Upload folder using huggingface_hub

adapter_config.json

@@ -31,13 +31,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "k_proj",
-    "o_proj",
     "down_proj",
-    "gate_proj",
     "up_proj",
-    "v_proj"
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "gate_proj",
+    "q_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76fba2a2cbc0153d9e8cf2d8e607e939db4f192defb829aabe6488b0b349c79f
+oid sha256:6d1a57e9aa23e560e2276e0aa0af5260077e945ac721c19b7dc0508e0ba2004b
 size 73911112

checkpoint-532/adapter_config.json

@@ -31,13 +31,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "k_proj",
-    "o_proj",
     "down_proj",
-    "gate_proj",
     "up_proj",
-    "v_proj"
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "gate_proj",
+    "q_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

checkpoint-532/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76fba2a2cbc0153d9e8cf2d8e607e939db4f192defb829aabe6488b0b349c79f
+oid sha256:6d1a57e9aa23e560e2276e0aa0af5260077e945ac721c19b7dc0508e0ba2004b
 size 73911112

checkpoint-532/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cff43d6e5e15ebde6b18210b8a20bafe3827abaee7515938de328f303338dc0b
+oid sha256:922c3ce6e354e1f1e3c23afab873ae8a9473446d676105a0ca2be1919365e5f2
 size 37974661

checkpoint-532/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 532,
-  "best_metric": 0.061251163482666016,
+  "best_metric": 0.05935819819569588,
   "best_model_checkpoint": "/content/AdapterOps/checkpoints/intent/checkpoint-532",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -11,389 +11,389 @@
   "log_history": [
     {
       "epoch": 0.03766478342749529,
-      "grad_norm": 6.20037317276001,
+      "grad_norm": 6.231619834899902,
       "learning_rate": 7.82608695652174e-05,
-      "loss": 7.226158905029297,
+      "loss": 7.225121307373047,
       "step": 10
     },
     {
       "epoch": 0.07532956685499058,
-      "grad_norm": 4.269313335418701,
+      "grad_norm": 4.445438861846924,
       "learning_rate": 0.00016521739130434784,
-      "loss": 2.931910514831543,
+      "loss": 2.9320711135864257,
       "step": 20
     },
     {
       "epoch": 0.11299435028248588,
-      "grad_norm": 4.2698187828063965,
+      "grad_norm": 4.574609756469727,
       "learning_rate": 0.00019997042345528464,
-      "loss": 1.316406536102295,
+      "loss": 1.325807762145996,
       "step": 30
     },
     {
       "epoch": 0.15065913370998116,
-      "grad_norm": 2.978565216064453,
+      "grad_norm": 3.1792845726013184,
       "learning_rate": 0.00019978974125833138,
-      "loss": 0.6114205360412598,
+      "loss": 0.623131275177002,
       "step": 40
     },
     {
       "epoch": 0.18832391713747645,
-      "grad_norm": 2.545992374420166,
+      "grad_norm": 2.6347761154174805,
       "learning_rate": 0.00019944510477851702,
-      "loss": 0.3492830753326416,
+      "loss": 0.3416164398193359,
       "step": 50
     },
     {
       "epoch": 0.22598870056497175,
-      "grad_norm": 1.7521356344223022,
+      "grad_norm": 1.986854910850525,
       "learning_rate": 0.00019893708025267305,
-      "loss": 0.2880478143692017,
+      "loss": 0.2902896642684937,
       "step": 60
     },
     {
       "epoch": 0.263653483992467,
-      "grad_norm": 2.126889228820801,
+      "grad_norm": 1.8776882886886597,
       "learning_rate": 0.00019826650236376135,
-      "loss": 0.21670234203338623,
+      "loss": 0.22792267799377441,
       "step": 70
     },
     {
       "epoch": 0.3013182674199623,
-      "grad_norm": 2.089568614959717,
+      "grad_norm": 2.289062023162842,
       "learning_rate": 0.00019743447286949272,
-      "loss": 0.20672128200531006,
+      "loss": 0.21074450016021729,
       "step": 80
     },
     {
       "epoch": 0.3389830508474576,
-      "grad_norm": 1.7552835941314697,
+      "grad_norm": 1.6699317693710327,
       "learning_rate": 0.00019644235879214144,
-      "loss": 0.2067953586578369,
+      "loss": 0.19599651098251342,
       "step": 90
     },
     {
       "epoch": 0.3766478342749529,
-      "grad_norm": 1.6600853204727173,
+      "grad_norm": 1.7410376071929932,
       "learning_rate": 0.00019529179017253157,
-      "loss": 0.1796424627304077,
+      "loss": 0.17224516868591308,
       "step": 100
     },
     {
       "epoch": 0.4143126177024482,
-      "grad_norm": 2.1523630619049072,
+      "grad_norm": 2.3447396755218506,
       "learning_rate": 0.00019398465739188392,
-      "loss": 0.16519591808319092,
+      "loss": 0.1530194640159607,
       "step": 110
     },
     {
       "epoch": 0.4519774011299435,
-      "grad_norm": 1.8311508893966675,
+      "grad_norm": 1.5996431112289429,
       "learning_rate": 0.0001925231080659249,
-      "loss": 0.12011874914169311,
+      "loss": 0.1293744921684265,
       "step": 120
     },
     {
       "epoch": 0.4896421845574388,
-      "grad_norm": 2.920830011367798,
+      "grad_norm": 3.0032854080200195,
       "learning_rate": 0.0001909095435163593,
-      "loss": 0.19478075504302977,
+      "loss": 0.20522007942199708,
       "step": 130
     },
     {
       "epoch": 0.527306967984934,
-      "grad_norm": 2.6382505893707275,
+      "grad_norm": 3.551349639892578,
       "learning_rate": 0.00018914661482550524,
-      "loss": 0.12657949924468995,
+      "loss": 0.13586968183517456,
       "step": 140
     },
     {
       "epoch": 0.5649717514124294,
-      "grad_norm": 1.2078700065612793,
+      "grad_norm": 1.3218621015548706,
       "learning_rate": 0.0001872372184805727,
-      "loss": 0.12965470552444458,
+      "loss": 0.13858824968338013,
       "step": 150
     },
     {
       "epoch": 0.6026365348399246,
-      "grad_norm": 0.9879769086837769,
+      "grad_norm": 1.2914279699325562,
       "learning_rate": 0.00018518449161474272,
-      "loss": 0.10555235147476197,
+      "loss": 0.11016701459884644,
       "step": 160
     },
     {
       "epoch": 0.64030131826742,
-      "grad_norm": 1.425346851348877,
+      "grad_norm": 1.2859493494033813,
       "learning_rate": 0.00018299180685286585,
-      "loss": 0.10113340616226196,
+      "loss": 0.10769171714782715,
       "step": 170
     },
     {
       "epoch": 0.6779661016949152,
-      "grad_norm": 1.8217041492462158,
+      "grad_norm": 1.906212329864502,
       "learning_rate": 0.00018066276677024834,
-      "loss": 0.13109639883041382,
+      "loss": 0.14465000629425048,
       "step": 180
     },
     {
       "epoch": 0.7156308851224106,
-      "grad_norm": 1.5895063877105713,
+      "grad_norm": 1.5840388536453247,
       "learning_rate": 0.00017820119797363037,
-      "loss": 0.1346671462059021,
+      "loss": 0.13583729267120362,
       "step": 190
     },
     {
       "epoch": 0.7532956685499058,
-      "grad_norm": 1.4222699403762817,
+      "grad_norm": 1.3889877796173096,
       "learning_rate": 0.00017561114481408124,
-      "loss": 0.10248379707336426,
+      "loss": 0.0995070457458496,
       "step": 200
     },
     {
       "epoch": 0.7909604519774012,
-      "grad_norm": 1.1675019264221191,
+      "grad_norm": 1.0385699272155762,
       "learning_rate": 0.00017289686274214118,
-      "loss": 0.10411225557327271,
+      "loss": 0.09645452499389648,
       "step": 210
     },
     {
       "epoch": 0.8286252354048964,
-      "grad_norm": 1.4915348291397095,
+      "grad_norm": 1.157353401184082,
       "learning_rate": 0.0001700628113161272,
-      "loss": 0.09461969137191772,
+      "loss": 0.0947035312652588,
       "step": 220
     },
     {
       "epoch": 0.8662900188323918,
-      "grad_norm": 0.9363121390342712,
+      "grad_norm": 0.9945691823959351,
       "learning_rate": 0.00016711364687509072,
-      "loss": 0.10223276615142822,
+      "loss": 0.10556681156158447,
       "step": 230
     },
     {
       "epoch": 0.903954802259887,
-      "grad_norm": 1.248064637184143,
+      "grad_norm": 0.9192282557487488,
       "learning_rate": 0.00016405421488846446,
-      "loss": 0.1066591501235962,
+      "loss": 0.11258301734924317,
       "step": 240
     },
     {
       "epoch": 0.9416195856873822,
-      "grad_norm": 1.4022786617279053,
+      "grad_norm": 1.4237569570541382,
       "learning_rate": 0.00016088954199496903,
-      "loss": 0.07587293386459351,
+      "loss": 0.06957031488418579,
       "step": 250
     },
     {
       "epoch": 0.9792843691148776,
-      "grad_norm": 1.2635411024093628,
+      "grad_norm": 1.4789296388626099,
       "learning_rate": 0.00015762482774385868,
-      "loss": 0.08321117162704468,
+      "loss": 0.08124288320541381,
       "step": 260
     },
     {
       "epoch": 1.0,
-      "eval_loss": 0.09874702244997025,
-      "eval_runtime": 34.4183,
-      "eval_samples_per_second": 43.523,
-      "eval_steps_per_second": 5.462,
+      "eval_loss": 0.1032918393611908,
+      "eval_runtime": 33.1563,
+      "eval_samples_per_second": 45.18,
+      "eval_steps_per_second": 5.67,
       "step": 266
     },
     {
       "epoch": 1.015065913370998,
-      "grad_norm": 0.7147154211997986,
+      "grad_norm": 0.9435644745826721,
       "learning_rate": 0.00015426543605207553,
-      "loss": 0.05910256505012512,
+      "loss": 0.05932101607322693,
       "step": 270
     },
     {
       "epoch": 1.0527306967984935,
-      "grad_norm": 0.6574576497077942,
+      "grad_norm": 1.3172110319137573,
       "learning_rate": 0.00015081688639134822,
-      "loss": 0.050995242595672605,
+      "loss": 0.05661671757698059,
       "step": 280
     },
     {
       "epoch": 1.0903954802259888,
-      "grad_norm": 1.4628405570983887,
+      "grad_norm": 1.552285075187683,
       "learning_rate": 0.00014728484471971412,
-      "loss": 0.058632773160934445,
+      "loss": 0.07390114665031433,
       "step": 290
     },
     {
       "epoch": 1.128060263653484,
-      "grad_norm": 1.517785668373108,
+      "grad_norm": 1.7628093957901,
       "learning_rate": 0.00014367511417236534,
-      "loss": 0.07628054618835449,
+      "loss": 0.07481516003608704,
       "step": 300
     },
     {
       "epoch": 1.1657250470809792,
-      "grad_norm": 1.1749475002288818,
+      "grad_norm": 0.9738607406616211,
       "learning_rate": 0.00013999362552711281,
-      "loss": 0.0642553687095642,
+      "loss": 0.07062948942184448,
       "step": 310
     },
     {
       "epoch": 1.2033898305084745,
-      "grad_norm": 1.3391507863998413,
+      "grad_norm": 1.2930378913879395,
       "learning_rate": 0.0001362464274601339,
-      "loss": 0.061815792322158815,
+      "loss": 0.06054446697235107,
       "step": 320
     },
     {
       "epoch": 1.24105461393597,
-      "grad_norm": 0.9363414645195007,
+      "grad_norm": 0.8408831357955933,
       "learning_rate": 0.00013243967660801328,
-      "loss": 0.051718562841415405,
+      "loss": 0.04705864489078522,
       "step": 330
     },
     {
       "epoch": 1.2787193973634652,
-      "grad_norm": 0.8903082609176636,
+      "grad_norm": 1.0044398307800293,
       "learning_rate": 0.00012857962745240533,
-      "loss": 0.07189121842384338,
+      "loss": 0.06349811553955079,
       "step": 340
     },
     {
       "epoch": 1.3163841807909604,
-      "grad_norm": 1.0371206998825073,
+      "grad_norm": 0.9429423213005066,
       "learning_rate": 0.00012467262204393702,
-      "loss": 0.05366076231002807,
+      "loss": 0.05153215527534485,
       "step": 350
     },
     {
       "epoch": 1.3540489642184557,
-      "grad_norm": 0.4973160922527313,
+      "grad_norm": 0.2149270921945572,
       "learning_rate": 0.00012072507958223552,
-      "loss": 0.05838419198989868,
+      "loss": 0.051507169008255006,
       "step": 360
     },
     {
       "epoch": 1.3917137476459511,
-      "grad_norm": 0.9904340505599976,
+      "grad_norm": 0.8501299023628235,
       "learning_rate": 0.00011674348586919997,
-      "loss": 0.030171194672584535,
+      "loss": 0.03359214067459106,
       "step": 370
     },
     {
       "epoch": 1.4293785310734464,
-      "grad_norm": 0.23608283698558807,
+      "grad_norm": 0.1692386418581009,
       "learning_rate": 0.00011273438265284614,
-      "loss": 0.041062140464782716,
+      "loss": 0.04313917756080628,
       "step": 380
     },
     {
       "epoch": 1.4670433145009416,
-      "grad_norm": 1.248146891593933,
+      "grad_norm": 1.53399658203125,
       "learning_rate": 0.00010870435687923144,
-      "loss": 0.05552493929862976,
+      "loss": 0.04919858276844025,
       "step": 390
     },
     {
       "epoch": 1.5047080979284368,
-      "grad_norm": 0.9870753884315491,
+      "grad_norm": 1.1897956132888794,
       "learning_rate": 0.00010466002987011989,
-      "loss": 0.05117332935333252,
+      "loss": 0.0537997305393219,
       "step": 400
     },
     {
       "epoch": 1.542372881355932,
-      "grad_norm": 0.8861913084983826,
+      "grad_norm": 0.8689746260643005,
       "learning_rate": 0.00010060804644416803,
-      "loss": 0.023644092679023742,
+      "loss": 0.01916753053665161,
       "step": 410
     },
     {
       "epoch": 1.5800376647834273,
-      "grad_norm": 1.5339771509170532,
+      "grad_norm": 1.2721374034881592,
       "learning_rate": 9.655506399950495e-05,
-      "loss": 0.0456025093793869,
+      "loss": 0.036153167486190796,
       "step": 420
     },
     {
       "epoch": 1.6177024482109228,
-      "grad_norm": 0.25646108388900757,
+      "grad_norm": 0.12184591591358185,
       "learning_rate": 9.250774157564502e-05,
-      "loss": 0.05388702750205994,
+      "loss": 0.053461313247680664,
       "step": 430
     },
     {
       "epoch": 1.655367231638418,
-      "grad_norm": 1.134123682975769,
+      "grad_norm": 1.0807650089263916,
       "learning_rate": 8.84727289127033e-05,
-      "loss": 0.0578171968460083,
+      "loss": 0.05760731101036072,
       "step": 440
     },
     {
       "epoch": 1.6930320150659135,
-      "grad_norm": 1.1316230297088623,
+      "grad_norm": 0.7574961185455322,
       "learning_rate": 8.445665552588983e-05,
-      "loss": 0.04443947970867157,
+      "loss": 0.043853914737701415,
       "step": 450
     },
     {
       "epoch": 1.7306967984934087,
-      "grad_norm": 0.5303252339363098,
+      "grad_norm": 0.5318535566329956,
       "learning_rate": 8.046611981323386e-05,
-      "loss": 0.028096905350685118,
+      "loss": 0.03305320739746094,
       "step": 460
     },
     {
       "epoch": 1.768361581920904,
-      "grad_norm": 1.101265788078308,
+      "grad_norm": 1.0630979537963867,
       "learning_rate": 7.650767821443301e-05,
-      "loss": 0.042096829414367674,
+      "loss": 0.03825681805610657,
       "step": 470
     },
     {
       "epoch": 1.8060263653483992,
-      "grad_norm": 0.6456757187843323,
+      "grad_norm": 0.5549096465110779,
       "learning_rate": 7.258783443864003e-05,
-      "loss": 0.021614205837249757,
+      "loss": 0.021957556903362273,
       "step": 480
     },
     {
       "epoch": 1.8436911487758945,
-      "grad_norm": 0.9631510972976685,
+      "grad_norm": 0.940976619720459,
       "learning_rate": 6.871302877888591e-05,
-      "loss": 0.05064235925674439,
+      "loss": 0.05064261555671692,
       "step": 490
     },
     {
       "epoch": 1.8813559322033897,
-      "grad_norm": 1.0680396556854248,
+      "grad_norm": 1.263758897781372,
       "learning_rate": 6.488962753069528e-05,
-      "loss": 0.06349914669990539,
+      "loss": 0.06854571104049682,
       "step": 500
     },
     {
       "epoch": 1.9190207156308852,
-      "grad_norm": 1.2046319246292114,
+      "grad_norm": 1.0090230703353882,
       "learning_rate": 6.112391253227939e-05,
-      "loss": 0.04726651012897491,
+      "loss": 0.04064035415649414,
       "step": 510
     },
     {
       "epoch": 1.9566854990583804,
-      "grad_norm": 0.7896116375923157,
+      "grad_norm": 0.840827465057373,
       "learning_rate": 5.7422070843492734e-05,
-      "loss": 0.05674504637718201,
+      "loss": 0.05464768409729004,
       "step": 520
     },
     {
       "epoch": 1.9943502824858759,
-      "grad_norm": 1.1105738878250122,
+      "grad_norm": 1.2041618824005127,
       "learning_rate": 5.3790184580509736e-05,
-      "loss": 0.042013120651245114,
+      "loss": 0.044507384300231934,
       "step": 530
     },
     {
       "epoch": 2.0,
-      "eval_loss": 0.061251163482666016,
-      "eval_runtime": 34.3237,
-      "eval_samples_per_second": 43.643,
-      "eval_steps_per_second": 5.477,
+      "eval_loss": 0.05935819819569588,
+      "eval_runtime": 33.1876,
+      "eval_samples_per_second": 45.137,
+      "eval_steps_per_second": 5.665,
       "step": 532
     }
   ],

checkpoint-798/adapter_config.json

@@ -31,13 +31,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "k_proj",
-    "o_proj",
     "down_proj",
-    "gate_proj",
     "up_proj",
-    "v_proj"
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "gate_proj",
+    "q_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

checkpoint-798/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c80eac4a7e9c851cf809133f4016b128e300914796571fdbfb2ea5ebf68c1f3
+oid sha256:f1b9623a1067d2a874568e910e1119fb73f6225b08d2ea7f207bf0e5fecac2e4
 size 73911112

checkpoint-798/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4315f7cf8aac9f361ac3d5c8faa2319cb3109fbf33f3d8b636170ce2bcfcc523
+oid sha256:22c831136d8fc45bca0820738a27734d5ada4a486f6721ac00d3a3939e33d50a
 size 37974661

checkpoint-798/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 532,
-  "best_metric": 0.061251163482666016,
+  "best_metric": 0.05935819819569588,
   "best_model_checkpoint": "/content/AdapterOps/checkpoints/intent/checkpoint-532",
   "epoch": 3.0,
   "eval_steps": 500,
@@ -11,579 +11,579 @@
   "log_history": [
     {
       "epoch": 0.03766478342749529,
-      "grad_norm": 6.20037317276001,
+      "grad_norm": 6.231619834899902,
       "learning_rate": 7.82608695652174e-05,
-      "loss": 7.226158905029297,
+      "loss": 7.225121307373047,
       "step": 10
     },
     {
       "epoch": 0.07532956685499058,
-      "grad_norm": 4.269313335418701,
+      "grad_norm": 4.445438861846924,
       "learning_rate": 0.00016521739130434784,
-      "loss": 2.931910514831543,
+      "loss": 2.9320711135864257,
       "step": 20
     },
     {
       "epoch": 0.11299435028248588,
-      "grad_norm": 4.2698187828063965,
+      "grad_norm": 4.574609756469727,
       "learning_rate": 0.00019997042345528464,
-      "loss": 1.316406536102295,
+      "loss": 1.325807762145996,
       "step": 30
     },
     {
       "epoch": 0.15065913370998116,
-      "grad_norm": 2.978565216064453,
+      "grad_norm": 3.1792845726013184,
       "learning_rate": 0.00019978974125833138,
-      "loss": 0.6114205360412598,
+      "loss": 0.623131275177002,
       "step": 40
     },
     {
       "epoch": 0.18832391713747645,
-      "grad_norm": 2.545992374420166,
+      "grad_norm": 2.6347761154174805,
       "learning_rate": 0.00019944510477851702,
-      "loss": 0.3492830753326416,
+      "loss": 0.3416164398193359,
       "step": 50
     },
     {
       "epoch": 0.22598870056497175,
-      "grad_norm": 1.7521356344223022,
+      "grad_norm": 1.986854910850525,
       "learning_rate": 0.00019893708025267305,
-      "loss": 0.2880478143692017,
+      "loss": 0.2902896642684937,
       "step": 60
     },
     {
       "epoch": 0.263653483992467,
-      "grad_norm": 2.126889228820801,
+      "grad_norm": 1.8776882886886597,
       "learning_rate": 0.00019826650236376135,
-      "loss": 0.21670234203338623,
+      "loss": 0.22792267799377441,
       "step": 70
     },
     {
       "epoch": 0.3013182674199623,
-      "grad_norm": 2.089568614959717,
+      "grad_norm": 2.289062023162842,
       "learning_rate": 0.00019743447286949272,
-      "loss": 0.20672128200531006,
+      "loss": 0.21074450016021729,
       "step": 80
     },
     {
       "epoch": 0.3389830508474576,
-      "grad_norm": 1.7552835941314697,
+      "grad_norm": 1.6699317693710327,
       "learning_rate": 0.00019644235879214144,
-      "loss": 0.2067953586578369,
+      "loss": 0.19599651098251342,
       "step": 90
     },
     {
       "epoch": 0.3766478342749529,
-      "grad_norm": 1.6600853204727173,
+      "grad_norm": 1.7410376071929932,
       "learning_rate": 0.00019529179017253157,
-      "loss": 0.1796424627304077,
+      "loss": 0.17224516868591308,
       "step": 100
     },
     {
       "epoch": 0.4143126177024482,
-      "grad_norm": 2.1523630619049072,
+      "grad_norm": 2.3447396755218506,
       "learning_rate": 0.00019398465739188392,
-      "loss": 0.16519591808319092,
+      "loss": 0.1530194640159607,
       "step": 110
     },
     {
       "epoch": 0.4519774011299435,
-      "grad_norm": 1.8311508893966675,
+      "grad_norm": 1.5996431112289429,
       "learning_rate": 0.0001925231080659249,
-      "loss": 0.12011874914169311,
+      "loss": 0.1293744921684265,
       "step": 120
     },
     {
       "epoch": 0.4896421845574388,
-      "grad_norm": 2.920830011367798,
+      "grad_norm": 3.0032854080200195,
       "learning_rate": 0.0001909095435163593,
-      "loss": 0.19478075504302977,
+      "loss": 0.20522007942199708,
       "step": 130
     },
     {
       "epoch": 0.527306967984934,
-      "grad_norm": 2.6382505893707275,
+      "grad_norm": 3.551349639892578,
       "learning_rate": 0.00018914661482550524,
-      "loss": 0.12657949924468995,
+      "loss": 0.13586968183517456,
       "step": 140
     },
     {
       "epoch": 0.5649717514124294,
-      "grad_norm": 1.2078700065612793,
+      "grad_norm": 1.3218621015548706,
       "learning_rate": 0.0001872372184805727,
-      "loss": 0.12965470552444458,
+      "loss": 0.13858824968338013,
       "step": 150
     },
     {
       "epoch": 0.6026365348399246,
-      "grad_norm": 0.9879769086837769,
+      "grad_norm": 1.2914279699325562,
       "learning_rate": 0.00018518449161474272,
-      "loss": 0.10555235147476197,
+      "loss": 0.11016701459884644,
       "step": 160
     },
     {
       "epoch": 0.64030131826742,
-      "grad_norm": 1.425346851348877,
+      "grad_norm": 1.2859493494033813,
       "learning_rate": 0.00018299180685286585,
-      "loss": 0.10113340616226196,
+      "loss": 0.10769171714782715,
       "step": 170
     },
     {
       "epoch": 0.6779661016949152,
-      "grad_norm": 1.8217041492462158,
+      "grad_norm": 1.906212329864502,
       "learning_rate": 0.00018066276677024834,
-      "loss": 0.13109639883041382,
+      "loss": 0.14465000629425048,
       "step": 180
     },
     {
       "epoch": 0.7156308851224106,
-      "grad_norm": 1.5895063877105713,
+      "grad_norm": 1.5840388536453247,
       "learning_rate": 0.00017820119797363037,
-      "loss": 0.1346671462059021,
+      "loss": 0.13583729267120362,
       "step": 190
     },
     {
       "epoch": 0.7532956685499058,
-      "grad_norm": 1.4222699403762817,
+      "grad_norm": 1.3889877796173096,
       "learning_rate": 0.00017561114481408124,
-      "loss": 0.10248379707336426,
+      "loss": 0.0995070457458496,
       "step": 200
     },
     {
       "epoch": 0.7909604519774012,
-      "grad_norm": 1.1675019264221191,
+      "grad_norm": 1.0385699272155762,
       "learning_rate": 0.00017289686274214118,
-      "loss": 0.10411225557327271,
+      "loss": 0.09645452499389648,
       "step": 210
     },
     {
       "epoch": 0.8286252354048964,
-      "grad_norm": 1.4915348291397095,
+      "grad_norm": 1.157353401184082,
       "learning_rate": 0.0001700628113161272,
-      "loss": 0.09461969137191772,
+      "loss": 0.0947035312652588,
       "step": 220
     },
     {
       "epoch": 0.8662900188323918,
-      "grad_norm": 0.9363121390342712,
+      "grad_norm": 0.9945691823959351,
       "learning_rate": 0.00016711364687509072,
-      "loss": 0.10223276615142822,
+      "loss": 0.10556681156158447,
       "step": 230
     },
     {
       "epoch": 0.903954802259887,
-      "grad_norm": 1.248064637184143,
+      "grad_norm": 0.9192282557487488,
       "learning_rate": 0.00016405421488846446,
-      "loss": 0.1066591501235962,
+      "loss": 0.11258301734924317,
       "step": 240
     },
     {
       "epoch": 0.9416195856873822,
-      "grad_norm": 1.4022786617279053,
+      "grad_norm": 1.4237569570541382,
       "learning_rate": 0.00016088954199496903,
-      "loss": 0.07587293386459351,
+      "loss": 0.06957031488418579,
       "step": 250
     },
     {
       "epoch": 0.9792843691148776,
-      "grad_norm": 1.2635411024093628,
+      "grad_norm": 1.4789296388626099,
       "learning_rate": 0.00015762482774385868,
-      "loss": 0.08321117162704468,
+      "loss": 0.08124288320541381,
       "step": 260
     },
     {
       "epoch": 1.0,
-      "eval_loss": 0.09874702244997025,
-      "eval_runtime": 34.4183,
-      "eval_samples_per_second": 43.523,
-      "eval_steps_per_second": 5.462,
+      "eval_loss": 0.1032918393611908,
+      "eval_runtime": 33.1563,
+      "eval_samples_per_second": 45.18,
+      "eval_steps_per_second": 5.67,
       "step": 266
     },
     {
       "epoch": 1.015065913370998,
-      "grad_norm": 0.7147154211997986,
+      "grad_norm": 0.9435644745826721,
       "learning_rate": 0.00015426543605207553,
-      "loss": 0.05910256505012512,
+      "loss": 0.05932101607322693,
       "step": 270
     },
     {
       "epoch": 1.0527306967984935,
-      "grad_norm": 0.6574576497077942,
+      "grad_norm": 1.3172110319137573,
       "learning_rate": 0.00015081688639134822,
-      "loss": 0.050995242595672605,
+      "loss": 0.05661671757698059,
       "step": 280
     },
     {
       "epoch": 1.0903954802259888,
-      "grad_norm": 1.4628405570983887,
+      "grad_norm": 1.552285075187683,
       "learning_rate": 0.00014728484471971412,
-      "loss": 0.058632773160934445,
+      "loss": 0.07390114665031433,
       "step": 290
     },
     {
       "epoch": 1.128060263653484,
-      "grad_norm": 1.517785668373108,
+      "grad_norm": 1.7628093957901,
       "learning_rate": 0.00014367511417236534,
-      "loss": 0.07628054618835449,
+      "loss": 0.07481516003608704,
       "step": 300
     },
     {
       "epoch": 1.1657250470809792,
-      "grad_norm": 1.1749475002288818,
+      "grad_norm": 0.9738607406616211,
       "learning_rate": 0.00013999362552711281,
-      "loss": 0.0642553687095642,
+      "loss": 0.07062948942184448,
       "step": 310
     },
     {
       "epoch": 1.2033898305084745,
-      "grad_norm": 1.3391507863998413,
+      "grad_norm": 1.2930378913879395,
       "learning_rate": 0.0001362464274601339,
-      "loss": 0.061815792322158815,
+      "loss": 0.06054446697235107,
       "step": 320
     },
     {
       "epoch": 1.24105461393597,
-      "grad_norm": 0.9363414645195007,
+      "grad_norm": 0.8408831357955933,
       "learning_rate": 0.00013243967660801328,
-      "loss": 0.051718562841415405,
+      "loss": 0.04705864489078522,
       "step": 330
     },
     {
       "epoch": 1.2787193973634652,
-      "grad_norm": 0.8903082609176636,
+      "grad_norm": 1.0044398307800293,
       "learning_rate": 0.00012857962745240533,
-      "loss": 0.07189121842384338,
+      "loss": 0.06349811553955079,
       "step": 340
     },
     {
       "epoch": 1.3163841807909604,
-      "grad_norm": 1.0371206998825073,
+      "grad_norm": 0.9429423213005066,
       "learning_rate": 0.00012467262204393702,
-      "loss": 0.05366076231002807,
+      "loss": 0.05153215527534485,
       "step": 350
     },
     {
       "epoch": 1.3540489642184557,
-      "grad_norm": 0.4973160922527313,
+      "grad_norm": 0.2149270921945572,
       "learning_rate": 0.00012072507958223552,
-      "loss": 0.05838419198989868,
+      "loss": 0.051507169008255006,
       "step": 360
     },
     {
       "epoch": 1.3917137476459511,
-      "grad_norm": 0.9904340505599976,
+      "grad_norm": 0.8501299023628235,
       "learning_rate": 0.00011674348586919997,
-      "loss": 0.030171194672584535,
+      "loss": 0.03359214067459106,
       "step": 370
     },
     {
       "epoch": 1.4293785310734464,
-      "grad_norm": 0.23608283698558807,
+      "grad_norm": 0.1692386418581009,
       "learning_rate": 0.00011273438265284614,
-      "loss": 0.041062140464782716,
+      "loss": 0.04313917756080628,
       "step": 380
     },
     {
       "epoch": 1.4670433145009416,
-      "grad_norm": 1.248146891593933,
+      "grad_norm": 1.53399658203125,
       "learning_rate": 0.00010870435687923144,
-      "loss": 0.05552493929862976,
+      "loss": 0.04919858276844025,
       "step": 390
     },
     {
       "epoch": 1.5047080979284368,
-      "grad_norm": 0.9870753884315491,
+      "grad_norm": 1.1897956132888794,
       "learning_rate": 0.00010466002987011989,
-      "loss": 0.05117332935333252,
+      "loss": 0.0537997305393219,
       "step": 400
     },
     {
       "epoch": 1.542372881355932,
-      "grad_norm": 0.8861913084983826,
+      "grad_norm": 0.8689746260643005,
       "learning_rate": 0.00010060804644416803,
-      "loss": 0.023644092679023742,
+      "loss": 0.01916753053665161,
       "step": 410
     },
     {
       "epoch": 1.5800376647834273,
-      "grad_norm": 1.5339771509170532,
+      "grad_norm": 1.2721374034881592,
       "learning_rate": 9.655506399950495e-05,
-      "loss": 0.0456025093793869,
+      "loss": 0.036153167486190796,
       "step": 420
     },
     {
       "epoch": 1.6177024482109228,
-      "grad_norm": 0.25646108388900757,
+      "grad_norm": 0.12184591591358185,
       "learning_rate": 9.250774157564502e-05,
-      "loss": 0.05388702750205994,
+      "loss": 0.053461313247680664,
       "step": 430
     },
     {
       "epoch": 1.655367231638418,
-      "grad_norm": 1.134123682975769,
+      "grad_norm": 1.0807650089263916,
       "learning_rate": 8.84727289127033e-05,
-      "loss": 0.0578171968460083,
+      "loss": 0.05760731101036072,
       "step": 440
     },
     {
       "epoch": 1.6930320150659135,
-      "grad_norm": 1.1316230297088623,
+      "grad_norm": 0.7574961185455322,
       "learning_rate": 8.445665552588983e-05,
-      "loss": 0.04443947970867157,
+      "loss": 0.043853914737701415,
       "step": 450
     },
     {
       "epoch": 1.7306967984934087,
-      "grad_norm": 0.5303252339363098,
+      "grad_norm": 0.5318535566329956,
       "learning_rate": 8.046611981323386e-05,
-      "loss": 0.028096905350685118,
+      "loss": 0.03305320739746094,
       "step": 460
     },
     {
       "epoch": 1.768361581920904,
-      "grad_norm": 1.101265788078308,
+      "grad_norm": 1.0630979537963867,
       "learning_rate": 7.650767821443301e-05,
-      "loss": 0.042096829414367674,
+      "loss": 0.03825681805610657,
       "step": 470
     },
     {
       "epoch": 1.8060263653483992,
-      "grad_norm": 0.6456757187843323,
+      "grad_norm": 0.5549096465110779,
       "learning_rate": 7.258783443864003e-05,
-      "loss": 0.021614205837249757,
+      "loss": 0.021957556903362273,
       "step": 480
     },
     {
       "epoch": 1.8436911487758945,
-      "grad_norm": 0.9631510972976685,
+      "grad_norm": 0.940976619720459,
       "learning_rate": 6.871302877888591e-05,
-      "loss": 0.05064235925674439,
+      "loss": 0.05064261555671692,
       "step": 490
     },
     {
       "epoch": 1.8813559322033897,
-      "grad_norm": 1.0680396556854248,
+      "grad_norm": 1.263758897781372,
       "learning_rate": 6.488962753069528e-05,
-      "loss": 0.06349914669990539,
+      "loss": 0.06854571104049682,
       "step": 500
     },
     {
       "epoch": 1.9190207156308852,
-      "grad_norm": 1.2046319246292114,
+      "grad_norm": 1.0090230703353882,
       "learning_rate": 6.112391253227939e-05,
-      "loss": 0.04726651012897491,
+      "loss": 0.04064035415649414,
       "step": 510
     },
     {
       "epoch": 1.9566854990583804,
-      "grad_norm": 0.7896116375923157,
+      "grad_norm": 0.840827465057373,
       "learning_rate": 5.7422070843492734e-05,
-      "loss": 0.05674504637718201,
+      "loss": 0.05464768409729004,
       "step": 520
     },
     {
       "epoch": 1.9943502824858759,
-      "grad_norm": 1.1105738878250122,
+      "grad_norm": 1.2041618824005127,
       "learning_rate": 5.3790184580509736e-05,
-      "loss": 0.042013120651245114,
+      "loss": 0.044507384300231934,
       "step": 530
     },
     {
       "epoch": 2.0,
-      "eval_loss": 0.061251163482666016,
-      "eval_runtime": 34.3237,
-      "eval_samples_per_second": 43.643,
-      "eval_steps_per_second": 5.477,
+      "eval_loss": 0.05935819819569588,
+      "eval_runtime": 33.1876,
+      "eval_samples_per_second": 45.137,
+      "eval_steps_per_second": 5.665,
       "step": 532
     },
     {
       "epoch": 2.030131826741996,
-      "grad_norm": 0.25343388319015503,
+      "grad_norm": 0.37746816873550415,
       "learning_rate": 5.023422092292398e-05,
-      "loss": 0.02057059109210968,
+      "loss": 0.021823622286319733,
       "step": 540
     },
     {
       "epoch": 2.0677966101694913,
-      "grad_norm": 0.10271047800779343,
+      "grad_norm": 0.1335442066192627,
       "learning_rate": 4.676002230968763e-05,
-      "loss": 0.01983402818441391,
+      "loss": 0.019007197022438048,
       "step": 550
     },
     {
       "epoch": 2.105461393596987,
-      "grad_norm": 0.285036563873291,
+      "grad_norm": 0.24414069950580597,
       "learning_rate": 4.3373296839999724e-05,
-      "loss": 0.018048064410686494,
+      "loss": 0.014083166420459748,
       "step": 560
     },
     {
       "epoch": 2.1431261770244823,
-      "grad_norm": 0.4295148551464081,
+      "grad_norm": 0.44468846917152405,
       "learning_rate": 4.007960889491419e-05,
-      "loss": 0.01638820767402649,
+      "loss": 0.017849110066890717,
       "step": 570
     },
     {
       "epoch": 2.1807909604519775,
-      "grad_norm": 0.20831610262393951,
+      "grad_norm": 0.24019800126552582,
       "learning_rate": 3.688436999507623e-05,
-      "loss": 0.012473517656326294,
+      "loss": 0.011732180416584016,
       "step": 580
     },
     {
       "epoch": 2.2184557438794728,
-      "grad_norm": 0.09759155660867691,
+      "grad_norm": 0.07194910198450089,
       "learning_rate": 3.3792829909608305e-05,
-      "loss": 0.01275492161512375,
+      "loss": 0.01383921504020691,
       "step": 590
     },
     {
       "epoch": 2.256120527306968,
-      "grad_norm": 0.2636793553829193,
+      "grad_norm": 0.24043963849544525,
       "learning_rate": 3.0810068030753114e-05,
-      "loss": 0.013585537672042847,
+      "loss": 0.015586639940738677,
       "step": 600
     },
     {
       "epoch": 2.2937853107344632,
-      "grad_norm": 0.47027063369750977,
+      "grad_norm": 0.5626550316810608,
       "learning_rate": 2.7940985028445577e-05,
-      "loss": 0.020398685336112977,
+      "loss": 0.021509250998497008,
       "step": 610
     },
     {
       "epoch": 2.3314500941619585,
-      "grad_norm": 0.21346846222877502,
+      "grad_norm": 0.07269634306430817,
       "learning_rate": 2.5190294798524806e-05,
-      "loss": 0.011220137029886246,
+      "loss": 0.01105707734823227,
       "step": 620
     },
     {
       "epoch": 2.3691148775894537,
-      "grad_norm": 0.13497644662857056,
+      "grad_norm": 0.18274766206741333,
       "learning_rate": 2.2562516717815872e-05,
-      "loss": 0.009159407764673232,
+      "loss": 0.008608298748731614,
       "step": 630
     },
     {
       "epoch": 2.406779661016949,
-      "grad_norm": 0.2953830361366272,
+      "grad_norm": 0.34793245792388916,
       "learning_rate": 2.0061968218805437e-05,
-      "loss": 0.011393478512763977,
+      "loss": 0.014009971916675568,
       "step": 640
     },
     {
       "epoch": 2.4444444444444446,
-      "grad_norm": 0.829157292842865,
+      "grad_norm": 0.4050791561603546,
       "learning_rate": 1.769275769611143e-05,
-      "loss": 0.015227539837360382,
+      "loss": 0.016612522304058075,
       "step": 650
     },
     {
       "epoch": 2.48210922787194,
-      "grad_norm": 0.6772462725639343,
+      "grad_norm": 0.6629785299301147,
       "learning_rate": 1.545877775640162e-05,
-      "loss": 0.016884945333003998,
+      "loss": 0.01626759022474289,
       "step": 660
     },
     {
       "epoch": 2.519774011299435,
-      "grad_norm": 0.25403982400894165,
+      "grad_norm": 0.3090490400791168,
       "learning_rate": 1.3363698822851e-05,
-      "loss": 0.017729948461055755,
+      "loss": 0.016154083609580993,
       "step": 670
     },
     {
       "epoch": 2.5574387947269304,
-      "grad_norm": 0.05312857776880264,
+      "grad_norm": 0.04540738835930824,
       "learning_rate": 1.1410963104646144e-05,
-      "loss": 0.018179668486118315,
+      "loss": 0.015498106181621552,
       "step": 680
     },
     {
       "epoch": 2.5951035781544256,
-      "grad_norm": 0.4972018599510193,
+      "grad_norm": 0.5511218309402466,
       "learning_rate": 9.603778941444608e-06,
-      "loss": 0.0064490877091884615,
+      "loss": 0.006779346615076065,
       "step": 690
     },
     {
       "epoch": 2.632768361581921,
-      "grad_norm": 0.026564480736851692,
+      "grad_norm": 0.0463351272046566,
       "learning_rate": 7.945115532081272e-06,
-      "loss": 0.013317090272903443,
+      "loss": 0.009855969995260238,
       "step": 700
     },
     {
       "epoch": 2.670433145009416,
-      "grad_norm": 0.199781596660614,
+      "grad_norm": 0.23571087419986725,
       "learning_rate": 6.437698056182429e-06,
-      "loss": 0.009696292132139206,
+      "loss": 0.012080962210893631,
       "step": 710
     },
     {
       "epoch": 2.7080979284369113,
-      "grad_norm": 0.054052602499723434,
+      "grad_norm": 0.07977436482906342,
       "learning_rate": 5.084003196702802e-06,
-      "loss": 0.009346330910921097,
+      "loss": 0.011082074791193008,
       "step": 720
     },
     {
       "epoch": 2.7457627118644066,
-      "grad_norm": 0.9128419160842896,
+      "grad_norm": 0.7126105427742004,
       "learning_rate": 3.886255070742017e-06,
-      "loss": 0.010141096264123916,
+      "loss": 0.009956397861242295,
       "step": 730
     },
     {
       "epoch": 2.7834274952919023,
-      "grad_norm": 0.9610094428062439,
+      "grad_norm": 0.9239416122436523,
       "learning_rate": 2.846421575326019e-06,
-      "loss": 0.01998741179704666,
+      "loss": 0.017436034977436066,
       "step": 740
     },
     {
       "epoch": 2.8210922787193975,
-      "grad_norm": 0.5325580835342407,
+      "grad_norm": 0.386120080947876,
       "learning_rate": 1.9662111541573714e-06,
-      "loss": 0.011871393769979477,
+      "loss": 0.01102660819888115,
       "step": 750
     },
     {
       "epoch": 2.8587570621468927,
-      "grad_norm": 1.5276249647140503,
+      "grad_norm": 0.9826363325119019,
       "learning_rate": 1.2470699906468097e-06,
-      "loss": 0.014735010266304017,
+      "loss": 0.011629839986562729,
       "step": 760
     },
     {
       "epoch": 2.896421845574388,
-      "grad_norm": 0.07353071868419647,
+      "grad_norm": 0.202703595161438,
       "learning_rate": 6.901796318376951e-07,
-      "loss": 0.012803852558135986,
+      "loss": 0.013090962171554565,
       "step": 770
     },
     {
       "epoch": 2.934086629001883,
-      "grad_norm": 1.0715869665145874,
+      "grad_norm": 0.5159716010093689,
       "learning_rate": 2.9645504712726424e-07,
-      "loss": 0.013602715730667115,
+      "loss": 0.010157313197851181,
       "step": 780
     },
     {
       "epoch": 2.9717514124293785,
-      "grad_norm": 0.2556895911693573,
+      "grad_norm": 0.3126760721206665,
       "learning_rate": 6.654312497434978e-08,
-      "loss": 0.010806727409362792,
+      "loss": 0.01111321970820427,
       "step": 790
     },
     {
       "epoch": 3.0,
-      "eval_loss": 0.06183396279811859,
-      "eval_runtime": 34.741,
-      "eval_samples_per_second": 43.119,
-      "eval_steps_per_second": 5.411,
+      "eval_loss": 0.062141623347997665,
+      "eval_runtime": 33.1699,
+      "eval_samples_per_second": 45.161,
+      "eval_steps_per_second": 5.668,
       "step": 798
     }
   ],

train_summary.json

@@ -6,10 +6,10 @@
   "trainable_params": 18464768,
   "total_params": 907081216,
   "trainable_pct": 2.0356,
-  "train_loss": 0.2149,
+  "train_loss": 0.2153,
   "adapter_dir": "checkpoints/intent",
   "best_checkpoint": "/content/AdapterOps/checkpoints/intent/checkpoint-532",
-  "best_eval_loss": 0.061251163482666016,
+  "best_eval_loss": 0.05935819819569588,
   "prompt_tokens_masked": true,
   "m11_undertrained_dir": "checkpoints/intent-undertrained-m11",
   "seed": 20260909