Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "up_proj",
     "o_proj",
+    "up_proj",
     "k_proj",
-    "gate_proj",
-    "v_proj",
+    "q_proj",
     "down_proj",
-    "q_proj"
+    "v_proj",
+    "gate_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 sha256:3ab8e1026b6d9b19a38ede103038c39b0bf2de5340598a3e4f8d33a92be0660f
+oid sha256:ff289ede209eda3ee47df8b03b3be63597ee1f3709bdd8675eead684d7fd73ee
 size 70430032

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a4a35e84099827f3c89ce8ce7a069cab440163a46df8ba580862650d31f6a58
+oid sha256:80cfdcee4b3751ceedaff274027ae4a58d594b59d989ed430311a9d6d2b3087d
 size 141053442

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 3.2202842235565186,
+  "best_metric": 3.2185161113739014,
   "best_model_checkpoint": "miner_id_24/checkpoint-50",
   "epoch": 0.1983635011157947,
   "eval_steps": 25,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.003967270022315894,
-      "grad_norm": 0.8100353479385376,
+      "grad_norm": 0.8241645097732544,
       "learning_rate": 5e-05,
       "loss": 3.4866,
       "step": 1
@@ -18,368 +18,368 @@
     {
       "epoch": 0.003967270022315894,
       "eval_loss": 3.7426881790161133,
-      "eval_runtime": 11.5565,
-      "eval_samples_per_second": 147.016,
-      "eval_steps_per_second": 18.431,
+      "eval_runtime": 11.7309,
+      "eval_samples_per_second": 144.831,
+      "eval_steps_per_second": 18.157,
       "step": 1
     },
     {
       "epoch": 0.007934540044631788,
-      "grad_norm": 0.8411345481872559,
+      "grad_norm": 0.8499454855918884,
       "learning_rate": 0.0001,
       "loss": 3.4657,
       "step": 2
     },
     {
       "epoch": 0.011901810066947682,
-      "grad_norm": 0.9052443504333496,
+      "grad_norm": 0.9072771072387695,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 3.5402,
+      "loss": 3.5394,
       "step": 3
     },
     {
       "epoch": 0.015869080089263576,
-      "grad_norm": 0.7589433789253235,
+      "grad_norm": 0.7632107138633728,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 3.4572,
+      "loss": 3.4564,
       "step": 4
     },
     {
       "epoch": 0.01983635011157947,
-      "grad_norm": 0.8587386608123779,
+      "grad_norm": 0.8594216704368591,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 3.5274,
+      "loss": 3.5266,
       "step": 5
     },
     {
       "epoch": 0.023803620133895365,
-      "grad_norm": 0.9426459074020386,
+      "grad_norm": 0.9645074009895325,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 3.5125,
+      "loss": 3.512,
       "step": 6
     },
     {
       "epoch": 0.02777089015621126,
-      "grad_norm": 0.9467688798904419,
+      "grad_norm": 0.9713271260261536,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 3.5663,
+      "loss": 3.5646,
       "step": 7
     },
     {
       "epoch": 0.03173816017852715,
-      "grad_norm": 1.0104401111602783,
+      "grad_norm": 1.0277541875839233,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 3.4531,
+      "loss": 3.4512,
       "step": 8
     },
     {
       "epoch": 0.03570543020084305,
-      "grad_norm": 1.0469218492507935,
+      "grad_norm": 1.0516093969345093,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 3.4803,
+      "loss": 3.4789,
       "step": 9
     },
     {
       "epoch": 0.03967270022315894,
-      "grad_norm": 1.096985936164856,
+      "grad_norm": 1.1022090911865234,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 3.5365,
+      "loss": 3.5343,
       "step": 10
     },
     {
       "epoch": 0.043639970245474835,
-      "grad_norm": 1.0697236061096191,
+      "grad_norm": 1.0738914012908936,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 3.5104,
+      "loss": 3.5085,
       "step": 11
     },
     {
       "epoch": 0.04760724026779073,
-      "grad_norm": 1.215027928352356,
+      "grad_norm": 1.1916122436523438,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 3.472,
+      "loss": 3.4677,
       "step": 12
     },
     {
       "epoch": 0.05157451029010662,
-      "grad_norm": 1.1256529092788696,
+      "grad_norm": 1.1183475255966187,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 3.4123,
+      "loss": 3.4107,
       "step": 13
     },
     {
       "epoch": 0.05554178031242252,
-      "grad_norm": 1.0637106895446777,
+      "grad_norm": 1.0498988628387451,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 3.3022,
+      "loss": 3.3009,
       "step": 14
     },
     {
       "epoch": 0.05950905033473841,
-      "grad_norm": 1.1305065155029297,
+      "grad_norm": 1.125704288482666,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 3.3346,
+      "loss": 3.332,
       "step": 15
     },
     {
       "epoch": 0.0634763203570543,
-      "grad_norm": 0.8763689994812012,
+      "grad_norm": 0.8547013998031616,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 3.2566,
+      "loss": 3.2555,
       "step": 16
     },
     {
       "epoch": 0.0674435903793702,
-      "grad_norm": 0.7871897220611572,
+      "grad_norm": 0.7742959260940552,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 3.3037,
+      "loss": 3.3036,
       "step": 17
     },
     {
       "epoch": 0.0714108604016861,
-      "grad_norm": 0.8707305192947388,
+      "grad_norm": 0.8887235522270203,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 3.3393,
+      "loss": 3.3385,
       "step": 18
     },
     {
       "epoch": 0.07537813042400199,
-      "grad_norm": 0.8883684873580933,
+      "grad_norm": 0.8906424641609192,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 3.2778,
+      "loss": 3.2767,
       "step": 19
     },
     {
       "epoch": 0.07934540044631788,
-      "grad_norm": 1.0402384996414185,
+      "grad_norm": 1.0381428003311157,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 3.2926,
+      "loss": 3.2897,
       "step": 20
     },
     {
       "epoch": 0.08331267046863378,
-      "grad_norm": 1.2061767578125,
+      "grad_norm": 1.2209287881851196,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 3.4678,
+      "loss": 3.4663,
       "step": 21
     },
     {
       "epoch": 0.08727994049094967,
-      "grad_norm": 1.1715914011001587,
+      "grad_norm": 1.1723295450210571,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 3.3448,
+      "loss": 3.3431,
       "step": 22
     },
     {
       "epoch": 0.09124721051326556,
-      "grad_norm": 1.2525004148483276,
+      "grad_norm": 1.0842334032058716,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 3.3168,
+      "loss": 3.314,
       "step": 23
     },
     {
       "epoch": 0.09521448053558146,
-      "grad_norm": 1.1802704334259033,
+      "grad_norm": 1.0503602027893066,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 3.2211,
+      "loss": 3.2195,
       "step": 24
     },
     {
       "epoch": 0.09918175055789735,
-      "grad_norm": 1.308222770690918,
+      "grad_norm": 1.2791979312896729,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 3.2855,
+      "loss": 3.2802,
       "step": 25
     },
     {
       "epoch": 0.09918175055789735,
-      "eval_loss": 3.268176555633545,
-      "eval_runtime": 11.5559,
-      "eval_samples_per_second": 147.024,
-      "eval_steps_per_second": 18.432,
+      "eval_loss": 3.267420530319214,
+      "eval_runtime": 11.7144,
+      "eval_samples_per_second": 145.035,
+      "eval_steps_per_second": 18.183,
       "step": 25
     },
     {
       "epoch": 0.10314902058021325,
-      "grad_norm": 0.9733558893203735,
+      "grad_norm": 0.9879074096679688,
       "learning_rate": 5e-05,
-      "loss": 3.2624,
+      "loss": 3.263,
       "step": 26
     },
     {
       "epoch": 0.10711629060252914,
-      "grad_norm": 1.166012167930603,
+      "grad_norm": 1.1669400930404663,
       "learning_rate": 4.6729843538492847e-05,
-      "loss": 3.1748,
+      "loss": 3.1752,
       "step": 27
     },
     {
       "epoch": 0.11108356062484503,
-      "grad_norm": 1.3410450220108032,
+      "grad_norm": 1.3233245611190796,
       "learning_rate": 4.347369038899744e-05,
-      "loss": 3.2613,
+      "loss": 3.2607,
       "step": 28
     },
     {
       "epoch": 0.11505083064716093,
-      "grad_norm": 1.3012003898620605,
+      "grad_norm": 1.279555082321167,
       "learning_rate": 4.0245483899193595e-05,
-      "loss": 3.2793,
+      "loss": 3.2804,
       "step": 29
     },
     {
       "epoch": 0.11901810066947682,
-      "grad_norm": 1.0508790016174316,
+      "grad_norm": 1.0318185091018677,
       "learning_rate": 3.705904774487396e-05,
-      "loss": 3.2646,
+      "loss": 3.2639,
       "step": 30
     },
     {
       "epoch": 0.12298537069179272,
-      "grad_norm": 0.9772084355354309,
+      "grad_norm": 0.9323838949203491,
       "learning_rate": 3.392802673484193e-05,
-      "loss": 3.2421,
+      "loss": 3.2406,
       "step": 31
     },
     {
       "epoch": 0.1269526407141086,
-      "grad_norm": 0.8641189932823181,
+      "grad_norm": 0.8470538854598999,
       "learning_rate": 3.086582838174551e-05,
-      "loss": 3.2449,
+      "loss": 3.2429,
       "step": 32
     },
     {
       "epoch": 0.1309199107364245,
-      "grad_norm": 0.8817277550697327,
+      "grad_norm": 0.8595752120018005,
       "learning_rate": 2.7885565489049946e-05,
-      "loss": 3.3813,
+      "loss": 3.3786,
       "step": 33
     },
     {
       "epoch": 0.1348871807587404,
-      "grad_norm": 0.8557921648025513,
+      "grad_norm": 0.8525285124778748,
       "learning_rate": 2.500000000000001e-05,
-      "loss": 3.2187,
+      "loss": 3.2173,
       "step": 34
     },
     {
       "epoch": 0.1388544507810563,
-      "grad_norm": 0.9797968864440918,
+      "grad_norm": 0.9648301005363464,
       "learning_rate": 2.2221488349019903e-05,
-      "loss": 3.2734,
+      "loss": 3.2725,
       "step": 35
     },
     {
       "epoch": 0.1428217208033722,
-      "grad_norm": 0.9896147847175598,
+      "grad_norm": 0.9749354124069214,
       "learning_rate": 1.9561928549563968e-05,
-      "loss": 3.2438,
+      "loss": 3.2412,
       "step": 36
     },
     {
       "epoch": 0.14678899082568808,
-      "grad_norm": 1.0835036039352417,
+      "grad_norm": 1.177864909172058,
       "learning_rate": 1.703270924499656e-05,
-      "loss": 3.3002,
+      "loss": 3.3005,
       "step": 37
     },
     {
       "epoch": 0.15075626084800398,
-      "grad_norm": 0.842976987361908,
+      "grad_norm": 0.8285359740257263,
       "learning_rate": 1.4644660940672627e-05,
-      "loss": 3.2495,
+      "loss": 3.2479,
       "step": 38
     },
     {
       "epoch": 0.15472353087031987,
-      "grad_norm": 0.6773706078529358,
+      "grad_norm": 0.6497054100036621,
       "learning_rate": 1.2408009626051137e-05,
-      "loss": 3.2503,
+      "loss": 3.2486,
       "step": 39
     },
     {
       "epoch": 0.15869080089263576,
-      "grad_norm": 0.5887467861175537,
+      "grad_norm": 0.6034384965896606,
       "learning_rate": 1.0332332985438248e-05,
-      "loss": 3.2258,
+      "loss": 3.2252,
       "step": 40
     },
     {
       "epoch": 0.16265807091495166,
-      "grad_norm": 0.5568850040435791,
+      "grad_norm": 0.5653945207595825,
       "learning_rate": 8.426519384872733e-06,
-      "loss": 3.1648,
+      "loss": 3.1646,
       "step": 41
     },
     {
       "epoch": 0.16662534093726755,
-      "grad_norm": 0.5692043900489807,
+      "grad_norm": 0.5705766677856445,
       "learning_rate": 6.698729810778065e-06,
-      "loss": 3.2207,
+      "loss": 3.2201,
       "step": 42
     },
     {
       "epoch": 0.17059261095958345,
-      "grad_norm": 0.6383813619613647,
+      "grad_norm": 0.6292085647583008,
       "learning_rate": 5.156362923365588e-06,
-      "loss": 3.2575,
+      "loss": 3.2563,
       "step": 43
     },
     {
       "epoch": 0.17455988098189934,
-      "grad_norm": 0.6307452917098999,
+      "grad_norm": 0.6345009803771973,
       "learning_rate": 3.8060233744356633e-06,
-      "loss": 3.2423,
+      "loss": 3.2398,
       "step": 44
     },
     {
       "epoch": 0.17852715100421523,
-      "grad_norm": 0.7243000268936157,
+      "grad_norm": 0.7186870574951172,
       "learning_rate": 2.653493525244721e-06,
-      "loss": 3.2007,
+      "loss": 3.1997,
       "step": 45
     },
     {
       "epoch": 0.18249442102653113,
-      "grad_norm": 0.846513032913208,
+      "grad_norm": 0.8233199715614319,
       "learning_rate": 1.70370868554659e-06,
-      "loss": 3.2938,
+      "loss": 3.2917,
       "step": 46
     },
     {
       "epoch": 0.18646169104884702,
-      "grad_norm": 0.951248824596405,
+      "grad_norm": 0.9457505345344543,
       "learning_rate": 9.607359798384785e-07,
-      "loss": 3.3001,
+      "loss": 3.297,
       "step": 47
     },
     {
       "epoch": 0.19042896107116292,
-      "grad_norm": 0.9839473962783813,
+      "grad_norm": 0.9537065029144287,
       "learning_rate": 4.277569313094809e-07,
-      "loss": 3.1855,
+      "loss": 3.1839,
       "step": 48
     },
     {
       "epoch": 0.1943962310934788,
-      "grad_norm": 1.0755198001861572,
+      "grad_norm": 1.0764613151550293,
       "learning_rate": 1.0705383806982606e-07,
-      "loss": 3.1702,
+      "loss": 3.1668,
       "step": 49
     },
     {
       "epoch": 0.1983635011157947,
-      "grad_norm": 1.5319169759750366,
+      "grad_norm": 1.5574710369110107,
       "learning_rate": 0.0,
-      "loss": 3.2793,
+      "loss": 3.2757,
       "step": 50
     },
     {
       "epoch": 0.1983635011157947,
-      "eval_loss": 3.2202842235565186,
-      "eval_runtime": 11.5549,
-      "eval_samples_per_second": 147.037,
-      "eval_steps_per_second": 18.434,
+      "eval_loss": 3.2185161113739014,
+      "eval_runtime": 11.7074,
+      "eval_samples_per_second": 145.122,
+      "eval_steps_per_second": 18.194,
       "step": 50
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:870dfab1563a1ea9b9c6847f153d550987da1ab2c43fc66f0f15ea702901d060
+oid sha256:41423001aa5ceb2d23235d9dee62fad89e061cbc43f5e3a09904fb106ff3b870
 size 6776