{
  "best_global_step": 2667,
  "best_metric": 0.8418311044589408,
  "best_model_checkpoint": "3class_classification_results/run_20250610_161638/checkpoints/checkpoint-2667",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 8890,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.056274620146314014,
      "grad_norm": 1.2899854183197021,
      "learning_rate": 1.1023622047244096e-06,
      "loss": 1.1147,
      "step": 50
    },
    {
      "epoch": 0.11254924029262803,
      "grad_norm": 2.2132718563079834,
      "learning_rate": 2.227221597300338e-06,
      "loss": 1.0842,
      "step": 100
    },
    {
      "epoch": 0.16882386043894204,
      "grad_norm": 2.2031383514404297,
      "learning_rate": 3.3520809898762658e-06,
      "loss": 0.9439,
      "step": 150
    },
    {
      "epoch": 0.22509848058525606,
      "grad_norm": 2.492532730102539,
      "learning_rate": 4.476940382452194e-06,
      "loss": 0.7497,
      "step": 200
    },
    {
      "epoch": 0.28137310073157007,
      "grad_norm": 5.415732383728027,
      "learning_rate": 5.6017997750281215e-06,
      "loss": 0.5953,
      "step": 250
    },
    {
      "epoch": 0.3376477208778841,
      "grad_norm": 3.040374755859375,
      "learning_rate": 6.726659167604051e-06,
      "loss": 0.47,
      "step": 300
    },
    {
      "epoch": 0.3939223410241981,
      "grad_norm": 2.420459747314453,
      "learning_rate": 7.851518560179978e-06,
      "loss": 0.4128,
      "step": 350
    },
    {
      "epoch": 0.4501969611705121,
      "grad_norm": 2.526780128479004,
      "learning_rate": 8.976377952755906e-06,
      "loss": 0.3899,
      "step": 400
    },
    {
      "epoch": 0.5064715813168261,
      "grad_norm": 6.837190628051758,
      "learning_rate": 1.0101237345331833e-05,
      "loss": 0.4092,
      "step": 450
    },
    {
      "epoch": 0.5627462014631401,
      "grad_norm": 2.692833185195923,
      "learning_rate": 1.1226096737907763e-05,
      "loss": 0.3887,
      "step": 500
    },
    {
      "epoch": 0.6190208216094542,
      "grad_norm": 1.8825379610061646,
      "learning_rate": 1.2350956130483691e-05,
      "loss": 0.3891,
      "step": 550
    },
    {
      "epoch": 0.6752954417557682,
      "grad_norm": 3.277323007583618,
      "learning_rate": 1.3453318335208101e-05,
      "loss": 0.3878,
      "step": 600
    },
    {
      "epoch": 0.7315700619020822,
      "grad_norm": 2.1386890411376953,
      "learning_rate": 1.4578177727784028e-05,
      "loss": 0.3823,
      "step": 650
    },
    {
      "epoch": 0.7878446820483962,
      "grad_norm": 5.464720249176025,
      "learning_rate": 1.5703037120359956e-05,
      "loss": 0.3854,
      "step": 700
    },
    {
      "epoch": 0.8441193021947102,
      "grad_norm": 2.195767402648926,
      "learning_rate": 1.6827896512935883e-05,
      "loss": 0.383,
      "step": 750
    },
    {
      "epoch": 0.9003939223410242,
      "grad_norm": 3.1674859523773193,
      "learning_rate": 1.7952755905511813e-05,
      "loss": 0.3622,
      "step": 800
    },
    {
      "epoch": 0.9566685424873382,
      "grad_norm": 2.137552499771118,
      "learning_rate": 1.9077615298087743e-05,
      "loss": 0.3679,
      "step": 850
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8270935960591133,
      "eval_f1_fix": 0.9979395604395604,
      "eval_f1_issue": 0.8027574563871693,
      "eval_f1_symptom": 0.77491961414791,
      "eval_f1_weighted": 0.8262810094793769,
      "eval_loss": 0.3870583176612854,
      "eval_precision_fix": 0.9965706447187929,
      "eval_precision_issue": 0.7557615894039735,
      "eval_precision_symptom": 0.8347765846899896,
      "eval_precision_weighted": 0.8313149571029931,
      "eval_recall_fix": 0.9993122420907841,
      "eval_recall_issue": 0.855985598559856,
      "eval_recall_symptom": 0.723072307230723,
      "eval_recall_weighted": 0.8270935960591133,
      "eval_runtime": 50.1716,
      "eval_samples_per_second": 161.845,
      "eval_steps_per_second": 2.531,
      "step": 889
    },
    {
      "epoch": 1.012380416432189,
      "grad_norm": 2.139568567276001,
      "learning_rate": 1.9977502812148483e-05,
      "loss": 0.3606,
      "step": 900
    },
    {
      "epoch": 1.068655036578503,
      "grad_norm": 1.6121031045913696,
      "learning_rate": 1.9852518435195604e-05,
      "loss": 0.3454,
      "step": 950
    },
    {
      "epoch": 1.124929656724817,
      "grad_norm": 2.136821985244751,
      "learning_rate": 1.972753405824272e-05,
      "loss": 0.3622,
      "step": 1000
    },
    {
      "epoch": 1.181204276871131,
      "grad_norm": 2.97175669670105,
      "learning_rate": 1.960254968128984e-05,
      "loss": 0.3678,
      "step": 1050
    },
    {
      "epoch": 1.237478897017445,
      "grad_norm": 2.862729549407959,
      "learning_rate": 1.947756530433696e-05,
      "loss": 0.3742,
      "step": 1100
    },
    {
      "epoch": 1.293753517163759,
      "grad_norm": 1.6733105182647705,
      "learning_rate": 1.935258092738408e-05,
      "loss": 0.3575,
      "step": 1150
    },
    {
      "epoch": 1.350028137310073,
      "grad_norm": 2.2922866344451904,
      "learning_rate": 1.92275965504312e-05,
      "loss": 0.3593,
      "step": 1200
    },
    {
      "epoch": 1.406302757456387,
      "grad_norm": 1.7616901397705078,
      "learning_rate": 1.9102612173478316e-05,
      "loss": 0.371,
      "step": 1250
    },
    {
      "epoch": 1.4625773776027011,
      "grad_norm": 3.089038610458374,
      "learning_rate": 1.8977627796525436e-05,
      "loss": 0.3577,
      "step": 1300
    },
    {
      "epoch": 1.5188519977490151,
      "grad_norm": 2.3721745014190674,
      "learning_rate": 1.8852643419572557e-05,
      "loss": 0.3542,
      "step": 1350
    },
    {
      "epoch": 1.5751266178953292,
      "grad_norm": 4.346639156341553,
      "learning_rate": 1.8727659042619674e-05,
      "loss": 0.3562,
      "step": 1400
    },
    {
      "epoch": 1.6314012380416432,
      "grad_norm": 1.6947968006134033,
      "learning_rate": 1.8602674665666794e-05,
      "loss": 0.344,
      "step": 1450
    },
    {
      "epoch": 1.6876758581879572,
      "grad_norm": 3.7869679927825928,
      "learning_rate": 1.847769028871391e-05,
      "loss": 0.3459,
      "step": 1500
    },
    {
      "epoch": 1.7439504783342712,
      "grad_norm": 3.708456039428711,
      "learning_rate": 1.835270591176103e-05,
      "loss": 0.3502,
      "step": 1550
    },
    {
      "epoch": 1.8002250984805852,
      "grad_norm": 1.927826166152954,
      "learning_rate": 1.8227721534808152e-05,
      "loss": 0.343,
      "step": 1600
    },
    {
      "epoch": 1.8564997186268992,
      "grad_norm": 3.222015142440796,
      "learning_rate": 1.810273715785527e-05,
      "loss": 0.3358,
      "step": 1650
    },
    {
      "epoch": 1.9127743387732132,
      "grad_norm": 2.057603597640991,
      "learning_rate": 1.797775278090239e-05,
      "loss": 0.3532,
      "step": 1700
    },
    {
      "epoch": 1.9690489589195272,
      "grad_norm": 3.3761789798736572,
      "learning_rate": 1.7852768403949506e-05,
      "loss": 0.3392,
      "step": 1750
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8397783251231528,
      "eval_f1_fix": 0.9979395604395604,
      "eval_f1_issue": 0.7946936197094125,
      "eval_f1_symptom": 0.8147512864493996,
      "eval_f1_weighted": 0.839320688806212,
      "eval_loss": 0.3833693265914917,
      "eval_precision_fix": 0.9965706447187929,
      "eval_precision_issue": 0.8389463154384795,
      "eval_precision_symptom": 0.778050778050778,
      "eval_precision_weighted": 0.8421754963079828,
      "eval_recall_fix": 0.9993122420907841,
      "eval_recall_issue": 0.7548754875487549,
      "eval_recall_symptom": 0.8550855085508551,
      "eval_recall_weighted": 0.8397783251231528,
      "eval_runtime": 51.072,
      "eval_samples_per_second": 158.991,
      "eval_steps_per_second": 2.487,
      "step": 1778
    },
    {
      "epoch": 2.024760832864378,
      "grad_norm": 1.9975008964538574,
      "learning_rate": 1.7727784026996627e-05,
      "loss": 0.3269,
      "step": 1800
    },
    {
      "epoch": 2.081035453010692,
      "grad_norm": 1.5703338384628296,
      "learning_rate": 1.7602799650043747e-05,
      "loss": 0.3253,
      "step": 1850
    },
    {
      "epoch": 2.137310073157006,
      "grad_norm": 1.7055504322052002,
      "learning_rate": 1.7477815273090864e-05,
      "loss": 0.3076,
      "step": 1900
    },
    {
      "epoch": 2.19358469330332,
      "grad_norm": 2.055391550064087,
      "learning_rate": 1.7352830896137985e-05,
      "loss": 0.3175,
      "step": 1950
    },
    {
      "epoch": 2.249859313449634,
      "grad_norm": 2.3307132720947266,
      "learning_rate": 1.72278465191851e-05,
      "loss": 0.3166,
      "step": 2000
    },
    {
      "epoch": 2.306133933595948,
      "grad_norm": 1.9831819534301758,
      "learning_rate": 1.7102862142232222e-05,
      "loss": 0.3272,
      "step": 2050
    },
    {
      "epoch": 2.362408553742262,
      "grad_norm": 2.037224292755127,
      "learning_rate": 1.6977877765279342e-05,
      "loss": 0.3217,
      "step": 2100
    },
    {
      "epoch": 2.418683173888576,
      "grad_norm": 2.2483575344085693,
      "learning_rate": 1.6852893388326463e-05,
      "loss": 0.3483,
      "step": 2150
    },
    {
      "epoch": 2.47495779403489,
      "grad_norm": 2.124046564102173,
      "learning_rate": 1.672790901137358e-05,
      "loss": 0.3032,
      "step": 2200
    },
    {
      "epoch": 2.531232414181204,
      "grad_norm": 2.424370288848877,
      "learning_rate": 1.6602924634420697e-05,
      "loss": 0.3173,
      "step": 2250
    },
    {
      "epoch": 2.587507034327518,
      "grad_norm": 3.231492757797241,
      "learning_rate": 1.6477940257467817e-05,
      "loss": 0.3346,
      "step": 2300
    },
    {
      "epoch": 2.643781654473832,
      "grad_norm": 1.850974440574646,
      "learning_rate": 1.6355455568053996e-05,
      "loss": 0.3341,
      "step": 2350
    },
    {
      "epoch": 2.700056274620146,
      "grad_norm": 2.1213784217834473,
      "learning_rate": 1.6230471191101113e-05,
      "loss": 0.3137,
      "step": 2400
    },
    {
      "epoch": 2.75633089476646,
      "grad_norm": 1.9203131198883057,
      "learning_rate": 1.6105486814148233e-05,
      "loss": 0.3434,
      "step": 2450
    },
    {
      "epoch": 2.812605514912774,
      "grad_norm": 2.571561336517334,
      "learning_rate": 1.598050243719535e-05,
      "loss": 0.311,
      "step": 2500
    },
    {
      "epoch": 2.8688801350590882,
      "grad_norm": 1.646263837814331,
      "learning_rate": 1.585551806024247e-05,
      "loss": 0.3265,
      "step": 2550
    },
    {
      "epoch": 2.9251547552054022,
      "grad_norm": 3.992446184158325,
      "learning_rate": 1.573053368328959e-05,
      "loss": 0.3231,
      "step": 2600
    },
    {
      "epoch": 2.9814293753517163,
      "grad_norm": 3.225480794906616,
      "learning_rate": 1.5605549306336708e-05,
      "loss": 0.3164,
      "step": 2650
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8419950738916256,
      "eval_f1_fix": 0.998282377189969,
      "eval_f1_issue": 0.8020994133991973,
      "eval_f1_symptom": 0.8133119252663845,
      "eval_f1_weighted": 0.8418311044589408,
      "eval_loss": 0.3592279255390167,
      "eval_precision_fix": 0.997254632807138,
      "eval_precision_issue": 0.8260731319554849,
      "eval_precision_symptom": 0.7919272313814667,
      "eval_precision_weighted": 0.8427097841260638,
      "eval_recall_fix": 0.9993122420907841,
      "eval_recall_issue": 0.7794779477947795,
      "eval_recall_symptom": 0.8358835883588359,
      "eval_recall_weighted": 0.8419950738916256,
      "eval_runtime": 51.3031,
      "eval_samples_per_second": 158.275,
      "eval_steps_per_second": 2.475,
      "step": 2667
    },
    {
      "epoch": 3.037141249296567,
      "grad_norm": 2.5545248985290527,
      "learning_rate": 1.548056492938383e-05,
      "loss": 0.3147,
      "step": 2700
    },
    {
      "epoch": 3.093415869442881,
      "grad_norm": 1.628539800643921,
      "learning_rate": 1.5355580552430946e-05,
      "loss": 0.2979,
      "step": 2750
    },
    {
      "epoch": 3.1496904895891955,
      "grad_norm": 1.9369264841079712,
      "learning_rate": 1.5230596175478068e-05,
      "loss": 0.2869,
      "step": 2800
    },
    {
      "epoch": 3.205965109735509,
      "grad_norm": 2.5903725624084473,
      "learning_rate": 1.5105611798525186e-05,
      "loss": 0.272,
      "step": 2850
    },
    {
      "epoch": 3.2622397298818235,
      "grad_norm": 2.560624361038208,
      "learning_rate": 1.4980627421572305e-05,
      "loss": 0.2914,
      "step": 2900
    },
    {
      "epoch": 3.3185143500281375,
      "grad_norm": 3.008889675140381,
      "learning_rate": 1.4855643044619424e-05,
      "loss": 0.2858,
      "step": 2950
    },
    {
      "epoch": 3.3747889701744516,
      "grad_norm": 2.466245651245117,
      "learning_rate": 1.4730658667666542e-05,
      "loss": 0.3021,
      "step": 3000
    },
    {
      "epoch": 3.4310635903207656,
      "grad_norm": 2.1926143169403076,
      "learning_rate": 1.4605674290713663e-05,
      "loss": 0.2961,
      "step": 3050
    },
    {
      "epoch": 3.4873382104670796,
      "grad_norm": 2.4998419284820557,
      "learning_rate": 1.4480689913760782e-05,
      "loss": 0.2851,
      "step": 3100
    },
    {
      "epoch": 3.5436128306133936,
      "grad_norm": 2.801224946975708,
      "learning_rate": 1.43557055368079e-05,
      "loss": 0.3003,
      "step": 3150
    },
    {
      "epoch": 3.5998874507597076,
      "grad_norm": 1.9596877098083496,
      "learning_rate": 1.4230721159855019e-05,
      "loss": 0.3019,
      "step": 3200
    },
    {
      "epoch": 3.6561620709060216,
      "grad_norm": 1.780312418937683,
      "learning_rate": 1.4105736782902138e-05,
      "loss": 0.2962,
      "step": 3250
    },
    {
      "epoch": 3.7124366910523356,
      "grad_norm": 3.569025754928589,
      "learning_rate": 1.3980752405949258e-05,
      "loss": 0.2923,
      "step": 3300
    },
    {
      "epoch": 3.7687113111986497,
      "grad_norm": 2.3551599979400635,
      "learning_rate": 1.3855768028996377e-05,
      "loss": 0.3113,
      "step": 3350
    },
    {
      "epoch": 3.8249859313449637,
      "grad_norm": 1.77907133102417,
      "learning_rate": 1.3730783652043495e-05,
      "loss": 0.3056,
      "step": 3400
    },
    {
      "epoch": 3.8812605514912777,
      "grad_norm": 4.4045610427856445,
      "learning_rate": 1.3605799275090614e-05,
      "loss": 0.2942,
      "step": 3450
    },
    {
      "epoch": 3.9375351716375917,
      "grad_norm": 1.8489115238189697,
      "learning_rate": 1.3480814898137733e-05,
      "loss": 0.3136,
      "step": 3500
    },
    {
      "epoch": 3.9938097917839057,
      "grad_norm": 1.9342581033706665,
      "learning_rate": 1.3355830521184853e-05,
      "loss": 0.3061,
      "step": 3550
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.84064039408867,
      "eval_f1_fix": 0.998282377189969,
      "eval_f1_issue": 0.7974306752310826,
      "eval_f1_symptom": 0.8142816009213936,
      "eval_f1_weighted": 0.8403127577401992,
      "eval_loss": 0.3966975510120392,
      "eval_precision_fix": 0.997254632807138,
      "eval_precision_issue": 0.8344262295081967,
      "eval_precision_symptom": 0.782729034043731,
      "eval_precision_weighted": 0.8423628977241568,
      "eval_recall_fix": 0.9993122420907841,
      "eval_recall_issue": 0.7635763576357636,
      "eval_recall_symptom": 0.8484848484848485,
      "eval_recall_weighted": 0.84064039408867,
      "eval_runtime": 50.3322,
      "eval_samples_per_second": 161.328,
      "eval_steps_per_second": 2.523,
      "step": 3556
    },
    {
      "epoch": 4.049521665728756,
      "grad_norm": 2.619023084640503,
      "learning_rate": 1.3230846144231972e-05,
      "loss": 0.2784,
      "step": 3600
    },
    {
      "epoch": 4.10579628587507,
      "grad_norm": 2.1567516326904297,
      "learning_rate": 1.310586176727909e-05,
      "loss": 0.2836,
      "step": 3650
    },
    {
      "epoch": 4.162070906021384,
      "grad_norm": 1.734316110610962,
      "learning_rate": 1.298087739032621e-05,
      "loss": 0.2725,
      "step": 3700
    },
    {
      "epoch": 4.218345526167698,
      "grad_norm": 2.4064743518829346,
      "learning_rate": 1.2855893013373328e-05,
      "loss": 0.2674,
      "step": 3750
    },
    {
      "epoch": 4.274620146314012,
      "grad_norm": 4.878196716308594,
      "learning_rate": 1.2730908636420448e-05,
      "loss": 0.2614,
      "step": 3800
    },
    {
      "epoch": 4.330894766460326,
      "grad_norm": 3.776557683944702,
      "learning_rate": 1.2605924259467567e-05,
      "loss": 0.2759,
      "step": 3850
    },
    {
      "epoch": 4.38716938660664,
      "grad_norm": 2.9769930839538574,
      "learning_rate": 1.2480939882514686e-05,
      "loss": 0.27,
      "step": 3900
    },
    {
      "epoch": 4.443444006752954,
      "grad_norm": 2.369786500930786,
      "learning_rate": 1.2355955505561805e-05,
      "loss": 0.2661,
      "step": 3950
    },
    {
      "epoch": 4.499718626899268,
      "grad_norm": 3.501908540725708,
      "learning_rate": 1.2230971128608923e-05,
      "loss": 0.2756,
      "step": 4000
    },
    {
      "epoch": 4.555993247045582,
      "grad_norm": 3.2472310066223145,
      "learning_rate": 1.2105986751656045e-05,
      "loss": 0.2638,
      "step": 4050
    },
    {
      "epoch": 4.612267867191896,
      "grad_norm": 1.875646710395813,
      "learning_rate": 1.1981002374703164e-05,
      "loss": 0.2455,
      "step": 4100
    },
    {
      "epoch": 4.66854248733821,
      "grad_norm": 2.917334794998169,
      "learning_rate": 1.1856017997750283e-05,
      "loss": 0.2608,
      "step": 4150
    },
    {
      "epoch": 4.724817107484524,
      "grad_norm": 2.597090482711792,
      "learning_rate": 1.1731033620797401e-05,
      "loss": 0.2565,
      "step": 4200
    },
    {
      "epoch": 4.781091727630838,
      "grad_norm": 2.790834903717041,
      "learning_rate": 1.160604924384452e-05,
      "loss": 0.2724,
      "step": 4250
    },
    {
      "epoch": 4.837366347777152,
      "grad_norm": 2.6425554752349854,
      "learning_rate": 1.148106486689164e-05,
      "loss": 0.2826,
      "step": 4300
    },
    {
      "epoch": 4.893640967923466,
      "grad_norm": 2.039649486541748,
      "learning_rate": 1.135608048993876e-05,
      "loss": 0.2584,
      "step": 4350
    },
    {
      "epoch": 4.94991558806978,
      "grad_norm": 3.3404409885406494,
      "learning_rate": 1.1231096112985878e-05,
      "loss": 0.2704,
      "step": 4400
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8385467980295567,
      "eval_f1_fix": 0.9986254295532646,
      "eval_f1_issue": 0.7967101179391682,
      "eval_f1_symptom": 0.8100493755445832,
      "eval_f1_weighted": 0.8383412273832256,
      "eval_loss": 0.3984265625476837,
      "eval_precision_fix": 0.9979395604395604,
      "eval_precision_issue": 0.825136612021858,
      "eval_precision_symptom": 0.7849704475091472,
      "eval_precision_weighted": 0.839592481563542,
      "eval_recall_fix": 0.9993122420907841,
      "eval_recall_issue": 0.7701770177017702,
      "eval_recall_symptom": 0.8367836783678367,
      "eval_recall_weighted": 0.8385467980295567,
      "eval_runtime": 103.4526,
      "eval_samples_per_second": 78.49,
      "eval_steps_per_second": 1.228,
      "step": 4445
    },
    {
      "epoch": 5.0056274620146315,
      "grad_norm": 1.412779688835144,
      "learning_rate": 1.1106111736032997e-05,
      "loss": 0.2558,
      "step": 4450
    },
    {
      "epoch": 5.0619020821609455,
      "grad_norm": 3.5638949871063232,
      "learning_rate": 1.0981127359080115e-05,
      "loss": 0.2201,
      "step": 4500
    },
    {
      "epoch": 5.1181767023072595,
      "grad_norm": 2.8439533710479736,
      "learning_rate": 1.0856142982127236e-05,
      "loss": 0.2406,
      "step": 4550
    },
    {
      "epoch": 5.1744513224535735,
      "grad_norm": 3.5135629177093506,
      "learning_rate": 1.0731158605174354e-05,
      "loss": 0.2282,
      "step": 4600
    },
    {
      "epoch": 5.2307259425998875,
      "grad_norm": 3.6669528484344482,
      "learning_rate": 1.0606174228221473e-05,
      "loss": 0.2266,
      "step": 4650
    },
    {
      "epoch": 5.287000562746202,
      "grad_norm": 3.3074584007263184,
      "learning_rate": 1.0481189851268592e-05,
      "loss": 0.2635,
      "step": 4700
    },
    {
      "epoch": 5.343275182892516,
      "grad_norm": 2.5459887981414795,
      "learning_rate": 1.035620547431571e-05,
      "loss": 0.2471,
      "step": 4750
    },
    {
      "epoch": 5.39954980303883,
      "grad_norm": 3.6159589290618896,
      "learning_rate": 1.0231221097362831e-05,
      "loss": 0.2443,
      "step": 4800
    },
    {
      "epoch": 5.455824423185144,
      "grad_norm": 4.78350830078125,
      "learning_rate": 1.010623672040995e-05,
      "loss": 0.2465,
      "step": 4850
    },
    {
      "epoch": 5.512099043331458,
      "grad_norm": 3.147178888320923,
      "learning_rate": 9.981252343457068e-06,
      "loss": 0.2523,
      "step": 4900
    },
    {
      "epoch": 5.568373663477772,
      "grad_norm": 3.182088613510132,
      "learning_rate": 9.856267966504189e-06,
      "loss": 0.2382,
      "step": 4950
    },
    {
      "epoch": 5.624648283624086,
      "grad_norm": 4.077524662017822,
      "learning_rate": 9.733783277090364e-06,
      "loss": 0.2342,
      "step": 5000
    },
    {
      "epoch": 5.6809229037704,
      "grad_norm": 4.5491509437561035,
      "learning_rate": 9.608798900137485e-06,
      "loss": 0.2438,
      "step": 5050
    },
    {
      "epoch": 5.737197523916714,
      "grad_norm": 3.9461286067962646,
      "learning_rate": 9.483814523184603e-06,
      "loss": 0.2294,
      "step": 5100
    },
    {
      "epoch": 5.793472144063028,
      "grad_norm": 2.644874095916748,
      "learning_rate": 9.358830146231722e-06,
      "loss": 0.2431,
      "step": 5150
    },
    {
      "epoch": 5.849746764209342,
      "grad_norm": 3.191740036010742,
      "learning_rate": 9.23384576927884e-06,
      "loss": 0.2455,
      "step": 5200
    },
    {
      "epoch": 5.906021384355656,
      "grad_norm": 2.2712607383728027,
      "learning_rate": 9.10886139232596e-06,
      "loss": 0.2407,
      "step": 5250
    },
    {
      "epoch": 5.96229600450197,
      "grad_norm": 2.775583505630493,
      "learning_rate": 8.98387701537308e-06,
      "loss": 0.2551,
      "step": 5300
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8370689655172414,
      "eval_f1_fix": 0.9979395604395604,
      "eval_f1_issue": 0.8008436275986743,
      "eval_f1_symptom": 0.8029895366218236,
      "eval_f1_weighted": 0.8370172484268523,
      "eval_loss": 0.4547211527824402,
      "eval_precision_fix": 0.9965706447187929,
      "eval_precision_issue": 0.8042360060514372,
      "eval_precision_symptom": 0.8001191540065534,
      "eval_precision_weighted": 0.8369863874254196,
      "eval_recall_fix": 0.9993122420907841,
      "eval_recall_issue": 0.7974797479747975,
      "eval_recall_symptom": 0.8058805880588059,
      "eval_recall_weighted": 0.8370689655172414,
      "eval_runtime": 52.2924,
      "eval_samples_per_second": 155.281,
      "eval_steps_per_second": 2.429,
      "step": 5334
    },
    {
      "epoch": 6.01800787844682,
      "grad_norm": 2.528533697128296,
      "learning_rate": 8.858892638420198e-06,
      "loss": 0.2138,
      "step": 5350
    },
    {
      "epoch": 6.074282498593134,
      "grad_norm": 5.1876091957092285,
      "learning_rate": 8.733908261467317e-06,
      "loss": 0.2319,
      "step": 5400
    },
    {
      "epoch": 6.130557118739448,
      "grad_norm": 1.4345381259918213,
      "learning_rate": 8.608923884514436e-06,
      "loss": 0.2118,
      "step": 5450
    },
    {
      "epoch": 6.186831738885762,
      "grad_norm": 6.736217021942139,
      "learning_rate": 8.483939507561556e-06,
      "loss": 0.2171,
      "step": 5500
    },
    {
      "epoch": 6.243106359032076,
      "grad_norm": 3.6418821811676025,
      "learning_rate": 8.358955130608675e-06,
      "loss": 0.2339,
      "step": 5550
    },
    {
      "epoch": 6.299380979178391,
      "grad_norm": 3.062988758087158,
      "learning_rate": 8.233970753655794e-06,
      "loss": 0.2121,
      "step": 5600
    },
    {
      "epoch": 6.355655599324704,
      "grad_norm": 3.603121519088745,
      "learning_rate": 8.108986376702912e-06,
      "loss": 0.211,
      "step": 5650
    },
    {
      "epoch": 6.411930219471018,
      "grad_norm": 3.1732518672943115,
      "learning_rate": 7.984001999750031e-06,
      "loss": 0.1988,
      "step": 5700
    },
    {
      "epoch": 6.468204839617332,
      "grad_norm": 4.5081071853637695,
      "learning_rate": 7.859017622797151e-06,
      "loss": 0.2006,
      "step": 5750
    },
    {
      "epoch": 6.524479459763647,
      "grad_norm": 3.192850112915039,
      "learning_rate": 7.73403324584427e-06,
      "loss": 0.2225,
      "step": 5800
    },
    {
      "epoch": 6.58075407990996,
      "grad_norm": 4.354557514190674,
      "learning_rate": 7.609048868891389e-06,
      "loss": 0.2232,
      "step": 5850
    },
    {
      "epoch": 6.637028700056275,
      "grad_norm": 2.804410934448242,
      "learning_rate": 7.484064491938508e-06,
      "loss": 0.2207,
      "step": 5900
    },
    {
      "epoch": 6.693303320202588,
      "grad_norm": 4.296571731567383,
      "learning_rate": 7.359080114985627e-06,
      "loss": 0.21,
      "step": 5950
    },
    {
      "epoch": 6.749577940348903,
      "grad_norm": 2.1833271980285645,
      "learning_rate": 7.234095738032747e-06,
      "loss": 0.2101,
      "step": 6000
    },
    {
      "epoch": 6.805852560495216,
      "grad_norm": 4.7521138191223145,
      "learning_rate": 7.109111361079865e-06,
      "loss": 0.2263,
      "step": 6050
    },
    {
      "epoch": 6.862127180641531,
      "grad_norm": 4.192446231842041,
      "learning_rate": 6.984126984126984e-06,
      "loss": 0.2127,
      "step": 6100
    },
    {
      "epoch": 6.918401800787844,
      "grad_norm": 3.3775792121887207,
      "learning_rate": 6.859142607174104e-06,
      "loss": 0.2254,
      "step": 6150
    },
    {
      "epoch": 6.974676420934159,
      "grad_norm": 3.5780911445617676,
      "learning_rate": 6.734158230221223e-06,
      "loss": 0.1976,
      "step": 6200
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8341133004926108,
      "eval_f1_fix": 0.998282377189969,
      "eval_f1_issue": 0.7951293759512937,
      "eval_f1_symptom": 0.8013019677467081,
      "eval_f1_weighted": 0.8340404242585783,
      "eval_loss": 0.45761045813560486,
      "eval_precision_fix": 0.997254632807138,
      "eval_precision_issue": 0.8069199876428792,
      "eval_precision_symptom": 0.7904261529480444,
      "eval_precision_weighted": 0.8342318870309269,
      "eval_recall_fix": 0.9993122420907841,
      "eval_recall_issue": 0.7836783678367837,
      "eval_recall_symptom": 0.8124812481248125,
      "eval_recall_weighted": 0.8341133004926108,
      "eval_runtime": 50.9303,
      "eval_samples_per_second": 159.433,
      "eval_steps_per_second": 2.494,
      "step": 6223
    },
    {
      "epoch": 7.0303882948790095,
      "grad_norm": 6.685973167419434,
      "learning_rate": 6.609173853268343e-06,
      "loss": 0.197,
      "step": 6250
    },
    {
      "epoch": 7.0866629150253235,
      "grad_norm": 3.8742218017578125,
      "learning_rate": 6.484189476315461e-06,
      "loss": 0.183,
      "step": 6300
    },
    {
      "epoch": 7.1429375351716375,
      "grad_norm": 5.564633846282959,
      "learning_rate": 6.35920509936258e-06,
      "loss": 0.2029,
      "step": 6350
    },
    {
      "epoch": 7.1992121553179516,
      "grad_norm": 4.8350677490234375,
      "learning_rate": 6.2342207224097e-06,
      "loss": 0.1889,
      "step": 6400
    },
    {
      "epoch": 7.255486775464266,
      "grad_norm": 2.719174385070801,
      "learning_rate": 6.109236345456818e-06,
      "loss": 0.2021,
      "step": 6450
    },
    {
      "epoch": 7.31176139561058,
      "grad_norm": 3.9081249237060547,
      "learning_rate": 5.984251968503938e-06,
      "loss": 0.1916,
      "step": 6500
    },
    {
      "epoch": 7.368036015756894,
      "grad_norm": 4.187448024749756,
      "learning_rate": 5.859267591551057e-06,
      "loss": 0.2077,
      "step": 6550
    },
    {
      "epoch": 7.424310635903208,
      "grad_norm": 2.6563658714294434,
      "learning_rate": 5.734283214598175e-06,
      "loss": 0.2,
      "step": 6600
    },
    {
      "epoch": 7.480585256049522,
      "grad_norm": 5.274929523468018,
      "learning_rate": 5.609298837645295e-06,
      "loss": 0.203,
      "step": 6650
    },
    {
      "epoch": 7.536859876195836,
      "grad_norm": 4.843599319458008,
      "learning_rate": 5.4843144606924135e-06,
      "loss": 0.1909,
      "step": 6700
    },
    {
      "epoch": 7.59313449634215,
      "grad_norm": 3.966811418533325,
      "learning_rate": 5.359330083739533e-06,
      "loss": 0.1953,
      "step": 6750
    },
    {
      "epoch": 7.649409116488464,
      "grad_norm": 3.038121461868286,
      "learning_rate": 5.234345706786652e-06,
      "loss": 0.2036,
      "step": 6800
    },
    {
      "epoch": 7.705683736634778,
      "grad_norm": 4.253869533538818,
      "learning_rate": 5.1093613298337705e-06,
      "loss": 0.1942,
      "step": 6850
    },
    {
      "epoch": 7.761958356781092,
      "grad_norm": 2.9078307151794434,
      "learning_rate": 4.984376952880891e-06,
      "loss": 0.1957,
      "step": 6900
    },
    {
      "epoch": 7.818232976927406,
      "grad_norm": 4.637989044189453,
      "learning_rate": 4.85939257592801e-06,
      "loss": 0.1803,
      "step": 6950
    },
    {
      "epoch": 7.87450759707372,
      "grad_norm": 4.39274787902832,
      "learning_rate": 4.734408198975128e-06,
      "loss": 0.2025,
      "step": 7000
    },
    {
      "epoch": 7.930782217220034,
      "grad_norm": 4.669205665588379,
      "learning_rate": 4.609423822022248e-06,
      "loss": 0.195,
      "step": 7050
    },
    {
      "epoch": 7.987056837366348,
      "grad_norm": 5.979293346405029,
      "learning_rate": 4.4844394450693665e-06,
      "loss": 0.1966,
      "step": 7100
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8283251231527093,
      "eval_f1_fix": 0.998282377189969,
      "eval_f1_issue": 0.7821735731039875,
      "eval_f1_symptom": 0.799538505912893,
      "eval_f1_weighted": 0.8279986374134825,
      "eval_loss": 0.5127028226852417,
      "eval_precision_fix": 0.997254632807138,
      "eval_precision_issue": 0.8167864141084259,
      "eval_precision_symptom": 0.7697861705081922,
      "eval_precision_weighted": 0.8298096872695525,
      "eval_recall_fix": 0.9993122420907841,
      "eval_recall_issue": 0.7503750375037503,
      "eval_recall_symptom": 0.8316831683168316,
      "eval_recall_weighted": 0.8283251231527093,
      "eval_runtime": 51.6048,
      "eval_samples_per_second": 157.35,
      "eval_steps_per_second": 2.461,
      "step": 7112
    },
    {
      "epoch": 8.042768711311199,
      "grad_norm": 4.806234836578369,
      "learning_rate": 4.359455068116486e-06,
      "loss": 0.1907,
      "step": 7150
    },
    {
      "epoch": 8.099043331457512,
      "grad_norm": 4.466273784637451,
      "learning_rate": 4.234470691163605e-06,
      "loss": 0.1822,
      "step": 7200
    },
    {
      "epoch": 8.155317951603827,
      "grad_norm": 6.873985767364502,
      "learning_rate": 4.1094863142107235e-06,
      "loss": 0.174,
      "step": 7250
    },
    {
      "epoch": 8.21159257175014,
      "grad_norm": 8.243498802185059,
      "learning_rate": 3.984501937257843e-06,
      "loss": 0.1684,
      "step": 7300
    },
    {
      "epoch": 8.267867191896455,
      "grad_norm": 4.151490211486816,
      "learning_rate": 3.859517560304963e-06,
      "loss": 0.1736,
      "step": 7350
    },
    {
      "epoch": 8.324141812042768,
      "grad_norm": 5.426071643829346,
      "learning_rate": 3.7345331833520813e-06,
      "loss": 0.1849,
      "step": 7400
    },
    {
      "epoch": 8.380416432189083,
      "grad_norm": 3.12210750579834,
      "learning_rate": 3.6095488063992e-06,
      "loss": 0.1703,
      "step": 7450
    },
    {
      "epoch": 8.436691052335396,
      "grad_norm": 4.037646770477295,
      "learning_rate": 3.4845644294463195e-06,
      "loss": 0.1632,
      "step": 7500
    },
    {
      "epoch": 8.492965672481711,
      "grad_norm": 3.56674861907959,
      "learning_rate": 3.3595800524934387e-06,
      "loss": 0.1709,
      "step": 7550
    },
    {
      "epoch": 8.549240292628024,
      "grad_norm": 4.258781433105469,
      "learning_rate": 3.234595675540558e-06,
      "loss": 0.1754,
      "step": 7600
    },
    {
      "epoch": 8.605514912774339,
      "grad_norm": 4.218986511230469,
      "learning_rate": 3.109611298587677e-06,
      "loss": 0.1683,
      "step": 7650
    },
    {
      "epoch": 8.661789532920652,
      "grad_norm": 4.169147491455078,
      "learning_rate": 2.984626921634796e-06,
      "loss": 0.1865,
      "step": 7700
    },
    {
      "epoch": 8.718064153066967,
      "grad_norm": 2.8161799907684326,
      "learning_rate": 2.8596425446819147e-06,
      "loss": 0.1877,
      "step": 7750
    },
    {
      "epoch": 8.77433877321328,
      "grad_norm": 3.580389976501465,
      "learning_rate": 2.734658167729034e-06,
      "loss": 0.1694,
      "step": 7800
    },
    {
      "epoch": 8.830613393359595,
      "grad_norm": 5.852592468261719,
      "learning_rate": 2.6096737907761534e-06,
      "loss": 0.1801,
      "step": 7850
    },
    {
      "epoch": 8.886888013505908,
      "grad_norm": 3.4175920486450195,
      "learning_rate": 2.484689413823272e-06,
      "loss": 0.1768,
      "step": 7900
    },
    {
      "epoch": 8.943162633652223,
      "grad_norm": 6.1865081787109375,
      "learning_rate": 2.3597050368703912e-06,
      "loss": 0.1811,
      "step": 7950
    },
    {
      "epoch": 8.999437253798536,
      "grad_norm": 5.233773708343506,
      "learning_rate": 2.2347206599175104e-06,
      "loss": 0.1829,
      "step": 8000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8298029556650246,
      "eval_f1_fix": 0.9986254295532646,
      "eval_f1_issue": 0.7910425177787865,
      "eval_f1_symptom": 0.7948221990775183,
      "eval_f1_weighted": 0.8297645906222303,
      "eval_loss": 0.5134788751602173,
      "eval_precision_fix": 0.9979395604395604,
      "eval_precision_issue": 0.797924297924298,
      "eval_precision_symptom": 0.788370720188902,
      "eval_precision_weighted": 0.8298184010160612,
      "eval_recall_fix": 0.9993122420907841,
      "eval_recall_issue": 0.7842784278427842,
      "eval_recall_symptom": 0.8013801380138014,
      "eval_recall_weighted": 0.8298029556650246,
      "eval_runtime": 53.8444,
      "eval_samples_per_second": 150.805,
      "eval_steps_per_second": 2.359,
      "step": 8001
    },
    {
      "epoch": 9.055149127743388,
      "grad_norm": 3.5759809017181396,
      "learning_rate": 2.1097362829646295e-06,
      "loss": 0.1629,
      "step": 8050
    },
    {
      "epoch": 9.111423747889702,
      "grad_norm": 3.6214778423309326,
      "learning_rate": 1.987251593550806e-06,
      "loss": 0.1694,
      "step": 8100
    },
    {
      "epoch": 9.167698368036016,
      "grad_norm": 6.99995756149292,
      "learning_rate": 1.8622672165979255e-06,
      "loss": 0.1775,
      "step": 8150
    },
    {
      "epoch": 9.22397298818233,
      "grad_norm": 4.065187931060791,
      "learning_rate": 1.7372828396450444e-06,
      "loss": 0.182,
      "step": 8200
    },
    {
      "epoch": 9.280247608328644,
      "grad_norm": 2.9803073406219482,
      "learning_rate": 1.6122984626921637e-06,
      "loss": 0.168,
      "step": 8250
    },
    {
      "epoch": 9.336522228474958,
      "grad_norm": 5.370956897735596,
      "learning_rate": 1.4873140857392826e-06,
      "loss": 0.1493,
      "step": 8300
    },
    {
      "epoch": 9.392796848621272,
      "grad_norm": 5.8644890785217285,
      "learning_rate": 1.3623297087864017e-06,
      "loss": 0.1511,
      "step": 8350
    },
    {
      "epoch": 9.449071468767587,
      "grad_norm": 4.59880256652832,
      "learning_rate": 1.2373453318335209e-06,
      "loss": 0.1626,
      "step": 8400
    },
    {
      "epoch": 9.5053460889139,
      "grad_norm": 3.982597827911377,
      "learning_rate": 1.11236095488064e-06,
      "loss": 0.165,
      "step": 8450
    },
    {
      "epoch": 9.561620709060215,
      "grad_norm": 6.357668876647949,
      "learning_rate": 9.873765779277591e-07,
      "loss": 0.1659,
      "step": 8500
    },
    {
      "epoch": 9.617895329206528,
      "grad_norm": 8.072972297668457,
      "learning_rate": 8.623922009748782e-07,
      "loss": 0.186,
      "step": 8550
    },
    {
      "epoch": 9.674169949352843,
      "grad_norm": 4.884040832519531,
      "learning_rate": 7.374078240219973e-07,
      "loss": 0.1702,
      "step": 8600
    },
    {
      "epoch": 9.730444569499156,
      "grad_norm": 7.7868123054504395,
      "learning_rate": 6.124234470691164e-07,
      "loss": 0.1603,
      "step": 8650
    },
    {
      "epoch": 9.78671918964547,
      "grad_norm": 5.416253089904785,
      "learning_rate": 4.874390701162355e-07,
      "loss": 0.152,
      "step": 8700
    },
    {
      "epoch": 9.842993809791784,
      "grad_norm": 5.989937782287598,
      "learning_rate": 3.6245469316335464e-07,
      "loss": 0.1558,
      "step": 8750
    },
    {
      "epoch": 9.899268429938099,
      "grad_norm": 5.493338108062744,
      "learning_rate": 2.374703162104737e-07,
      "loss": 0.1645,
      "step": 8800
    },
    {
      "epoch": 9.955543050084412,
      "grad_norm": 6.584114074707031,
      "learning_rate": 1.124859392575928e-07,
      "loss": 0.1569,
      "step": 8850
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8299261083743842,
      "eval_f1_fix": 0.998282377189969,
      "eval_f1_issue": 0.7886029411764706,
      "eval_f1_symptom": 0.7975297750330834,
      "eval_f1_weighted": 0.8298131674335786,
      "eval_loss": 0.5292249321937561,
      "eval_precision_fix": 0.997254632807138,
      "eval_precision_issue": 0.8056338028169014,
      "eval_precision_symptom": 0.7820069204152249,
      "eval_precision_weighted": 0.8302481239697359,
      "eval_recall_fix": 0.9993122420907841,
      "eval_recall_issue": 0.7722772277227723,
      "eval_recall_symptom": 0.8136813681368137,
      "eval_recall_weighted": 0.8299261083743842,
      "eval_runtime": 51.5276,
      "eval_samples_per_second": 157.585,
      "eval_steps_per_second": 2.465,
      "step": 8890
    }
  ],
  "logging_steps": 50,
  "max_steps": 8890,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.73930339431168e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}