Training in progress, step 50, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "up_proj",
-    "q_proj",
+    "k_proj",
     "v_proj",
+    "q_proj",
     "gate_proj",
-    "k_proj",
-    "down_proj"
+    "up_proj",
+    "down_proj",
+    "o_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 sha256:0c40cb370fc2ce37f6f547ee5bcf78a20052b793ca315c4480c5e3d23ebc7ac5
+oid sha256:41d7e231427887b895fb766c3b49b50cb85cc8e90e4d3f1054722665314eaddf
 size 319876032

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b1f81f3d6252dc6a155ed015f1175765178bf46e423eabe4fdfcf2ecf11b8ceb
+oid sha256:d02cc0a50d9734c5b6eea8fe8ea5b2afe4049f6921136edd8069c95d886d5511
 size 640009682

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fde16eea6e7a77484d531f0284da65a2e0c32520ab376fadfdf6bdb6044ef5dd
+oid sha256:2bed50a18dc514f557238b0fec4e86e21b12fd8e46f37856fdc365ea74e9e033
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:739b2d91799fe74cd6d3c730a0720ac64ff56a4311ade1aa1fad00175a42c5c0
+oid sha256:c9347fc83d27f60d82d319315ca5be66998d01d098ef78b2b83f111bb9b0169b
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54a0dce6ed6cc1a73936cf80e3d8c2a80f6c7a1c669a1b0befa8d7d07e596dda
+oid sha256:75576fae3b13b59a6a322f3e77c55ae5ea39e946939481747fb8251f1c0d6803
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50e5dece309d0144af69ddb9d5ef7c77b64f868b71b5395b07bdbd1dea95f0c2
+oid sha256:0100b8b3f988766216eddb62ba8b5711798bb5cf2af3e82fca13fe8a3559eb29
 size 15024

last-checkpoint/scheduler.pt

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

last-checkpoint/trainer_state.json

@@ -1,16 +1,16 @@
 {
-  "best_metric": 1.1423113346099854,
-  "best_model_checkpoint": "miner_id_24/checkpoint-25",
-  "epoch": 0.05272176090681429,
+  "best_metric": 1.0235497951507568,
+  "best_model_checkpoint": "miner_id_24/checkpoint-50",
+  "epoch": 0.10544352181362858,
   "eval_steps": 25,
-  "global_step": 25,
+  "global_step": 50,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.0021088704362725716,
-      "grad_norm": 1.0445363521575928,
+      "grad_norm": 0.9816077351570129,
       "learning_rate": 5e-05,
       "loss": 2.0005,
       "step": 1
@@ -18,186 +18,369 @@
     {
       "epoch": 0.0021088704362725716,
       "eval_loss": 2.172279119491577,
-      "eval_runtime": 91.0927,
-      "eval_samples_per_second": 35.074,
-      "eval_steps_per_second": 4.391,
+      "eval_runtime": 91.0195,
+      "eval_samples_per_second": 35.102,
+      "eval_steps_per_second": 4.395,
       "step": 1
     },
     {
       "epoch": 0.004217740872545143,
-      "grad_norm": 1.0849013328552246,
+      "grad_norm": 1.0070489645004272,
       "learning_rate": 0.0001,
       "loss": 2.0592,
       "step": 2
     },
     {
       "epoch": 0.006326611308817715,
-      "grad_norm": 1.02006995677948,
+      "grad_norm": 0.957038938999176,
       "learning_rate": 9.989294616193017e-05,
-      "loss": 2.047,
+      "loss": 2.0489,
       "step": 3
     },
     {
       "epoch": 0.008435481745090286,
-      "grad_norm": 0.9143255949020386,
+      "grad_norm": 0.8758676648139954,
       "learning_rate": 9.957224306869053e-05,
-      "loss": 1.8777,
+      "loss": 1.8789,
       "step": 4
     },
     {
       "epoch": 0.010544352181362857,
-      "grad_norm": 0.7838174104690552,
+      "grad_norm": 0.7380896806716919,
       "learning_rate": 9.903926402016153e-05,
       "loss": 1.7968,
       "step": 5
     },
     {
       "epoch": 0.01265322261763543,
-      "grad_norm": 0.9264847040176392,
+      "grad_norm": 0.8705281019210815,
       "learning_rate": 9.829629131445342e-05,
-      "loss": 1.7519,
+      "loss": 1.7523,
       "step": 6
     },
     {
       "epoch": 0.014762093053908,
-      "grad_norm": 0.7748655676841736,
+      "grad_norm": 0.7527739405632019,
       "learning_rate": 9.73465064747553e-05,
-      "loss": 1.6999,
+      "loss": 1.7027,
       "step": 7
     },
     {
       "epoch": 0.016870963490180572,
-      "grad_norm": 0.752051055431366,
+      "grad_norm": 0.7234334349632263,
       "learning_rate": 9.619397662556435e-05,
-      "loss": 1.6492,
+      "loss": 1.6509,
       "step": 8
     },
     {
       "epoch": 0.018979833926453145,
-      "grad_norm": 0.6525676846504211,
+      "grad_norm": 0.6203216910362244,
       "learning_rate": 9.484363707663442e-05,
-      "loss": 1.5985,
+      "loss": 1.5996,
       "step": 9
     },
     {
       "epoch": 0.021088704362725714,
-      "grad_norm": 0.6811321377754211,
+      "grad_norm": 0.646403968334198,
       "learning_rate": 9.330127018922194e-05,
-      "loss": 1.5414,
+      "loss": 1.5424,
       "step": 10
     },
     {
       "epoch": 0.023197574798998286,
-      "grad_norm": 0.6379256248474121,
+      "grad_norm": 0.6088539361953735,
       "learning_rate": 9.157348061512727e-05,
-      "loss": 1.4974,
+      "loss": 1.4988,
       "step": 11
     },
     {
       "epoch": 0.02530644523527086,
-      "grad_norm": 0.6451159715652466,
+      "grad_norm": 0.6200858354568481,
       "learning_rate": 8.966766701456177e-05,
-      "loss": 1.4877,
+      "loss": 1.4886,
       "step": 12
     },
     {
       "epoch": 0.02741531567154343,
-      "grad_norm": 0.5045467615127563,
+      "grad_norm": 0.48125016689300537,
       "learning_rate": 8.759199037394887e-05,
-      "loss": 1.3994,
+      "loss": 1.4003,
       "step": 13
     },
     {
       "epoch": 0.029524186107816,
-      "grad_norm": 0.4902452528476715,
+      "grad_norm": 0.46856382489204407,
       "learning_rate": 8.535533905932738e-05,
-      "loss": 1.4029,
+      "loss": 1.4054,
       "step": 14
     },
     {
       "epoch": 0.03163305654408857,
-      "grad_norm": 0.4506430923938751,
+      "grad_norm": 0.42640671133995056,
       "learning_rate": 8.296729075500344e-05,
-      "loss": 1.3466,
+      "loss": 1.3487,
       "step": 15
     },
     {
       "epoch": 0.033741926980361145,
-      "grad_norm": 0.44114139676094055,
+      "grad_norm": 0.4178040027618408,
       "learning_rate": 8.043807145043604e-05,
-      "loss": 1.3213,
+      "loss": 1.3236,
       "step": 16
     },
     {
       "epoch": 0.03585079741663372,
-      "grad_norm": 0.444820761680603,
+      "grad_norm": 0.42343956232070923,
       "learning_rate": 7.777851165098012e-05,
-      "loss": 1.3173,
+      "loss": 1.3199,
       "step": 17
     },
     {
       "epoch": 0.03795966785290629,
-      "grad_norm": 0.47539570927619934,
+      "grad_norm": 0.4505631923675537,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 1.2578,
+      "loss": 1.2593,
       "step": 18
     },
     {
       "epoch": 0.040068538289178855,
-      "grad_norm": 0.42036089301109314,
+      "grad_norm": 0.40388721227645874,
       "learning_rate": 7.211443451095007e-05,
-      "loss": 1.2263,
+      "loss": 1.2285,
       "step": 19
     },
     {
       "epoch": 0.04217740872545143,
-      "grad_norm": 0.48343217372894287,
+      "grad_norm": 0.4622335731983185,
       "learning_rate": 6.91341716182545e-05,
-      "loss": 1.2371,
+      "loss": 1.2403,
       "step": 20
     },
     {
       "epoch": 0.044286279161724,
-      "grad_norm": 0.46945393085479736,
+      "grad_norm": 0.4492970108985901,
       "learning_rate": 6.607197326515808e-05,
-      "loss": 1.2132,
+      "loss": 1.2165,
       "step": 21
     },
     {
       "epoch": 0.04639514959799657,
-      "grad_norm": 0.4912544786930084,
+      "grad_norm": 0.4659884572029114,
       "learning_rate": 6.294095225512603e-05,
-      "loss": 1.1808,
+      "loss": 1.1846,
       "step": 22
     },
     {
       "epoch": 0.048504020034269145,
-      "grad_norm": 0.5027343034744263,
+      "grad_norm": 0.4740868806838989,
       "learning_rate": 5.9754516100806423e-05,
-      "loss": 1.1536,
+      "loss": 1.1572,
       "step": 23
     },
     {
       "epoch": 0.05061289047054172,
-      "grad_norm": 0.4939061999320984,
+      "grad_norm": 0.47230803966522217,
       "learning_rate": 5.6526309611002594e-05,
-      "loss": 1.1453,
+      "loss": 1.1504,
       "step": 24
     },
     {
       "epoch": 0.05272176090681429,
-      "grad_norm": 0.5463296175003052,
+      "grad_norm": 0.5276089906692505,
       "learning_rate": 5.327015646150716e-05,
-      "loss": 1.1425,
+      "loss": 1.1447,
       "step": 25
     },
     {
       "epoch": 0.05272176090681429,
-      "eval_loss": 1.1423113346099854,
-      "eval_runtime": 91.1944,
-      "eval_samples_per_second": 35.035,
-      "eval_steps_per_second": 4.386,
+      "eval_loss": 1.1456960439682007,
+      "eval_runtime": 91.1288,
+      "eval_samples_per_second": 35.06,
+      "eval_steps_per_second": 4.389,
       "step": 25
+    },
+    {
+      "epoch": 0.05483063134308686,
+      "grad_norm": 0.4547205865383148,
+      "learning_rate": 5e-05,
+      "loss": 1.198,
+      "step": 26
+    },
+    {
+      "epoch": 0.05693950177935943,
+      "grad_norm": 0.4535289406776428,
+      "learning_rate": 4.6729843538492847e-05,
+      "loss": 1.173,
+      "step": 27
+    },
+    {
+      "epoch": 0.059048372215632,
+      "grad_norm": 0.405439168214798,
+      "learning_rate": 4.347369038899744e-05,
+      "loss": 1.1636,
+      "step": 28
+    },
+    {
+      "epoch": 0.06115724265190457,
+      "grad_norm": 0.39680033922195435,
+      "learning_rate": 4.0245483899193595e-05,
+      "loss": 1.0978,
+      "step": 29
+    },
+    {
+      "epoch": 0.06326611308817714,
+      "grad_norm": 0.3968617618083954,
+      "learning_rate": 3.705904774487396e-05,
+      "loss": 1.1111,
+      "step": 30
+    },
+    {
+      "epoch": 0.06537498352444972,
+      "grad_norm": 0.4227343797683716,
+      "learning_rate": 3.392802673484193e-05,
+      "loss": 1.0829,
+      "step": 31
+    },
+    {
+      "epoch": 0.06748385396072229,
+      "grad_norm": 0.495728999376297,
+      "learning_rate": 3.086582838174551e-05,
+      "loss": 1.0851,
+      "step": 32
+    },
+    {
+      "epoch": 0.06959272439699486,
+      "grad_norm": 0.4811357855796814,
+      "learning_rate": 2.7885565489049946e-05,
+      "loss": 1.0681,
+      "step": 33
+    },
+    {
+      "epoch": 0.07170159483326743,
+      "grad_norm": 0.49746838212013245,
+      "learning_rate": 2.500000000000001e-05,
+      "loss": 1.0742,
+      "step": 34
+    },
+    {
+      "epoch": 0.07381046526954001,
+      "grad_norm": 0.4783548414707184,
+      "learning_rate": 2.2221488349019903e-05,
+      "loss": 1.0255,
+      "step": 35
+    },
+    {
+      "epoch": 0.07591933570581258,
+      "grad_norm": 0.5429141521453857,
+      "learning_rate": 1.9561928549563968e-05,
+      "loss": 1.0484,
+      "step": 36
+    },
+    {
+      "epoch": 0.07802820614208515,
+      "grad_norm": 0.5357297658920288,
+      "learning_rate": 1.703270924499656e-05,
+      "loss": 1.0614,
+      "step": 37
+    },
+    {
+      "epoch": 0.08013707657835771,
+      "grad_norm": 0.4397980272769928,
+      "learning_rate": 1.4644660940672627e-05,
+      "loss": 1.0864,
+      "step": 38
+    },
+    {
+      "epoch": 0.08224594701463028,
+      "grad_norm": 0.45920154452323914,
+      "learning_rate": 1.2408009626051137e-05,
+      "loss": 1.1013,
+      "step": 39
+    },
+    {
+      "epoch": 0.08435481745090286,
+      "grad_norm": 0.4466278553009033,
+      "learning_rate": 1.0332332985438248e-05,
+      "loss": 1.0761,
+      "step": 40
+    },
+    {
+      "epoch": 0.08646368788717543,
+      "grad_norm": 0.48130419850349426,
+      "learning_rate": 8.426519384872733e-06,
+      "loss": 1.0738,
+      "step": 41
+    },
+    {
+      "epoch": 0.088572558323448,
+      "grad_norm": 0.45974889397621155,
+      "learning_rate": 6.698729810778065e-06,
+      "loss": 1.0774,
+      "step": 42
+    },
+    {
+      "epoch": 0.09068142875972057,
+      "grad_norm": 0.4616631865501404,
+      "learning_rate": 5.156362923365588e-06,
+      "loss": 1.0489,
+      "step": 43
+    },
+    {
+      "epoch": 0.09279029919599315,
+      "grad_norm": 0.4217536747455597,
+      "learning_rate": 3.8060233744356633e-06,
+      "loss": 1.0136,
+      "step": 44
+    },
+    {
+      "epoch": 0.09489916963226572,
+      "grad_norm": 0.4349440038204193,
+      "learning_rate": 2.653493525244721e-06,
+      "loss": 1.0296,
+      "step": 45
+    },
+    {
+      "epoch": 0.09700804006853829,
+      "grad_norm": 0.4788382351398468,
+      "learning_rate": 1.70370868554659e-06,
+      "loss": 1.0326,
+      "step": 46
+    },
+    {
+      "epoch": 0.09911691050481086,
+      "grad_norm": 0.4514710605144501,
+      "learning_rate": 9.607359798384785e-07,
+      "loss": 1.0064,
+      "step": 47
+    },
+    {
+      "epoch": 0.10122578094108343,
+      "grad_norm": 0.4754306972026825,
+      "learning_rate": 4.277569313094809e-07,
+      "loss": 0.995,
+      "step": 48
+    },
+    {
+      "epoch": 0.10333465137735601,
+      "grad_norm": 0.5506452322006226,
+      "learning_rate": 1.0705383806982606e-07,
+      "loss": 0.9761,
+      "step": 49
+    },
+    {
+      "epoch": 0.10544352181362858,
+      "grad_norm": 0.59029620885849,
+      "learning_rate": 0.0,
+      "loss": 1.0103,
+      "step": 50
+    },
+    {
+      "epoch": 0.10544352181362858,
+      "eval_loss": 1.0235497951507568,
+      "eval_runtime": 91.03,
+      "eval_samples_per_second": 35.098,
+      "eval_steps_per_second": 4.394,
+      "step": 50
     }
   ],
   "logging_steps": 1,
@@ -221,12 +404,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 2.652642132099072e+17,
+  "total_flos": 5.305284264198144e+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 sha256:683d4cbeb722b54ff6db428ea473c68efc5977acfa89741fe44cd067e6661259
+oid sha256:7e48dd2262fdb84040320dc1ca35ffb4ba84f4356c28393139018be13df34c92
 size 6776