Training in progress, step 25, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
     "gate_proj",
-    "q_proj",
-    "v_proj",
     "up_proj",
     "o_proj",
-    "down_proj"
+    "v_proj",
+    "q_proj",
+    "down_proj",
+    "k_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 sha256:61949b472b952e5d15d8c65d000321fff6d83cfe540a369f7716dca28fcdde29
+oid sha256:25b1f94eaf6d39551a78ad778692a7d42f8ff5c178880051219ea14aeb94cdf2
 size 319876032

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:166dd08e065ace45675f5a7ac1b69f679edb526cd2380283c3a706cccb14c783
+oid sha256:b29ebc56513c0dba57f09b3e0433e759241e786c657aaf5e284a0f7e4efcf0c9
 size 640009682

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d4bc7bd355a3ddeda1b0d0c83f49b30fb3a31890641ae70f7ebf862afe3fd92
+oid sha256:801d7d83c24c40916f58076e00c4f26fea64e52541ba92058ab844df3c09d4d6
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae91023a435252b323366bd5c1a9210da3d37d0abed5518699297bbd167a2ea6
+oid sha256:05d983889f868d5371022e69eaef589aee748a9dbe66ba73c77df8dd836e299c
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:228191e181c69e263b04f93740d502c760e8ee44a9a2a3be15cfef6781d4cce3
+oid sha256:b4d97584e062f4742bfc7f75f541ca4dee1675e2a1e19a646f806031bbb56085
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d226d8b101d1c380fc8dff8b7d458949a74c1584eb730496e1977723528db09
+oid sha256:249d461cf3b5745c3c53b789a803f560f925013a6b22f7b4940b1b6926de54e6
 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": 2.3266286007128656e-05,
-  "best_model_checkpoint": "miner_id_24/checkpoint-50",
-  "epoch": 0.16263468184590363,
+  "best_metric": 3.5550496249925345e-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": 27.655288696289062,
+      "grad_norm": 25.735654830932617,
       "learning_rate": 5e-05,
       "loss": 13.8956,
       "step": 1
@@ -18,369 +18,186 @@
     {
       "epoch": 0.0032526936369180726,
       "eval_loss": 1.115818738937378,
-      "eval_runtime": 1.7162,
-      "eval_samples_per_second": 29.133,
-      "eval_steps_per_second": 7.575,
+      "eval_runtime": 1.7267,
+      "eval_samples_per_second": 28.956,
+      "eval_steps_per_second": 7.529,
       "step": 1
     },
     {
       "epoch": 0.006505387273836145,
-      "grad_norm": 31.58814811706543,
+      "grad_norm": 29.343381881713867,
       "learning_rate": 0.0001,
       "loss": 16.437,
       "step": 2
     },
     {
       "epoch": 0.00975808091075422,
-      "grad_norm": 29.703227996826172,
+      "grad_norm": 27.868852615356445,
       "learning_rate": 9.990365154573717e-05,
-      "loss": 15.5317,
+      "loss": 15.591,
       "step": 3
     },
     {
       "epoch": 0.01301077454767229,
-      "grad_norm": 28.027442932128906,
+      "grad_norm": 26.958335876464844,
       "learning_rate": 9.961501876182148e-05,
-      "loss": 11.9267,
+      "loss": 12.0222,
       "step": 4
     },
     {
       "epoch": 0.016263468184590364,
-      "grad_norm": 22.9697208404541,
+      "grad_norm": 22.245113372802734,
       "learning_rate": 9.913533761814537e-05,
-      "loss": 6.5239,
+      "loss": 6.5918,
       "step": 5
     },
     {
       "epoch": 0.01951616182150844,
-      "grad_norm": 14.957622528076172,
+      "grad_norm": 14.600166320800781,
       "learning_rate": 9.846666218300807e-05,
-      "loss": 3.2191,
+      "loss": 3.2538,
       "step": 6
     },
     {
       "epoch": 0.02276885545842651,
-      "grad_norm": 9.402341842651367,
+      "grad_norm": 9.080986976623535,
       "learning_rate": 9.761185582727977e-05,
-      "loss": 1.6066,
+      "loss": 1.6262,
       "step": 7
     },
     {
       "epoch": 0.02602154909534458,
-      "grad_norm": 6.7407050132751465,
+      "grad_norm": 6.721291542053223,
       "learning_rate": 9.657457896300791e-05,
-      "loss": 0.8577,
+      "loss": 0.8675,
       "step": 8
     },
     {
       "epoch": 0.029274242732262656,
-      "grad_norm": 2.6831045150756836,
+      "grad_norm": 2.7155210971832275,
       "learning_rate": 9.535927336897098e-05,
-      "loss": 0.2921,
+      "loss": 0.2972,
       "step": 9
     },
     {
       "epoch": 0.03252693636918073,
-      "grad_norm": 1.0659644603729248,
+      "grad_norm": 1.0281790494918823,
       "learning_rate": 9.397114317029975e-05,
-      "loss": 0.0745,
+      "loss": 0.0723,
       "step": 10
     },
     {
       "epoch": 0.0357796300060988,
-      "grad_norm": 0.2215704321861267,
+      "grad_norm": 0.1787344217300415,
       "learning_rate": 9.241613255361455e-05,
-      "loss": 0.0105,
+      "loss": 0.0097,
       "step": 11
     },
     {
       "epoch": 0.03903232364301688,
-      "grad_norm": 0.3032132089138031,
+      "grad_norm": 0.2821807265281677,
       "learning_rate": 9.070090031310558e-05,
-      "loss": 0.0058,
+      "loss": 0.0057,
       "step": 12
     },
     {
       "epoch": 0.042285017279934944,
-      "grad_norm": 3.7234013080596924,
+      "grad_norm": 7.026083946228027,
       "learning_rate": 8.883279133655399e-05,
-      "loss": 0.0414,
+      "loss": 0.0591,
       "step": 13
     },
     {
       "epoch": 0.04553771091685302,
-      "grad_norm": 1.0705058574676514,
+      "grad_norm": 1.1730057001113892,
       "learning_rate": 8.681980515339464e-05,
-      "loss": 0.01,
+      "loss": 0.0117,
       "step": 14
     },
     {
       "epoch": 0.048790404553771094,
-      "grad_norm": 0.7245180606842041,
+      "grad_norm": 0.7102124094963074,
       "learning_rate": 8.467056167950311e-05,
-      "loss": 0.014,
+      "loss": 0.0168,
       "step": 15
     },
     {
       "epoch": 0.05204309819068916,
-      "grad_norm": 0.06175830215215683,
+      "grad_norm": 0.05078607797622681,
       "learning_rate": 8.239426430539243e-05,
       "loss": 0.0019,
       "step": 16
     },
     {
       "epoch": 0.05529579182760724,
-      "grad_norm": 0.022349433973431587,
+      "grad_norm": 0.020944418385624886,
       "learning_rate": 8.000066048588211e-05,
       "loss": 0.0013,
       "step": 17
     },
     {
       "epoch": 0.05854848546452531,
-      "grad_norm": 0.011071293614804745,
+      "grad_norm": 0.021133584901690483,
       "learning_rate": 7.75e-05,
-      "loss": 0.0008,
+      "loss": 0.0009,
       "step": 18
     },
     {
       "epoch": 0.061801179101443386,
-      "grad_norm": 0.010401671752333641,
+      "grad_norm": 0.010110766626894474,
       "learning_rate": 7.490299105985507e-05,
       "loss": 0.0006,
       "step": 19
     },
     {
       "epoch": 0.06505387273836145,
-      "grad_norm": 0.1783440113067627,
+      "grad_norm": 0.5465399622917175,
       "learning_rate": 7.222075445642904e-05,
-      "loss": 0.0026,
+      "loss": 0.0042,
       "step": 20
     },
     {
       "epoch": 0.06830656637527953,
-      "grad_norm": 10.063867568969727,
+      "grad_norm": 10.054618835449219,
       "learning_rate": 6.946477593864228e-05,
-      "loss": 0.0462,
+      "loss": 0.0962,
       "step": 21
     },
     {
       "epoch": 0.0715592600121976,
-      "grad_norm": 0.8846852779388428,
+      "grad_norm": 0.1642262190580368,
       "learning_rate": 6.664685702961344e-05,
-      "loss": 0.006,
+      "loss": 0.0022,
       "step": 22
     },
     {
       "epoch": 0.07481195364911568,
-      "grad_norm": 0.5929298996925354,
+      "grad_norm": 0.0208920706063509,
       "learning_rate": 6.377906449072578e-05,
-      "loss": 0.0028,
+      "loss": 0.0007,
       "step": 23
     },
     {
       "epoch": 0.07806464728603375,
-      "grad_norm": 2.6959142684936523,
+      "grad_norm": 1.8869060277938843,
       "learning_rate": 6.087367864990233e-05,
-      "loss": 0.0253,
+      "loss": 0.0154,
       "step": 24
     },
     {
       "epoch": 0.08131734092295181,
-      "grad_norm": 0.7000852227210999,
+      "grad_norm": 5.405251502990723,
       "learning_rate": 5.794314081535644e-05,
-      "loss": 0.0033,
+      "loss": 0.0084,
       "step": 25
     },
     {
       "epoch": 0.08131734092295181,
-      "eval_loss": 0.0007573559996671975,
-      "eval_runtime": 1.7752,
-      "eval_samples_per_second": 28.165,
-      "eval_steps_per_second": 7.323,
+      "eval_loss": 3.5550496249925345e-05,
+      "eval_runtime": 1.7506,
+      "eval_samples_per_second": 28.561,
+      "eval_steps_per_second": 7.426,
       "step": 25
-    },
-    {
-      "epoch": 0.08457003455986989,
-      "grad_norm": 6.518759250640869,
-      "learning_rate": 5.500000000000001e-05,
-      "loss": 0.0149,
-      "step": 26
-    },
-    {
-      "epoch": 0.08782272819678796,
-      "grad_norm": 0.00867572333663702,
-      "learning_rate": 5.205685918464356e-05,
-      "loss": 0.0007,
-      "step": 27
-    },
-    {
-      "epoch": 0.09107542183370604,
-      "grad_norm": 0.02392137609422207,
-      "learning_rate": 4.912632135009769e-05,
-      "loss": 0.0009,
-      "step": 28
-    },
-    {
-      "epoch": 0.09432811547062411,
-      "grad_norm": 0.037521857768297195,
-      "learning_rate": 4.6220935509274235e-05,
-      "loss": 0.0018,
-      "step": 29
-    },
-    {
-      "epoch": 0.09758080910754219,
-      "grad_norm": 0.00691900635138154,
-      "learning_rate": 4.3353142970386564e-05,
-      "loss": 0.0006,
-      "step": 30
-    },
-    {
-      "epoch": 0.10083350274446026,
-      "grad_norm": 0.04174826294183731,
-      "learning_rate": 4.053522406135775e-05,
-      "loss": 0.0009,
-      "step": 31
-    },
-    {
-      "epoch": 0.10408619638137832,
-      "grad_norm": 0.37340205907821655,
-      "learning_rate": 3.777924554357096e-05,
-      "loss": 0.0037,
-      "step": 32
-    },
-    {
-      "epoch": 0.1073388900182964,
-      "grad_norm": 0.010010410100221634,
-      "learning_rate": 3.509700894014496e-05,
-      "loss": 0.0008,
-      "step": 33
-    },
-    {
-      "epoch": 0.11059158365521447,
-      "grad_norm": 0.03361556679010391,
-      "learning_rate": 3.250000000000001e-05,
-      "loss": 0.0011,
-      "step": 34
-    },
-    {
-      "epoch": 0.11384427729213255,
-      "grad_norm": 0.06904581934213638,
-      "learning_rate": 2.9999339514117912e-05,
-      "loss": 0.0011,
-      "step": 35
-    },
-    {
-      "epoch": 0.11709697092905062,
-      "grad_norm": 0.05339711904525757,
-      "learning_rate": 2.760573569460757e-05,
-      "loss": 0.0019,
-      "step": 36
-    },
-    {
-      "epoch": 0.1203496645659687,
-      "grad_norm": 0.024666300043463707,
-      "learning_rate": 2.53294383204969e-05,
-      "loss": 0.0011,
-      "step": 37
-    },
-    {
-      "epoch": 0.12360235820288677,
-      "grad_norm": 0.3888329565525055,
-      "learning_rate": 2.3180194846605367e-05,
-      "loss": 0.0023,
-      "step": 38
-    },
-    {
-      "epoch": 0.12685505183980483,
-      "grad_norm": 0.8227237462997437,
-      "learning_rate": 2.1167208663446025e-05,
-      "loss": 0.0052,
-      "step": 39
-    },
-    {
-      "epoch": 0.1301077454767229,
-      "grad_norm": 0.010262459516525269,
-      "learning_rate": 1.9299099686894423e-05,
-      "loss": 0.0007,
-      "step": 40
-    },
-    {
-      "epoch": 0.13336043911364098,
-      "grad_norm": 0.0066667767241597176,
-      "learning_rate": 1.758386744638546e-05,
-      "loss": 0.0006,
-      "step": 41
-    },
-    {
-      "epoch": 0.13661313275055906,
-      "grad_norm": 0.0076936231926083565,
-      "learning_rate": 1.602885682970026e-05,
-      "loss": 0.0006,
-      "step": 42
-    },
-    {
-      "epoch": 0.13986582638747713,
-      "grad_norm": 0.0788809284567833,
-      "learning_rate": 1.464072663102903e-05,
-      "loss": 0.0019,
-      "step": 43
-    },
-    {
-      "epoch": 0.1431185200243952,
-      "grad_norm": 0.004542102105915546,
-      "learning_rate": 1.3425421036992098e-05,
-      "loss": 0.0005,
-      "step": 44
-    },
-    {
-      "epoch": 0.14637121366131328,
-      "grad_norm": 0.0091813113540411,
-      "learning_rate": 1.2388144172720251e-05,
-      "loss": 0.0007,
-      "step": 45
-    },
-    {
-      "epoch": 0.14962390729823136,
-      "grad_norm": 0.01032401155680418,
-      "learning_rate": 1.1533337816991932e-05,
-      "loss": 0.0007,
-      "step": 46
-    },
-    {
-      "epoch": 0.15287660093514943,
-      "grad_norm": 0.01623600535094738,
-      "learning_rate": 1.0864662381854632e-05,
-      "loss": 0.0007,
-      "step": 47
-    },
-    {
-      "epoch": 0.1561292945720675,
-      "grad_norm": 0.006202691234648228,
-      "learning_rate": 1.0384981238178534e-05,
-      "loss": 0.0005,
-      "step": 48
-    },
-    {
-      "epoch": 0.15938198820898555,
-      "grad_norm": 0.0056791468523442745,
-      "learning_rate": 1.0096348454262845e-05,
-      "loss": 0.0006,
-      "step": 49
-    },
-    {
-      "epoch": 0.16263468184590363,
-      "grad_norm": 0.01795625500380993,
-      "learning_rate": 1e-05,
-      "loss": 0.0006,
-      "step": 50
-    },
-    {
-      "epoch": 0.16263468184590363,
-      "eval_loss": 2.3266286007128656e-05,
-      "eval_runtime": 1.7581,
-      "eval_samples_per_second": 28.439,
-      "eval_steps_per_second": 7.394,
-      "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 sha256:9f550835c1df4b23d1bbd11263feaf2340f69c1a111acddc8f9b730d2d71d240
+oid sha256:3d040487211e78fab0f1bce4f06b3f411129f652fc8c78c88243eb33088386cd
 size 6840