Training in progress, step 25, checkpoint

last-checkpoint/adapter_config.json

@@ -22,11 +22,11 @@
   "target_modules": [
     "v_proj",
     "o_proj",
-    "down_proj",
     "k_proj",
-    "up_proj",
+    "gate_proj",
+    "down_proj",
     "q_proj",
-    "gate_proj"
+    "up_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 sha256:3be8c835d90ef6a0e5be353b98569d168ca6ca36f1a29ffce5a22464f044852d
+oid sha256:65405ae444df32afd0c2438b6ee731fab60b95b56c0219baefc70f8c4a40de03
 size 156926880

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bfaaaf38d45bae3d70de866ff183e3a4359c58acf627b11d07f24845a7e427f1
+oid sha256:4c3413eea9b89ef4a5e656c43cefa67fb42c62f8b0041cd7a0ac070009e1101b
 size 313998650

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:97cec626985927ebf5f926ef8db14111273baf09d11734b9764f6b920bb7ded1
+oid sha256:779e97e8ea7cb127409cbb724efbebd167605f0a190e1036b5903a084a92e655
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:21b8ebd07f99f4b9febf514224d316c5373a34ff4ad237c86f42e9212750f04a
+oid sha256:f3a53dddb2fb642bdb032ebaa463b6b4e474bec37af360b04fdb99dc58ec2a2a
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f3ab4bc2ab73c53f960cd00d8fa71e61ca98b331cd1d1d9fd5ca4ada15400168
+oid sha256:afd59f255cf47b03bae4462289eb19484151becbe4554c41e06e11f5a4bb1002
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4186250a1aca3322d717a26f3a6e78fedf802a8b67a62364cc6fb604a2aea6cc
+oid sha256:6a52f9ad0a74ffe421a11e307ffb467999cdc9fbeb2982b45bd492bbc12fa9f5
 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

@@ -1,16 +1,16 @@
 {
-  "best_metric": 1.5265823602676392,
-  "best_model_checkpoint": "miner_id_24/checkpoint-50",
-  "epoch": 0.2450229709035222,
+  "best_metric": 1.6192727088928223,
+  "best_model_checkpoint": "miner_id_24/checkpoint-25",
+  "epoch": 0.1225114854517611,
   "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": 0.004900459418070444,
-      "grad_norm": 23.471738815307617,
+      "grad_norm": 23.882152557373047,
       "learning_rate": 5e-05,
       "loss": 6.7852,
       "step": 1
@@ -18,369 +18,186 @@
     {
       "epoch": 0.004900459418070444,
       "eval_loss": 7.382022857666016,
-      "eval_runtime": 18.5323,
-      "eval_samples_per_second": 74.195,
-      "eval_steps_per_second": 9.281,
+      "eval_runtime": 18.4681,
+      "eval_samples_per_second": 74.453,
+      "eval_steps_per_second": 9.313,
       "step": 1
     },
     {
       "epoch": 0.009800918836140888,
-      "grad_norm": 23.077556610107422,
+      "grad_norm": 23.43764877319336,
       "learning_rate": 0.0001,
       "loss": 6.8171,
       "step": 2
     },
     {
       "epoch": 0.014701378254211332,
-      "grad_norm": 17.30784034729004,
+      "grad_norm": 17.595151901245117,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 5.2819,
+      "loss": 5.2748,
       "step": 3
     },
     {
       "epoch": 0.019601837672281776,
-      "grad_norm": 10.578380584716797,
+      "grad_norm": 10.711825370788574,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 3.4992,
+      "loss": 3.4875,
       "step": 4
     },
     {
       "epoch": 0.02450229709035222,
-      "grad_norm": 8.487051010131836,
+      "grad_norm": 8.532750129699707,
       "learning_rate": 9.903926402016153e-05,
-      "loss": 2.9079,
+      "loss": 2.892,
       "step": 5
     },
     {
       "epoch": 0.029402756508422664,
-      "grad_norm": 6.275716781616211,
+      "grad_norm": 6.257192134857178,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 2.6387,
+      "loss": 2.6199,
       "step": 6
     },
     {
       "epoch": 0.03430321592649311,
-      "grad_norm": 4.362668037414551,
+      "grad_norm": 4.383069038391113,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 2.48,
+      "loss": 2.4587,
       "step": 7
     },
     {
       "epoch": 0.03920367534456355,
-      "grad_norm": 3.9618260860443115,
+      "grad_norm": 4.003562927246094,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 2.3593,
+      "loss": 2.3455,
       "step": 8
     },
     {
       "epoch": 0.044104134762633994,
-      "grad_norm": 3.2072575092315674,
+      "grad_norm": 3.2436230182647705,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 2.3954,
+      "loss": 2.3851,
       "step": 9
     },
     {
       "epoch": 0.04900459418070444,
-      "grad_norm": 3.726250171661377,
+      "grad_norm": 3.570329427719116,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 2.1785,
+      "loss": 2.1683,
       "step": 10
     },
     {
       "epoch": 0.053905053598774885,
-      "grad_norm": 2.532313585281372,
+      "grad_norm": 2.575678586959839,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 2.2436,
+      "loss": 2.2367,
       "step": 11
     },
     {
       "epoch": 0.05880551301684533,
-      "grad_norm": 2.7389440536499023,
+      "grad_norm": 2.7599802017211914,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 2.1768,
+      "loss": 2.169,
       "step": 12
     },
     {
       "epoch": 0.06370597243491577,
-      "grad_norm": 2.995002508163452,
+      "grad_norm": 2.853652000427246,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 1.9088,
+      "loss": 1.8975,
       "step": 13
     },
     {
       "epoch": 0.06860643185298622,
-      "grad_norm": 2.0172295570373535,
+      "grad_norm": 2.076232671737671,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 1.4442,
+      "loss": 1.4363,
       "step": 14
     },
     {
       "epoch": 0.07350689127105667,
-      "grad_norm": 2.4108283519744873,
+      "grad_norm": 2.3847100734710693,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 1.5092,
+      "loss": 1.4934,
       "step": 15
     },
     {
       "epoch": 0.0784073506891271,
-      "grad_norm": 2.739496946334839,
+      "grad_norm": 2.6652495861053467,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 1.4547,
+      "loss": 1.4453,
       "step": 16
     },
     {
       "epoch": 0.08330781010719755,
-      "grad_norm": 3.4123828411102295,
+      "grad_norm": 3.1245956420898438,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 1.4325,
+      "loss": 1.4152,
       "step": 17
     },
     {
       "epoch": 0.08820826952526799,
-      "grad_norm": 2.2385709285736084,
+      "grad_norm": 2.274653673171997,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 1.6656,
+      "loss": 1.6554,
       "step": 18
     },
     {
       "epoch": 0.09310872894333844,
-      "grad_norm": 2.2674455642700195,
+      "grad_norm": 2.281479597091675,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 1.6064,
+      "loss": 1.5986,
       "step": 19
     },
     {
       "epoch": 0.09800918836140889,
-      "grad_norm": 2.166579484939575,
+      "grad_norm": 2.2254714965820312,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 1.707,
+      "loss": 1.7064,
       "step": 20
     },
     {
       "epoch": 0.10290964777947932,
-      "grad_norm": 2.1645350456237793,
+      "grad_norm": 2.2070491313934326,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 1.7495,
+      "loss": 1.745,
       "step": 21
     },
     {
       "epoch": 0.10781010719754977,
-      "grad_norm": 2.4724934101104736,
+      "grad_norm": 2.4654288291931152,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 2.0361,
+      "loss": 2.0296,
       "step": 22
     },
     {
       "epoch": 0.11271056661562022,
-      "grad_norm": 1.933280110359192,
+      "grad_norm": 1.9440032243728638,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 2.0074,
+      "loss": 2.0025,
       "step": 23
     },
     {
       "epoch": 0.11761102603369065,
-      "grad_norm": 2.0323431491851807,
+      "grad_norm": 2.069082021713257,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 2.0281,
+      "loss": 2.0276,
       "step": 24
     },
     {
       "epoch": 0.1225114854517611,
-      "grad_norm": 2.1172215938568115,
+      "grad_norm": 2.1394474506378174,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 2.1502,
+      "loss": 2.143,
       "step": 25
     },
     {
       "epoch": 0.1225114854517611,
-      "eval_loss": 1.6246213912963867,
-      "eval_runtime": 18.4789,
-      "eval_samples_per_second": 74.409,
-      "eval_steps_per_second": 9.308,
+      "eval_loss": 1.6192727088928223,
+      "eval_runtime": 18.4903,
+      "eval_samples_per_second": 74.363,
+      "eval_steps_per_second": 9.302,
       "step": 25
-    },
-    {
-      "epoch": 0.12741194486983154,
-      "grad_norm": 1.912805199623108,
-      "learning_rate": 5e-05,
-      "loss": 1.3076,
-      "step": 26
-    },
-    {
-      "epoch": 0.132312404287902,
-      "grad_norm": 1.7271080017089844,
-      "learning_rate": 4.6729843538492847e-05,
-      "loss": 1.5167,
-      "step": 27
-    },
-    {
-      "epoch": 0.13721286370597244,
-      "grad_norm": 1.879220724105835,
-      "learning_rate": 4.347369038899744e-05,
-      "loss": 1.1116,
-      "step": 28
-    },
-    {
-      "epoch": 0.14211332312404287,
-      "grad_norm": 1.5653444528579712,
-      "learning_rate": 4.0245483899193595e-05,
-      "loss": 1.0531,
-      "step": 29
-    },
-    {
-      "epoch": 0.14701378254211334,
-      "grad_norm": 1.661484718322754,
-      "learning_rate": 3.705904774487396e-05,
-      "loss": 1.345,
-      "step": 30
-    },
-    {
-      "epoch": 0.15191424196018377,
-      "grad_norm": 1.8263542652130127,
-      "learning_rate": 3.392802673484193e-05,
-      "loss": 1.4316,
-      "step": 31
-    },
-    {
-      "epoch": 0.1568147013782542,
-      "grad_norm": 2.0890257358551025,
-      "learning_rate": 3.086582838174551e-05,
-      "loss": 1.6723,
-      "step": 32
-    },
-    {
-      "epoch": 0.16171516079632464,
-      "grad_norm": 2.0878026485443115,
-      "learning_rate": 2.7885565489049946e-05,
-      "loss": 1.8243,
-      "step": 33
-    },
-    {
-      "epoch": 0.1666156202143951,
-      "grad_norm": 2.1519672870635986,
-      "learning_rate": 2.500000000000001e-05,
-      "loss": 1.8026,
-      "step": 34
-    },
-    {
-      "epoch": 0.17151607963246554,
-      "grad_norm": 2.1100552082061768,
-      "learning_rate": 2.2221488349019903e-05,
-      "loss": 1.6778,
-      "step": 35
-    },
-    {
-      "epoch": 0.17641653905053598,
-      "grad_norm": 2.2416326999664307,
-      "learning_rate": 1.9561928549563968e-05,
-      "loss": 1.9641,
-      "step": 36
-    },
-    {
-      "epoch": 0.18131699846860644,
-      "grad_norm": 2.2929885387420654,
-      "learning_rate": 1.703270924499656e-05,
-      "loss": 1.9768,
-      "step": 37
-    },
-    {
-      "epoch": 0.18621745788667687,
-      "grad_norm": 1.919622778892517,
-      "learning_rate": 1.4644660940672627e-05,
-      "loss": 1.6375,
-      "step": 38
-    },
-    {
-      "epoch": 0.1911179173047473,
-      "grad_norm": 1.6996771097183228,
-      "learning_rate": 1.2408009626051137e-05,
-      "loss": 1.3368,
-      "step": 39
-    },
-    {
-      "epoch": 0.19601837672281777,
-      "grad_norm": 1.4825369119644165,
-      "learning_rate": 1.0332332985438248e-05,
-      "loss": 1.0644,
-      "step": 40
-    },
-    {
-      "epoch": 0.2009188361408882,
-      "grad_norm": 1.6856578588485718,
-      "learning_rate": 8.426519384872733e-06,
-      "loss": 1.179,
-      "step": 41
-    },
-    {
-      "epoch": 0.20581929555895864,
-      "grad_norm": 1.5633989572525024,
-      "learning_rate": 6.698729810778065e-06,
-      "loss": 1.188,
-      "step": 42
-    },
-    {
-      "epoch": 0.2107197549770291,
-      "grad_norm": 1.8037993907928467,
-      "learning_rate": 5.156362923365588e-06,
-      "loss": 1.2002,
-      "step": 43
-    },
-    {
-      "epoch": 0.21562021439509954,
-      "grad_norm": 1.8860701322555542,
-      "learning_rate": 3.8060233744356633e-06,
-      "loss": 1.4013,
-      "step": 44
-    },
-    {
-      "epoch": 0.22052067381316998,
-      "grad_norm": 1.9994962215423584,
-      "learning_rate": 2.653493525244721e-06,
-      "loss": 1.532,
-      "step": 45
-    },
-    {
-      "epoch": 0.22542113323124044,
-      "grad_norm": 2.0619125366210938,
-      "learning_rate": 1.70370868554659e-06,
-      "loss": 1.5686,
-      "step": 46
-    },
-    {
-      "epoch": 0.23032159264931087,
-      "grad_norm": 1.921753168106079,
-      "learning_rate": 9.607359798384785e-07,
-      "loss": 1.712,
-      "step": 47
-    },
-    {
-      "epoch": 0.2352220520673813,
-      "grad_norm": 2.2192230224609375,
-      "learning_rate": 4.277569313094809e-07,
-      "loss": 1.7398,
-      "step": 48
-    },
-    {
-      "epoch": 0.24012251148545177,
-      "grad_norm": 1.9620671272277832,
-      "learning_rate": 1.0705383806982606e-07,
-      "loss": 1.9291,
-      "step": 49
-    },
-    {
-      "epoch": 0.2450229709035222,
-      "grad_norm": 2.0268142223358154,
-      "learning_rate": 0.0,
-      "loss": 1.8767,
-      "step": 50
-    },
-    {
-      "epoch": 0.2450229709035222,
-      "eval_loss": 1.5265823602676392,
-      "eval_runtime": 18.5545,
-      "eval_samples_per_second": 74.106,
-      "eval_steps_per_second": 9.27,
-      "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": 1.5904571953617306e+17,
+  "total_flos": 7.957253052799386e+16,
   "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:bc68cb827569788bb1f117b8f7f7c72a5cd15a42f06b5bf44da40189cb6ceefa
+oid sha256:ca7570c1a84ec9fd90550b9cc88334bb96c7e7a9f33b9061b066e49eb42d8735
 size 6776