       "epoch": 0.006744487678339819,
-      "grad_norm": 10.575514793395996,
+      "grad_norm": 10.77153491973877,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 0.5618,
+      "loss": 0.5624,
       "step": 13
     },
     {
       "epoch": 0.007263294422827497,
-      "grad_norm": 13.16678524017334,
+      "grad_norm": 13.511357307434082,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 0.564,
+      "loss": 0.5687,
       "step": 14
     },
     {
       "epoch": 0.007782101167315175,
-      "grad_norm": 9.636844635009766,
+      "grad_norm": 9.808670043945312,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 0.4095,
+      "loss": 0.4122,
       "step": 15
     },
     {
       "epoch": 0.008300907911802853,
-      "grad_norm": 7.153684139251709,
+      "grad_norm": 7.290343761444092,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 0.2974,
+      "loss": 0.2995,
       "step": 16
     },
     {
       "epoch": 0.008819714656290532,
-      "grad_norm": 7.988699913024902,
+      "grad_norm": 8.092347145080566,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 0.3181,
+      "loss": 0.3212,
       "step": 17
     },
     {
       "epoch": 0.00933852140077821,
-      "grad_norm": 9.76743221282959,
+      "grad_norm": 9.879920959472656,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 0.2727,
+      "loss": 0.2751,
       "step": 18
     },
     {
       "epoch": 0.009857328145265888,
-      "grad_norm": 7.5284318923950195,
+      "grad_norm": 7.612329483032227,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 0.2183,
+      "loss": 0.2201,
       "step": 19
     },
     {
       "epoch": 0.010376134889753566,
-      "grad_norm": 5.3470563888549805,
+      "grad_norm": 5.5150322914123535,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 0.1889,
+      "loss": 0.1909,
       "step": 20
     },
     {
       "epoch": 0.010894941634241245,
-      "grad_norm": 3.7029356956481934,
+      "grad_norm": 3.7366225719451904,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 0.1763,
+      "loss": 0.1765,
       "step": 21
     },
     {
       "epoch": 0.011413748378728923,
-      "grad_norm": 4.031704425811768,
+      "grad_norm": 4.036696434020996,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 0.1656,
+      "loss": 0.1671,
       "step": 22
     },
     {
       "epoch": 0.011932555123216601,
-      "grad_norm": 5.219779014587402,
+      "grad_norm": 5.270599365234375,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 0.1802,
+      "loss": 0.1808,
       "step": 23
     },
     {
       "epoch": 0.012451361867704281,
-      "grad_norm": 7.983980178833008,
+      "grad_norm": 8.023921012878418,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 0.2019,
+      "loss": 0.2025,
       "step": 24
     },
     {
       "epoch": 0.01297016861219196,
-      "grad_norm": 8.021879196166992,
+      "grad_norm": 8.021526336669922,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 0.2279,
+      "loss": 0.2274,
       "step": 25
     },
     {
       "epoch": 0.01297016861219196,
-      "eval_loss": 0.23452582955360413,
-      "eval_runtime": 439.5012,
-      "eval_samples_per_second": 29.545,
-      "eval_steps_per_second": 3.695,
+      "eval_loss": 0.235469788312912,
+      "eval_runtime": 437.5793,
+      "eval_samples_per_second": 29.675,
+      "eval_steps_per_second": 3.711,
       "step": 25
+    },
+    {
+      "epoch": 0.013488975356679637,
+      "grad_norm": 5.151494979858398,
+      "learning_rate": 5e-05,
+      "loss": 0.625,
+      "step": 26
+    },
+    {
+      "epoch": 0.014007782101167316,
+      "grad_norm": 5.601419448852539,
+      "learning_rate": 4.6729843538492847e-05,
+      "loss": 0.4202,
+      "step": 27
+    },
+    {
+      "epoch": 0.014526588845654994,
+      "grad_norm": 5.864795207977295,
+      "learning_rate": 4.347369038899744e-05,
+      "loss": 0.2736,
+      "step": 28
+    },
+    {
+      "epoch": 0.015045395590142672,
+      "grad_norm": 4.344137668609619,
+      "learning_rate": 4.0245483899193595e-05,
+      "loss": 0.2721,
+      "step": 29
+    },
+    {
+      "epoch": 0.01556420233463035,
+      "grad_norm": 3.193909168243408,
+      "learning_rate": 3.705904774487396e-05,
+      "loss": 0.2,
+      "step": 30
+    },
+    {
+      "epoch": 0.016083009079118027,
+      "grad_norm": 2.997635841369629,
+      "learning_rate": 3.392802673484193e-05,
+      "loss": 0.1727,
+      "step": 31
+    },
+    {
+      "epoch": 0.016601815823605707,
+      "grad_norm": 2.729241371154785,
+      "learning_rate": 3.086582838174551e-05,
+      "loss": 0.1658,
+      "step": 32
+    },
+    {
+      "epoch": 0.017120622568093387,
+      "grad_norm": 2.756976842880249,
+      "learning_rate": 2.7885565489049946e-05,
+      "loss": 0.1545,
+      "step": 33
+    },
+    {
+      "epoch": 0.017639429312581063,
+      "grad_norm": 2.4652767181396484,
+      "learning_rate": 2.500000000000001e-05,
+      "loss": 0.1408,
+      "step": 34
+    },
+    {
+      "epoch": 0.018158236057068743,
+      "grad_norm": 2.768718957901001,
+      "learning_rate": 2.2221488349019903e-05,
+      "loss": 0.1777,
+      "step": 35
+    },
+    {
+      "epoch": 0.01867704280155642,
+      "grad_norm": 2.951350212097168,
+      "learning_rate": 1.9561928549563968e-05,
+      "loss": 0.1508,
+      "step": 36
+    },
+    {
+      "epoch": 0.0191958495460441,
+      "grad_norm": 4.257266521453857,
+      "learning_rate": 1.703270924499656e-05,
+      "loss": 0.1418,
+      "step": 37
+    },
+    {
+      "epoch": 0.019714656290531776,
+      "grad_norm": 4.354798793792725,
+      "learning_rate": 1.4644660940672627e-05,
+      "loss": 0.421,
+      "step": 38
+    },
+    {
+      "epoch": 0.020233463035019456,
+      "grad_norm": 3.913156032562256,
+      "learning_rate": 1.2408009626051137e-05,
+      "loss": 0.378,
+      "step": 39
+    },
+    {
+      "epoch": 0.020752269779507133,
+      "grad_norm": 3.5373024940490723,
+      "learning_rate": 1.0332332985438248e-05,
+      "loss": 0.2926,
+      "step": 40
+    },
+    {
+      "epoch": 0.021271076523994813,
+      "grad_norm": 4.939075946807861,
+      "learning_rate": 8.426519384872733e-06,
+      "loss": 0.1924,
+      "step": 41
+    },
+    {
+      "epoch": 0.02178988326848249,
+      "grad_norm": 3.2499287128448486,
+      "learning_rate": 6.698729810778065e-06,
+      "loss": 0.2142,
+      "step": 42
+    },
+    {
+      "epoch": 0.02230869001297017,
+      "grad_norm": 2.4298489093780518,
+      "learning_rate": 5.156362923365588e-06,
+      "loss": 0.1793,
+      "step": 43
+    },
+    {
+      "epoch": 0.022827496757457846,
+      "grad_norm": 2.3704288005828857,
+      "learning_rate": 3.8060233744356633e-06,
+      "loss": 0.1485,
+      "step": 44
+    },
+    {
+      "epoch": 0.023346303501945526,
+      "grad_norm": 2.584822177886963,
+      "learning_rate": 2.653493525244721e-06,
+      "loss": 0.1585,
+      "step": 45
+    },
+    {
+      "epoch": 0.023865110246433202,
+      "grad_norm": 2.569350481033325,
+      "learning_rate": 1.70370868554659e-06,
+      "loss": 0.1276,
+      "step": 46
+    },
+    {
+      "epoch": 0.024383916990920882,
+      "grad_norm": 2.4440035820007324,
+      "learning_rate": 9.607359798384785e-07,
+      "loss": 0.1723,
+      "step": 47
+    },
+    {
+      "epoch": 0.024902723735408562,
+      "grad_norm": 2.5331945419311523,
+      "learning_rate": 4.277569313094809e-07,
+      "loss": 0.1476,
+      "step": 48
+    },
+    {
+      "epoch": 0.02542153047989624,
+      "grad_norm": 3.334847927093506,
+      "learning_rate": 1.0705383806982606e-07,
+      "loss": 0.1711,
+      "step": 49
+    },
+    {
+      "epoch": 0.02594033722438392,
+      "grad_norm": 3.929615020751953,
+      "learning_rate": 0.0,
+      "loss": 0.1231,
+      "step": 50
+    },
+    {
+      "epoch": 0.02594033722438392,
+      "eval_loss": 0.20182333886623383,
+      "eval_runtime": 437.3533,
+      "eval_samples_per_second": 29.69,
+      "eval_steps_per_second": 3.713,
+      "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": 1.4854232815855206e+17,
+  "total_flos": 2.9717737929231565e+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 sha256:32c9a617800bcf00617a974e9c1ac0e7ad0ab0a36eece5b682859f66cccab4d2
+oid sha256:3b4c867fc22c662d2eb55140f540ec9341dba7a58b87b7576d7182430af5852d
 size 6776