Training in progress, step 25, checkpoint

last-checkpoint/adapter_config.json

@@ -20,12 +20,12 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
     "v_proj",
     "o_proj",
+    "up_proj",
     "gate_proj",
+    "k_proj",
     "down_proj",
-    "up_proj",
     "q_proj"
   ],
   "task_type": "CAUSAL_LM",

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7a36d40f6204a5ec87a40e3f246efaafa4e68d50f531838ad5cd7c449cd36ca
+oid sha256:2ee41b9b4612c6e22740b56d928486b447441af09f94b7b26af3234d87b01bfa
 size 319876032

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f95329211f4372661b459a246ebea0c4211f05948296ae14f96bb9803ccbbe3
+oid sha256:82216ff0e830dc15595b941198f99a5c1c29be997505b99d58fa7654ffa6014e
 size 640009682

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:56216a6c58bc6ab3a1dfc75f31768d606f47c1006e7fa6dbe6575b10dfff1c28
+oid sha256:95fb48fd22e6c2a376769b4dbb15d468a6a253676a538f44df0b135f67fbbc41
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:99dd115d4304a37d308ad84e7b4bcd429693d63cf0ccda0df88328ba71d9f298
+oid sha256:9af0153594cc6d980b75778b7926f1dc7aad8f6cfaa494fef6200f9489aa4082
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e331ca5a81d68b4da49b6e4557e6306ef4d62f7e5131d2272125b0eba17d7a4
+oid sha256:c5f35d2221559e61bf5d837b931a6b3dc1911559d490b8418d64138f6c19888e
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aab9a08d28f671bc1d7a0dbbd5d6d65f25ae2f84d71fdc06449224f823b74309
+oid sha256:7414c99693e5554f6d248a904992a9312b6da10ca0a3b3a44f192dd6dd92e1a2
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5607f6de446164d9d9adb8b91c44cec55b14aa391e24ba5637c08b834eedda2a
+oid sha256:d271cdb95f63cd655315f063ca2e25c78dc5ae4275523c5d4f80f367586b3351
 size 1064

last-checkpoint/trainer_state.json

@@ -1,16 +1,16 @@
 {
-  "best_metric": 2.4370641767745838e-05,
-  "best_model_checkpoint": "miner_id_24/checkpoint-50",
-  "epoch": 0.16263468184590363,
+  "best_metric": 3.5030836443183944e-05,
+  "best_model_checkpoint": "miner_id_24/checkpoint-25",
+  "epoch": 0.08131734092295181,
   "eval_steps": 25,
-  "global_step": 50,
+  "global_step": 25,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.0032526936369180726,
-      "grad_norm": 26.938196182250977,
+      "grad_norm": 26.712371826171875,
       "learning_rate": 5e-05,
       "loss": 13.9096,
       "step": 1
@@ -18,369 +18,186 @@
     {
       "epoch": 0.0032526936369180726,
       "eval_loss": 1.1164056062698364,
-      "eval_runtime": 2.9495,
-      "eval_samples_per_second": 16.952,
-      "eval_steps_per_second": 4.407,
+      "eval_runtime": 2.9801,
+      "eval_samples_per_second": 16.778,
+      "eval_steps_per_second": 4.362,
       "step": 1
     },
     {
       "epoch": 0.006505387273836145,
-      "grad_norm": 30.800067901611328,
+      "grad_norm": 30.42289161682129,
       "learning_rate": 0.0001,
       "loss": 16.4353,
       "step": 2
     },
     {
       "epoch": 0.00975808091075422,
-      "grad_norm": 28.980260848999023,
+      "grad_norm": 28.80431365966797,
       "learning_rate": 9.990365154573717e-05,
-      "loss": 15.5479,
+      "loss": 15.5446,
       "step": 3
     },
     {
       "epoch": 0.01301077454767229,
-      "grad_norm": 27.222143173217773,
+      "grad_norm": 27.237125396728516,
       "learning_rate": 9.961501876182148e-05,
-      "loss": 11.9474,
+      "loss": 11.9457,
       "step": 4
     },
     {
       "epoch": 0.016263468184590364,
-      "grad_norm": 22.249568939208984,
+      "grad_norm": 22.341909408569336,
       "learning_rate": 9.913533761814537e-05,
-      "loss": 6.5466,
+      "loss": 6.5323,
       "step": 5
     },
     {
       "epoch": 0.01951616182150844,
-      "grad_norm": 15.059317588806152,
+      "grad_norm": 15.01369857788086,
       "learning_rate": 9.846666218300807e-05,
-      "loss": 3.252,
+      "loss": 3.2113,
       "step": 6
     },
     {
       "epoch": 0.02276885545842651,
-      "grad_norm": 9.444744110107422,
+      "grad_norm": 9.229114532470703,
       "learning_rate": 9.761185582727977e-05,
-      "loss": 1.6042,
+      "loss": 1.575,
       "step": 7
     },
     {
       "epoch": 0.02602154909534458,
-      "grad_norm": 6.794412612915039,
+      "grad_norm": 6.6448798179626465,
       "learning_rate": 9.657457896300791e-05,
-      "loss": 0.8631,
+      "loss": 0.8369,
       "step": 8
     },
     {
       "epoch": 0.029274242732262656,
-      "grad_norm": 2.7923848628997803,
+      "grad_norm": 2.6261746883392334,
       "learning_rate": 9.535927336897098e-05,
-      "loss": 0.2988,
+      "loss": 0.281,
       "step": 9
     },
     {
       "epoch": 0.03252693636918073,
-      "grad_norm": 1.0953826904296875,
+      "grad_norm": 1.0150703191757202,
       "learning_rate": 9.397114317029975e-05,
-      "loss": 0.0755,
+      "loss": 0.0699,
       "step": 10
     },
     {
       "epoch": 0.0357796300060988,
-      "grad_norm": 0.1885637640953064,
+      "grad_norm": 0.1931096911430359,
       "learning_rate": 9.241613255361455e-05,
-      "loss": 0.0101,
+      "loss": 0.0098,
       "step": 11
     },
     {
       "epoch": 0.03903232364301688,
-      "grad_norm": 0.19117560982704163,
+      "grad_norm": 0.23602637648582458,
       "learning_rate": 9.070090031310558e-05,
-      "loss": 0.0051,
+      "loss": 0.0052,
       "step": 12
     },
     {
       "epoch": 0.042285017279934944,
-      "grad_norm": 9.700277328491211,
+      "grad_norm": 8.177572250366211,
       "learning_rate": 8.883279133655399e-05,
-      "loss": 0.0693,
+      "loss": 0.0579,
       "step": 13
     },
     {
       "epoch": 0.04553771091685302,
-      "grad_norm": 1.511494755744934,
+      "grad_norm": 1.2461020946502686,
       "learning_rate": 8.681980515339464e-05,
-      "loss": 0.0186,
+      "loss": 0.0166,
       "step": 14
     },
     {
       "epoch": 0.048790404553771094,
-      "grad_norm": 0.6943612098693848,
+      "grad_norm": 0.680696964263916,
       "learning_rate": 8.467056167950311e-05,
-      "loss": 0.0167,
+      "loss": 0.0155,
       "step": 15
     },
     {
       "epoch": 0.05204309819068916,
-      "grad_norm": 0.07114950567483902,
+      "grad_norm": 0.06047433242201805,
       "learning_rate": 8.239426430539243e-05,
-      "loss": 0.0022,
+      "loss": 0.0021,
       "step": 16
     },
     {
       "epoch": 0.05529579182760724,
-      "grad_norm": 0.027020208537578583,
+      "grad_norm": 0.023200832307338715,
       "learning_rate": 8.000066048588211e-05,
       "loss": 0.0014,
       "step": 17
     },
     {
       "epoch": 0.05854848546452531,
-      "grad_norm": 0.8249042630195618,
+      "grad_norm": 3.277186870574951,
       "learning_rate": 7.75e-05,
-      "loss": 0.002,
+      "loss": 0.0114,
       "step": 18
     },
     {
       "epoch": 0.061801179101443386,
-      "grad_norm": 0.04792841151356697,
+      "grad_norm": 0.02421419695019722,
       "learning_rate": 7.490299105985507e-05,
       "loss": 0.0007,
       "step": 19
     },
     {
       "epoch": 0.06505387273836145,
-      "grad_norm": 0.5568363070487976,
+      "grad_norm": 0.599700391292572,
       "learning_rate": 7.222075445642904e-05,
-      "loss": 0.004,
+      "loss": 0.0041,
       "step": 20
     },
     {
       "epoch": 0.06830656637527953,
-      "grad_norm": 0.04415999725461006,
+      "grad_norm": 0.2749795615673065,
       "learning_rate": 6.946477593864228e-05,
-      "loss": 0.0009,
+      "loss": 0.0016,
       "step": 21
     },
     {
       "epoch": 0.0715592600121976,
-      "grad_norm": 0.12361333519220352,
+      "grad_norm": 0.12743216753005981,
       "learning_rate": 6.664685702961344e-05,
-      "loss": 0.0019,
+      "loss": 0.0018,
       "step": 22
     },
     {
       "epoch": 0.07481195364911568,
-      "grad_norm": 0.5670086145401001,
+      "grad_norm": 1.3546181917190552,
       "learning_rate": 6.377906449072578e-05,
-      "loss": 0.0017,
+      "loss": 0.0044,
       "step": 23
     },
     {
       "epoch": 0.07806464728603375,
-      "grad_norm": 1.2928516864776611,
+      "grad_norm": 1.1135410070419312,
       "learning_rate": 6.087367864990233e-05,
-      "loss": 0.0069,
+      "loss": 0.0136,
       "step": 24
     },
     {
       "epoch": 0.08131734092295181,
-      "grad_norm": 2.51751971244812,
+      "grad_norm": 4.629287242889404,
       "learning_rate": 5.794314081535644e-05,
-      "loss": 0.0133,
+      "loss": 0.0087,
       "step": 25
     },
     {
       "epoch": 0.08131734092295181,
-      "eval_loss": 3.148895120830275e-05,
-      "eval_runtime": 2.9982,
-      "eval_samples_per_second": 16.676,
-      "eval_steps_per_second": 4.336,
+      "eval_loss": 3.5030836443183944e-05,
+      "eval_runtime": 2.9973,
+      "eval_samples_per_second": 16.682,
+      "eval_steps_per_second": 4.337,
       "step": 25
-    },
-    {
-      "epoch": 0.08457003455986989,
-      "grad_norm": 0.2348364293575287,
-      "learning_rate": 5.500000000000001e-05,
-      "loss": 0.0016,
-      "step": 26
-    },
-    {
-      "epoch": 0.08782272819678796,
-      "grad_norm": 0.16821667551994324,
-      "learning_rate": 5.205685918464356e-05,
-      "loss": 0.0013,
-      "step": 27
-    },
-    {
-      "epoch": 0.09107542183370604,
-      "grad_norm": 0.030762257054448128,
-      "learning_rate": 4.912632135009769e-05,
-      "loss": 0.001,
-      "step": 28
-    },
-    {
-      "epoch": 0.09432811547062411,
-      "grad_norm": 0.03775183856487274,
-      "learning_rate": 4.6220935509274235e-05,
-      "loss": 0.0017,
-      "step": 29
-    },
-    {
-      "epoch": 0.09758080910754219,
-      "grad_norm": 0.006697505712509155,
-      "learning_rate": 4.3353142970386564e-05,
-      "loss": 0.0006,
-      "step": 30
-    },
-    {
-      "epoch": 0.10083350274446026,
-      "grad_norm": 0.024879690259695053,
-      "learning_rate": 4.053522406135775e-05,
-      "loss": 0.0008,
-      "step": 31
-    },
-    {
-      "epoch": 0.10408619638137832,
-      "grad_norm": 0.30223092436790466,
-      "learning_rate": 3.777924554357096e-05,
-      "loss": 0.0032,
-      "step": 32
-    },
-    {
-      "epoch": 0.1073388900182964,
-      "grad_norm": 0.014068452641367912,
-      "learning_rate": 3.509700894014496e-05,
-      "loss": 0.001,
-      "step": 33
-    },
-    {
-      "epoch": 0.11059158365521447,
-      "grad_norm": 0.023483309894800186,
-      "learning_rate": 3.250000000000001e-05,
-      "loss": 0.0009,
-      "step": 34
-    },
-    {
-      "epoch": 0.11384427729213255,
-      "grad_norm": 0.01585865579545498,
-      "learning_rate": 2.9999339514117912e-05,
-      "loss": 0.0009,
-      "step": 35
-    },
-    {
-      "epoch": 0.11709697092905062,
-      "grad_norm": 0.04529554396867752,
-      "learning_rate": 2.760573569460757e-05,
-      "loss": 0.0017,
-      "step": 36
-    },
-    {
-      "epoch": 0.1203496645659687,
-      "grad_norm": 0.25387975573539734,
-      "learning_rate": 2.53294383204969e-05,
-      "loss": 0.0021,
-      "step": 37
-    },
-    {
-      "epoch": 0.12360235820288677,
-      "grad_norm": 0.2627509832382202,
-      "learning_rate": 2.3180194846605367e-05,
-      "loss": 0.0035,
-      "step": 38
-    },
-    {
-      "epoch": 0.12685505183980483,
-      "grad_norm": 0.026653291657567024,
-      "learning_rate": 2.1167208663446025e-05,
-      "loss": 0.0006,
-      "step": 39
-    },
-    {
-      "epoch": 0.1301077454767229,
-      "grad_norm": 0.009845051914453506,
-      "learning_rate": 1.9299099686894423e-05,
-      "loss": 0.0008,
-      "step": 40
-    },
-    {
-      "epoch": 0.13336043911364098,
-      "grad_norm": 0.006667473819106817,
-      "learning_rate": 1.758386744638546e-05,
-      "loss": 0.0006,
-      "step": 41
-    },
-    {
-      "epoch": 0.13661313275055906,
-      "grad_norm": 0.04519423469901085,
-      "learning_rate": 1.602885682970026e-05,
-      "loss": 0.0008,
-      "step": 42
-    },
-    {
-      "epoch": 0.13986582638747713,
-      "grad_norm": 0.7729731798171997,
-      "learning_rate": 1.464072663102903e-05,
-      "loss": 0.0024,
-      "step": 43
-    },
-    {
-      "epoch": 0.1431185200243952,
-      "grad_norm": 0.004916268866509199,
-      "learning_rate": 1.3425421036992098e-05,
-      "loss": 0.0005,
-      "step": 44
-    },
-    {
-      "epoch": 0.14637121366131328,
-      "grad_norm": 0.009590042755007744,
-      "learning_rate": 1.2388144172720251e-05,
-      "loss": 0.0007,
-      "step": 45
-    },
-    {
-      "epoch": 0.14962390729823136,
-      "grad_norm": 0.014254098758101463,
-      "learning_rate": 1.1533337816991932e-05,
-      "loss": 0.0008,
-      "step": 46
-    },
-    {
-      "epoch": 0.15287660093514943,
-      "grad_norm": 1.3294068574905396,
-      "learning_rate": 1.0864662381854632e-05,
-      "loss": 0.0064,
-      "step": 47
-    },
-    {
-      "epoch": 0.1561292945720675,
-      "grad_norm": 0.00593988411128521,
-      "learning_rate": 1.0384981238178534e-05,
-      "loss": 0.0005,
-      "step": 48
-    },
-    {
-      "epoch": 0.15938198820898555,
-      "grad_norm": 0.008307484909892082,
-      "learning_rate": 1.0096348454262845e-05,
-      "loss": 0.0007,
-      "step": 49
-    },
-    {
-      "epoch": 0.16263468184590363,
-      "grad_norm": 0.01137072965502739,
-      "learning_rate": 1e-05,
-      "loss": 0.0006,
-      "step": 50
-    },
-    {
-      "epoch": 0.16263468184590363,
-      "eval_loss": 2.4370641767745838e-05,
-      "eval_runtime": 3.0145,
-      "eval_samples_per_second": 16.586,
-      "eval_steps_per_second": 4.312,
-      "step": 50
     }
   ],
   "logging_steps": 1,
@@ -404,12 +221,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 5.259156315439104e+17,
+  "total_flos": 2.629578157719552e+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 sha256:5fbddc24dc8e158f7195073d82d89af0ce54626aadbd80692ebbc80c82758115
+oid sha256:4b1f808783ce62c45f596b9f6176b1ca3fb6a402d62b2e7ed676ba4372464fd3
 size 6840