Training in progress, step 25, checkpoint

last-checkpoint/adapter_config.json

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

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c3f1085970e10fc875db136ec23891438513b032dbac0aa6c92740631c9d0e29
+oid sha256:aa54acbae13b542c0a44a2df9e8d9bc6910b4bfc0e1e72b4a1f5d31f2aa7535f
 size 319876032

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9638839cc7aa7ad4e1d815c7e910d5f798f763ac52b4ba04dba53d7f1afd7ae
+oid sha256:69bfe5eb1b9ac7d5621367d5756e3b934b0c07a345c3b297c22199373b289413
 size 640009682

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71f3e99bab5910a9eb9869e9cff396e4392854837bacbf6f33794bd4244c02c5
+oid sha256:cb407202bbcf1c0e8e88edfebb19feaed744e54b180e5617ade49956433eb099
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b3bb4016113e7b6d5fd848508e844409ddb242416dab96b071055a19c2cbd61
+oid sha256:0a9422ed0ba7efd63f8bb85a58f63aab48d3cc7faa7458846d7e794325280d10
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b0373480498ccb6580f361fbb9f0baac56bc12706dfb2fc14f4dbdfc5ad7b17e
+oid sha256:11ef8648f92511a0b6896f8f73ed395046d781b21635cc3b50fbdf69358efa45
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:45eae6409aa657abdfae0cf93ea9475a05d27d5dd786345e774660695f5157ba
+oid sha256:b5413b46c45f1bcddff958a622dbf892d22de4c0b0bee77779945545e2f6e409
 size 15024

last-checkpoint/scheduler.pt

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

last-checkpoint/trainer_state.json

@@ -1,386 +1,203 @@
 {
-  "best_metric": 0.986032247543335,
-  "best_model_checkpoint": "miner_id_24/checkpoint-50",
-  "epoch": 0.411946446961895,
+  "best_metric": 1.097769021987915,
+  "best_model_checkpoint": "miner_id_24/checkpoint-25",
+  "epoch": 0.2059732234809475,
   "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": 0.008238928939237899,
-      "grad_norm": 1.3493034839630127,
+      "grad_norm": 1.285041093826294,
       "learning_rate": 5e-05,
-      "loss": 1.3213,
+      "loss": 1.3212,
       "step": 1
     },
     {
       "epoch": 0.008238928939237899,
-      "eval_loss": 1.754831314086914,
-      "eval_runtime": 28.9839,
-      "eval_samples_per_second": 28.223,
-      "eval_steps_per_second": 3.554,
+      "eval_loss": 1.754696011543274,
+      "eval_runtime": 57.0263,
+      "eval_samples_per_second": 14.344,
+      "eval_steps_per_second": 1.806,
       "step": 1
     },
     {
       "epoch": 0.016477857878475798,
-      "grad_norm": 1.0823650360107422,
+      "grad_norm": 1.0165727138519287,
       "learning_rate": 0.0001,
-      "loss": 1.5151,
+      "loss": 1.5153,
       "step": 2
     },
     {
       "epoch": 0.024716786817713696,
-      "grad_norm": 1.0006831884384155,
+      "grad_norm": 0.9726688265800476,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 0.9711,
+      "loss": 0.9716,
       "step": 3
     },
     {
       "epoch": 0.032955715756951595,
-      "grad_norm": 0.5157923102378845,
+      "grad_norm": 0.4960082769393921,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 1.1883,
+      "loss": 1.1878,
       "step": 4
     },
     {
       "epoch": 0.0411946446961895,
-      "grad_norm": 0.5396899580955505,
+      "grad_norm": 0.518202006816864,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 1.2001,
+      "loss": 1.2005,
       "step": 5
     },
     {
       "epoch": 0.04943357363542739,
-      "grad_norm": 0.5691497921943665,
+      "grad_norm": 0.5391622185707092,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 1.1429,
+      "loss": 1.1447,
       "step": 6
     },
     {
       "epoch": 0.057672502574665295,
-      "grad_norm": 0.4316560626029968,
+      "grad_norm": 0.43155014514923096,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 1.1932,
+      "loss": 1.1931,
       "step": 7
     },
     {
       "epoch": 0.06591143151390319,
-      "grad_norm": 0.38552603125572205,
+      "grad_norm": 0.3879729211330414,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 1.2687,
+      "loss": 1.2691,
       "step": 8
     },
     {
       "epoch": 0.07415036045314109,
-      "grad_norm": 0.3694867491722107,
+      "grad_norm": 0.3610127568244934,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 1.4062,
+      "loss": 1.4065,
       "step": 9
     },
     {
       "epoch": 0.082389289392379,
-      "grad_norm": 0.4373791515827179,
+      "grad_norm": 0.4237038791179657,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 1.569,
+      "loss": 1.5679,
       "step": 10
     },
     {
       "epoch": 0.09062821833161688,
-      "grad_norm": 0.7083751559257507,
+      "grad_norm": 0.6687376499176025,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 1.857,
+      "loss": 1.8561,
       "step": 11
     },
     {
       "epoch": 0.09886714727085479,
-      "grad_norm": 3.521287441253662,
+      "grad_norm": 3.46882963180542,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 2.7373,
+      "loss": 2.7443,
       "step": 12
     },
     {
       "epoch": 0.10710607621009269,
-      "grad_norm": 2.4093360900878906,
+      "grad_norm": 2.3832569122314453,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 1.9829,
+      "loss": 1.9835,
       "step": 13
     },
     {
       "epoch": 0.11534500514933059,
-      "grad_norm": 1.6218130588531494,
+      "grad_norm": 1.5324733257293701,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 1.0176,
+      "loss": 1.0203,
       "step": 14
     },
     {
       "epoch": 0.12358393408856849,
-      "grad_norm": 0.743577778339386,
+      "grad_norm": 0.7405157685279846,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 0.4495,
+      "loss": 0.4499,
       "step": 15
     },
     {
       "epoch": 0.13182286302780638,
-      "grad_norm": 0.3841581344604492,
+      "grad_norm": 0.3928067088127136,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 0.6022,
+      "loss": 0.6033,
       "step": 16
     },
     {
       "epoch": 0.1400617919670443,
-      "grad_norm": 0.40283846855163574,
+      "grad_norm": 0.4208212196826935,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 1.036,
+      "loss": 1.0389,
       "step": 17
     },
     {
       "epoch": 0.14830072090628219,
-      "grad_norm": 0.3529587388038635,
+      "grad_norm": 0.37482503056526184,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 1.0628,
+      "loss": 1.0651,
       "step": 18
     },
     {
       "epoch": 0.15653964984552007,
-      "grad_norm": 0.3646421432495117,
+      "grad_norm": 0.39850133657455444,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 1.0915,
+      "loss": 1.0933,
       "step": 19
     },
     {
       "epoch": 0.164778578784758,
-      "grad_norm": 0.3263953924179077,
+      "grad_norm": 0.3676826059818268,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 1.1286,
+      "loss": 1.1298,
       "step": 20
     },
     {
       "epoch": 0.17301750772399588,
-      "grad_norm": 0.296126127243042,
+      "grad_norm": 0.29273325204849243,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 1.2489,
+      "loss": 1.2501,
       "step": 21
     },
     {
       "epoch": 0.18125643666323377,
-      "grad_norm": 0.3966723084449768,
+      "grad_norm": 0.38638484477996826,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 1.3886,
+      "loss": 1.3895,
       "step": 22
     },
     {
       "epoch": 0.18949536560247168,
-      "grad_norm": 0.4947672486305237,
+      "grad_norm": 0.5108461976051331,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 1.5394,
+      "loss": 1.541,
       "step": 23
     },
     {
       "epoch": 0.19773429454170957,
-      "grad_norm": 1.0129477977752686,
+      "grad_norm": 1.4381290674209595,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 1.7833,
+      "loss": 1.7898,
       "step": 24
     },
     {
       "epoch": 0.2059732234809475,
-      "grad_norm": 4.30376672744751,
+      "grad_norm": 4.901559829711914,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 2.4598,
+      "loss": 2.4978,
       "step": 25
     },
     {
       "epoch": 0.2059732234809475,
-      "eval_loss": 1.0918176174163818,
-      "eval_runtime": 27.4744,
-      "eval_samples_per_second": 29.773,
-      "eval_steps_per_second": 3.749,
+      "eval_loss": 1.097769021987915,
+      "eval_runtime": 55.7383,
+      "eval_samples_per_second": 14.676,
+      "eval_steps_per_second": 1.848,
       "step": 25
-    },
-    {
-      "epoch": 0.21421215242018538,
-      "grad_norm": 0.7323720455169678,
-      "learning_rate": 5e-05,
-      "loss": 1.5897,
-      "step": 26
-    },
-    {
-      "epoch": 0.22245108135942326,
-      "grad_norm": 1.0876057147979736,
-      "learning_rate": 4.6729843538492847e-05,
-      "loss": 0.6764,
-      "step": 27
-    },
-    {
-      "epoch": 0.23069001029866118,
-      "grad_norm": 0.4686364531517029,
-      "learning_rate": 4.347369038899744e-05,
-      "loss": 0.8033,
-      "step": 28
-    },
-    {
-      "epoch": 0.23892893923789907,
-      "grad_norm": 0.22451618313789368,
-      "learning_rate": 4.0245483899193595e-05,
-      "loss": 0.7197,
-      "step": 29
-    },
-    {
-      "epoch": 0.24716786817713698,
-      "grad_norm": 0.20337329804897308,
-      "learning_rate": 3.705904774487396e-05,
-      "loss": 0.9707,
-      "step": 30
-    },
-    {
-      "epoch": 0.2554067971163749,
-      "grad_norm": 0.22543476521968842,
-      "learning_rate": 3.392802673484193e-05,
-      "loss": 1.0043,
-      "step": 31
-    },
-    {
-      "epoch": 0.26364572605561276,
-      "grad_norm": 0.27064037322998047,
-      "learning_rate": 3.086582838174551e-05,
-      "loss": 1.1156,
-      "step": 32
-    },
-    {
-      "epoch": 0.27188465499485065,
-      "grad_norm": 0.32793888449668884,
-      "learning_rate": 2.7885565489049946e-05,
-      "loss": 1.2575,
-      "step": 33
-    },
-    {
-      "epoch": 0.2801235839340886,
-      "grad_norm": 0.49145063757896423,
-      "learning_rate": 2.500000000000001e-05,
-      "loss": 1.3977,
-      "step": 34
-    },
-    {
-      "epoch": 0.2883625128733265,
-      "grad_norm": 0.5390352010726929,
-      "learning_rate": 2.2221488349019903e-05,
-      "loss": 1.4835,
-      "step": 35
-    },
-    {
-      "epoch": 0.29660144181256437,
-      "grad_norm": 0.8575440049171448,
-      "learning_rate": 1.9561928549563968e-05,
-      "loss": 1.6991,
-      "step": 36
-    },
-    {
-      "epoch": 0.30484037075180226,
-      "grad_norm": 1.5018357038497925,
-      "learning_rate": 1.703270924499656e-05,
-      "loss": 1.7517,
-      "step": 37
-    },
-    {
-      "epoch": 0.31307929969104015,
-      "grad_norm": 1.204730749130249,
-      "learning_rate": 1.4644660940672627e-05,
-      "loss": 1.0741,
-      "step": 38
-    },
-    {
-      "epoch": 0.3213182286302781,
-      "grad_norm": 0.9195225834846497,
-      "learning_rate": 1.2408009626051137e-05,
-      "loss": 0.4584,
-      "step": 39
-    },
-    {
-      "epoch": 0.329557157569516,
-      "grad_norm": 0.4034794569015503,
-      "learning_rate": 1.0332332985438248e-05,
-      "loss": 0.5015,
-      "step": 40
-    },
-    {
-      "epoch": 0.33779608650875387,
-      "grad_norm": 0.39492160081863403,
-      "learning_rate": 8.426519384872733e-06,
-      "loss": 0.5094,
-      "step": 41
-    },
-    {
-      "epoch": 0.34603501544799176,
-      "grad_norm": 0.17655304074287415,
-      "learning_rate": 6.698729810778065e-06,
-      "loss": 0.9547,
-      "step": 42
-    },
-    {
-      "epoch": 0.35427394438722964,
-      "grad_norm": 0.17320753633975983,
-      "learning_rate": 5.156362923365588e-06,
-      "loss": 1.0494,
-      "step": 43
-    },
-    {
-      "epoch": 0.36251287332646753,
-      "grad_norm": 0.19076572358608246,
-      "learning_rate": 3.8060233744356633e-06,
-      "loss": 1.0784,
-      "step": 44
-    },
-    {
-      "epoch": 0.3707518022657055,
-      "grad_norm": 0.17125385999679565,
-      "learning_rate": 2.653493525244721e-06,
-      "loss": 1.0381,
-      "step": 45
-    },
-    {
-      "epoch": 0.37899073120494337,
-      "grad_norm": 0.22827577590942383,
-      "learning_rate": 1.70370868554659e-06,
-      "loss": 1.245,
-      "step": 46
-    },
-    {
-      "epoch": 0.38722966014418125,
-      "grad_norm": 0.2971423268318176,
-      "learning_rate": 9.607359798384785e-07,
-      "loss": 1.3483,
-      "step": 47
-    },
-    {
-      "epoch": 0.39546858908341914,
-      "grad_norm": 0.5259574055671692,
-      "learning_rate": 4.277569313094809e-07,
-      "loss": 1.6239,
-      "step": 48
-    },
-    {
-      "epoch": 0.40370751802265703,
-      "grad_norm": 0.8611096739768982,
-      "learning_rate": 1.0705383806982606e-07,
-      "loss": 1.7759,
-      "step": 49
-    },
-    {
-      "epoch": 0.411946446961895,
-      "grad_norm": 3.794422149658203,
-      "learning_rate": 0.0,
-      "loss": 1.5341,
-      "step": 50
-    },
-    {
-      "epoch": 0.411946446961895,
-      "eval_loss": 0.986032247543335,
-      "eval_runtime": 27.9322,
-      "eval_samples_per_second": 29.285,
-      "eval_steps_per_second": 3.688,
-      "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": 6.04008059433386e+17,
+  "total_flos": 3.0358346152699494e+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 sha256:1d297b5b1837367d13537f6cf72d45632d1103d798c9e93f31cc4a94f4615aa2
+oid sha256:2dfb0d4f32f51c493cac3b51f47102d63cd08e8ab6a489fabc604b74664bdefe
 size 6776