Upload 10 files

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da9c76f91c340c43d00e90b30b080373a0dd6615a7e02480016eb919b5c67da9
+oid sha256:6a3f02550d4f935dc69517ae713f60c984b68e1c4e6c70297462b4ac3502c96f
 size 3111656212

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5c5b28dd2354302b279aed74a13d3a719376bf9b72edbf8ba24e4989e51956c
+oid sha256:f5559e84ab4d20220b4e995bab1078649df513d51ef9ee25c55e8c0095675ed0
 size 6223436042

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44213e11b20ad35cc12ee23e9b37516197be0a22514fe7582c6f89a61de845f3
+oid sha256:3aa126ec760ffe2d99e9cdb113d1ce118c69d0c5b58b29cd42dcf9c28d3e67db
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31026445b691f1b7b313e7e4d39feecc6ba53ee700f381fcaf28d9a5411a611c
+oid sha256:3165a19a79ff0fa578a1a8f54ac4647e71f7ae2c9a8788aa153845e7e2fdb748
 size 1064

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.3108769357204437,
-  "best_model_checkpoint": "./checkpoint-xlm-v-base/checkpoint-15000",
-  "epoch": 0.6640694173897644,
+  "best_metric": 0.267339825630188,
+  "best_model_checkpoint": "./checkpoint-xlm-v-base/checkpoint-62000",
+  "epoch": 3.032583672746591,
   "eval_steps": 1000,
-  "global_step": 15000,
+  "global_step": 68500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -337,6 +337,1172 @@
       "eval_samples_per_second": 74.212,
       "eval_steps_per_second": 0.29,
       "step": 15000
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 36.22490692138672,
+      "learning_rate": 1.7976169394621674e-05,
+      "loss": 0.3332,
+      "step": 16000
+    },
+    {
+      "epoch": 0.71,
+      "eval_LOC_f1": 0.8066277986085525,
+      "eval_ORG_f1": 0.6983725665012176,
+      "eval_PER_f1": 0.8332057011578333,
+      "eval_loss": 0.31944143772125244,
+      "eval_overall_accuracy": 0.9018524580941595,
+      "eval_overall_f1": 0.7838363709114303,
+      "eval_overall_precision": 0.7856547450536315,
+      "eval_overall_recall": 0.7820263944768853,
+      "eval_runtime": 885.7636,
+      "eval_samples_per_second": 74.173,
+      "eval_steps_per_second": 0.29,
+      "step": 16000
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 12.729876518249512,
+      "learning_rate": 1.7849679981785527e-05,
+      "loss": 0.3324,
+      "step": 17000
+    },
+    {
+      "epoch": 0.75,
+      "eval_LOC_f1": 0.8091387580118001,
+      "eval_ORG_f1": 0.7010727593385756,
+      "eval_PER_f1": 0.8321347376235326,
+      "eval_loss": 0.3180248737335205,
+      "eval_overall_accuracy": 0.9008195115735422,
+      "eval_overall_f1": 0.7850356248119096,
+      "eval_overall_precision": 0.7839540384210074,
+      "eval_overall_recall": 0.7861201997589117,
+      "eval_runtime": 884.7821,
+      "eval_samples_per_second": 74.256,
+      "eval_steps_per_second": 0.29,
+      "step": 17000
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 8.6387939453125,
+      "learning_rate": 1.772319056894938e-05,
+      "loss": 0.3385,
+      "step": 18000
+    },
+    {
+      "epoch": 0.8,
+      "eval_LOC_f1": 0.7763382604086921,
+      "eval_ORG_f1": 0.6755694544225161,
+      "eval_PER_f1": 0.8256535378180551,
+      "eval_loss": 0.33690232038497925,
+      "eval_overall_accuracy": 0.8964328777930731,
+      "eval_overall_f1": 0.7651045588831604,
+      "eval_overall_precision": 0.755082284607938,
+      "eval_overall_recall": 0.7753964650813282,
+      "eval_runtime": 890.8661,
+      "eval_samples_per_second": 73.748,
+      "eval_steps_per_second": 0.288,
+      "step": 18000
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 9.216795921325684,
+      "learning_rate": 1.7596701156113234e-05,
+      "loss": 0.3267,
+      "step": 19000
+    },
+    {
+      "epoch": 0.84,
+      "eval_LOC_f1": 0.8015086633534421,
+      "eval_ORG_f1": 0.70105107410561,
+      "eval_PER_f1": 0.8364990020855292,
+      "eval_loss": 0.3270108103752136,
+      "eval_overall_accuracy": 0.9019283701744147,
+      "eval_overall_f1": 0.7856358496296815,
+      "eval_overall_precision": 0.7816849022099619,
+      "eval_overall_recall": 0.789626939273917,
+      "eval_runtime": 950.9161,
+      "eval_samples_per_second": 69.091,
+      "eval_steps_per_second": 0.27,
+      "step": 19000
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 10.781560897827148,
+      "learning_rate": 1.747021174327709e-05,
+      "loss": 0.3233,
+      "step": 20000
+    },
+    {
+      "epoch": 0.89,
+      "eval_LOC_f1": 0.8056958546374753,
+      "eval_ORG_f1": 0.6917519315097098,
+      "eval_PER_f1": 0.8345942771968653,
+      "eval_loss": 0.30491939187049866,
+      "eval_overall_accuracy": 0.9033815442821598,
+      "eval_overall_f1": 0.783789045768533,
+      "eval_overall_precision": 0.7798233379823338,
+      "eval_overall_recall": 0.7877952940808116,
+      "eval_runtime": 917.217,
+      "eval_samples_per_second": 71.63,
+      "eval_steps_per_second": 0.28,
+      "step": 20000
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 15.803696632385254,
+      "learning_rate": 1.7343722330440944e-05,
+      "loss": 0.3112,
+      "step": 21000
+    },
+    {
+      "epoch": 0.93,
+      "eval_LOC_f1": 0.8185129588612579,
+      "eval_ORG_f1": 0.7198663442992582,
+      "eval_PER_f1": 0.8407628833170422,
+      "eval_loss": 0.3230852782726288,
+      "eval_overall_accuracy": 0.9054094812832666,
+      "eval_overall_f1": 0.7978603713205666,
+      "eval_overall_precision": 0.7936613663999752,
+      "eval_overall_recall": 0.8021040437090032,
+      "eval_runtime": 886.1519,
+      "eval_samples_per_second": 74.141,
+      "eval_steps_per_second": 0.29,
+      "step": 21000
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 2.6037917137145996,
+      "learning_rate": 1.7217232917604798e-05,
+      "loss": 0.3256,
+      "step": 22000
+    },
+    {
+      "epoch": 0.97,
+      "eval_LOC_f1": 0.8135399708226737,
+      "eval_ORG_f1": 0.7164829968607569,
+      "eval_PER_f1": 0.8380035321776889,
+      "eval_loss": 0.3068985044956207,
+      "eval_overall_accuracy": 0.9048909746636654,
+      "eval_overall_f1": 0.7934551505253751,
+      "eval_overall_precision": 0.781072268142868,
+      "eval_overall_recall": 0.806236986708831,
+      "eval_runtime": 885.0947,
+      "eval_samples_per_second": 74.229,
+      "eval_steps_per_second": 0.29,
+      "step": 22000
+    },
+    {
+      "epoch": 1.02,
+      "grad_norm": 20.057506561279297,
+      "learning_rate": 1.709074350476865e-05,
+      "loss": 0.2931,
+      "step": 23000
+    },
+    {
+      "epoch": 1.02,
+      "eval_LOC_f1": 0.8090675101559834,
+      "eval_ORG_f1": 0.7063356635658061,
+      "eval_PER_f1": 0.8313781014023732,
+      "eval_loss": 0.3428688645362854,
+      "eval_overall_accuracy": 0.8993378704357015,
+      "eval_overall_f1": 0.7878607280931577,
+      "eval_overall_precision": 0.779137676333938,
+      "eval_overall_recall": 0.7967813140880129,
+      "eval_runtime": 913.4969,
+      "eval_samples_per_second": 71.921,
+      "eval_steps_per_second": 0.281,
+      "step": 23000
+    },
+    {
+      "epoch": 1.06,
+      "grad_norm": 10.215392112731934,
+      "learning_rate": 1.6964254091932504e-05,
+      "loss": 0.2936,
+      "step": 24000
+    },
+    {
+      "epoch": 1.06,
+      "eval_LOC_f1": 0.8279096724337782,
+      "eval_ORG_f1": 0.7213703143205346,
+      "eval_PER_f1": 0.8399587047500638,
+      "eval_loss": 0.3276561200618744,
+      "eval_overall_accuracy": 0.9048035402140856,
+      "eval_overall_f1": 0.8012747941254013,
+      "eval_overall_precision": 0.7938388261504187,
+      "eval_overall_recall": 0.8088513862579645,
+      "eval_runtime": 888.4065,
+      "eval_samples_per_second": 73.953,
+      "eval_steps_per_second": 0.289,
+      "step": 24000
+    },
+    {
+      "epoch": 1.11,
+      "grad_norm": 13.276248931884766,
+      "learning_rate": 1.683776467909636e-05,
+      "loss": 0.2797,
+      "step": 25000
+    },
+    {
+      "epoch": 1.11,
+      "eval_LOC_f1": 0.8137454037894649,
+      "eval_ORG_f1": 0.7157125069541951,
+      "eval_PER_f1": 0.8422392139071162,
+      "eval_loss": 0.30677318572998047,
+      "eval_overall_accuracy": 0.9050170429398038,
+      "eval_overall_f1": 0.7953167244565965,
+      "eval_overall_precision": 0.7764588335296837,
+      "eval_overall_recall": 0.8151134211061885,
+      "eval_runtime": 923.9779,
+      "eval_samples_per_second": 71.106,
+      "eval_steps_per_second": 0.278,
+      "step": 25000
+    },
+    {
+      "epoch": 1.15,
+      "grad_norm": 0.08989755064249039,
+      "learning_rate": 1.6711275266260215e-05,
+      "loss": 0.2792,
+      "step": 26000
+    },
+    {
+      "epoch": 1.15,
+      "eval_LOC_f1": 0.8237387838615486,
+      "eval_ORG_f1": 0.717827626918536,
+      "eval_PER_f1": 0.8449089917750373,
+      "eval_loss": 0.3350381851196289,
+      "eval_overall_accuracy": 0.9065115620198306,
+      "eval_overall_f1": 0.7990867935095444,
+      "eval_overall_precision": 0.7941203753807146,
+      "eval_overall_recall": 0.8041157224039952,
+      "eval_runtime": 950.4133,
+      "eval_samples_per_second": 69.128,
+      "eval_steps_per_second": 0.27,
+      "step": 26000
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 32.182918548583984,
+      "learning_rate": 1.658478585342407e-05,
+      "loss": 0.2698,
+      "step": 27000
+    },
+    {
+      "epoch": 1.2,
+      "eval_LOC_f1": 0.8267147626869356,
+      "eval_ORG_f1": 0.7322118816415036,
+      "eval_PER_f1": 0.8463502705378134,
+      "eval_loss": 0.3303050696849823,
+      "eval_overall_accuracy": 0.9053356025623038,
+      "eval_overall_f1": 0.8061476513209491,
+      "eval_overall_precision": 0.803686040812516,
+      "eval_overall_recall": 0.8086243874947164,
+      "eval_runtime": 903.1604,
+      "eval_samples_per_second": 72.745,
+      "eval_steps_per_second": 0.285,
+      "step": 27000
+    },
+    {
+      "epoch": 1.24,
+      "grad_norm": 27.704275131225586,
+      "learning_rate": 1.6458296440587925e-05,
+      "loss": 0.2846,
+      "step": 28000
+    },
+    {
+      "epoch": 1.24,
+      "eval_LOC_f1": 0.8198407012516331,
+      "eval_ORG_f1": 0.7170252756930114,
+      "eval_PER_f1": 0.8497716275494053,
+      "eval_loss": 0.3040228486061096,
+      "eval_overall_accuracy": 0.9088763588770705,
+      "eval_overall_f1": 0.7998951624404026,
+      "eval_overall_precision": 0.7879266486958503,
+      "eval_overall_recall": 0.8122328850760054,
+      "eval_runtime": 883.6437,
+      "eval_samples_per_second": 74.351,
+      "eval_steps_per_second": 0.291,
+      "step": 28000
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 1.532094120979309,
+      "learning_rate": 1.633180702775178e-05,
+      "loss": 0.2765,
+      "step": 29000
+    },
+    {
+      "epoch": 1.28,
+      "eval_LOC_f1": 0.8226323815533471,
+      "eval_ORG_f1": 0.7322743544720759,
+      "eval_PER_f1": 0.8458698818030955,
+      "eval_loss": 0.3010263741016388,
+      "eval_overall_accuracy": 0.9093941877102408,
+      "eval_overall_f1": 0.8048208514659728,
+      "eval_overall_precision": 0.799766577265244,
+      "eval_overall_recall": 0.8099394148128435,
+      "eval_runtime": 883.3843,
+      "eval_samples_per_second": 74.373,
+      "eval_steps_per_second": 0.291,
+      "step": 29000
+    },
+    {
+      "epoch": 1.33,
+      "grad_norm": 2.7648439407348633,
+      "learning_rate": 1.6205317614915632e-05,
+      "loss": 0.2758,
+      "step": 30000
+    },
+    {
+      "epoch": 1.33,
+      "eval_LOC_f1": 0.8158232882579698,
+      "eval_ORG_f1": 0.7211553763726063,
+      "eval_PER_f1": 0.8392668350824088,
+      "eval_loss": 0.2979504466056824,
+      "eval_overall_accuracy": 0.907270682822384,
+      "eval_overall_f1": 0.7967142515352101,
+      "eval_overall_precision": 0.7737616641463505,
+      "eval_overall_recall": 0.8210701817555615,
+      "eval_runtime": 884.2556,
+      "eval_samples_per_second": 74.3,
+      "eval_steps_per_second": 0.291,
+      "step": 30000
+    },
+    {
+      "epoch": 1.37,
+      "grad_norm": 2.9498727321624756,
+      "learning_rate": 1.607882820207949e-05,
+      "loss": 0.2745,
+      "step": 31000
+    },
+    {
+      "epoch": 1.37,
+      "eval_LOC_f1": 0.8269012485811577,
+      "eval_ORG_f1": 0.7277582167305856,
+      "eval_PER_f1": 0.856517895595802,
+      "eval_loss": 0.2944641709327698,
+      "eval_overall_accuracy": 0.9108385505943848,
+      "eval_overall_f1": 0.807803496021649,
+      "eval_overall_precision": 0.7947313807024321,
+      "eval_overall_recall": 0.8213128356059302,
+      "eval_runtime": 883.9066,
+      "eval_samples_per_second": 74.329,
+      "eval_steps_per_second": 0.291,
+      "step": 31000
+    },
+    {
+      "epoch": 1.42,
+      "grad_norm": 11.60289192199707,
+      "learning_rate": 1.5952338789243342e-05,
+      "loss": 0.2645,
+      "step": 32000
+    },
+    {
+      "epoch": 1.42,
+      "eval_LOC_f1": 0.8305319969159598,
+      "eval_ORG_f1": 0.7228604829282057,
+      "eval_PER_f1": 0.8315148384875288,
+      "eval_loss": 0.32325080037117004,
+      "eval_overall_accuracy": 0.9048618298471388,
+      "eval_overall_f1": 0.7998450483255535,
+      "eval_overall_precision": 0.7917570997998328,
+      "eval_overall_recall": 0.8080999420761776,
+      "eval_runtime": 933.3011,
+      "eval_samples_per_second": 70.395,
+      "eval_steps_per_second": 0.275,
+      "step": 32000
+    },
+    {
+      "epoch": 1.46,
+      "grad_norm": 42.618431091308594,
+      "learning_rate": 1.5825849376407196e-05,
+      "loss": 0.2779,
+      "step": 33000
+    },
+    {
+      "epoch": 1.46,
+      "eval_LOC_f1": 0.8264125401549256,
+      "eval_ORG_f1": 0.7424042624042624,
+      "eval_PER_f1": 0.8601716304896517,
+      "eval_loss": 0.2943771183490753,
+      "eval_overall_accuracy": 0.9127221190857203,
+      "eval_overall_f1": 0.8132353632361465,
+      "eval_overall_precision": 0.8138473840171838,
+      "eval_overall_recall": 0.8126242622540194,
+      "eval_runtime": 953.7502,
+      "eval_samples_per_second": 68.886,
+      "eval_steps_per_second": 0.269,
+      "step": 33000
+    },
+    {
+      "epoch": 1.51,
+      "grad_norm": 15.319729804992676,
+      "learning_rate": 1.569935996357105e-05,
+      "loss": 0.2709,
+      "step": 34000
+    },
+    {
+      "epoch": 1.51,
+      "eval_LOC_f1": 0.832774509183695,
+      "eval_ORG_f1": 0.7316936984844457,
+      "eval_PER_f1": 0.8539732494099136,
+      "eval_loss": 0.2914768159389496,
+      "eval_overall_accuracy": 0.9130203451152948,
+      "eval_overall_f1": 0.8107029247351679,
+      "eval_overall_precision": 0.7998217523118878,
+      "eval_overall_recall": 0.8218842462858306,
+      "eval_runtime": 894.9182,
+      "eval_samples_per_second": 73.415,
+      "eval_steps_per_second": 0.287,
+      "step": 34000
+    },
+    {
+      "epoch": 1.55,
+      "grad_norm": 1.931920051574707,
+      "learning_rate": 1.5572870550734906e-05,
+      "loss": 0.2631,
+      "step": 35000
+    },
+    {
+      "epoch": 1.55,
+      "eval_LOC_f1": 0.8323614548810673,
+      "eval_ORG_f1": 0.7279775567457282,
+      "eval_PER_f1": 0.8522675037838443,
+      "eval_loss": 0.3124816417694092,
+      "eval_overall_accuracy": 0.9096829247297835,
+      "eval_overall_f1": 0.8079843932416348,
+      "eval_overall_precision": 0.7857095311702623,
+      "eval_overall_recall": 0.8315590901263366,
+      "eval_runtime": 886.0992,
+      "eval_samples_per_second": 74.145,
+      "eval_steps_per_second": 0.29,
+      "step": 35000
+    },
+    {
+      "epoch": 1.59,
+      "grad_norm": 2.4540863037109375,
+      "learning_rate": 1.544638113789876e-05,
+      "loss": 0.2684,
+      "step": 36000
+    },
+    {
+      "epoch": 1.59,
+      "eval_LOC_f1": 0.8353469255313396,
+      "eval_ORG_f1": 0.743517370545253,
+      "eval_PER_f1": 0.8544175455688603,
+      "eval_loss": 0.31003931164741516,
+      "eval_overall_accuracy": 0.9140255023922472,
+      "eval_overall_f1": 0.8147700607298496,
+      "eval_overall_precision": 0.8114867383067271,
+      "eval_overall_recall": 0.8180800601155346,
+      "eval_runtime": 886.9365,
+      "eval_samples_per_second": 74.075,
+      "eval_steps_per_second": 0.29,
+      "step": 36000
+    },
+    {
+      "epoch": 1.64,
+      "grad_norm": 1.3480443954467773,
+      "learning_rate": 1.5319891725062616e-05,
+      "loss": 0.2546,
+      "step": 37000
+    },
+    {
+      "epoch": 1.64,
+      "eval_LOC_f1": 0.8268920250802105,
+      "eval_ORG_f1": 0.7359205250232403,
+      "eval_PER_f1": 0.8566762684569846,
+      "eval_loss": 0.3172565698623657,
+      "eval_overall_accuracy": 0.9102766656432092,
+      "eval_overall_f1": 0.8115168704156479,
+      "eval_overall_precision": 0.8111456076827428,
+      "eval_overall_recall": 0.8118884731593531,
+      "eval_runtime": 886.1348,
+      "eval_samples_per_second": 74.142,
+      "eval_steps_per_second": 0.29,
+      "step": 37000
+    },
+    {
+      "epoch": 1.68,
+      "grad_norm": 2.66180419921875,
+      "learning_rate": 1.5193402312226468e-05,
+      "loss": 0.2642,
+      "step": 38000
+    },
+    {
+      "epoch": 1.68,
+      "eval_LOC_f1": 0.8459391601383606,
+      "eval_ORG_f1": 0.7362593503366764,
+      "eval_PER_f1": 0.8645872824401172,
+      "eval_loss": 0.2804827094078064,
+      "eval_overall_accuracy": 0.913356527184997,
+      "eval_overall_f1": 0.8191614534186092,
+      "eval_overall_precision": 0.8128202954617264,
+      "eval_overall_recall": 0.8256023294769635,
+      "eval_runtime": 901.1127,
+      "eval_samples_per_second": 72.91,
+      "eval_steps_per_second": 0.285,
+      "step": 38000
+    },
+    {
+      "epoch": 1.73,
+      "grad_norm": 47.826175689697266,
+      "learning_rate": 1.5066912899390323e-05,
+      "loss": 0.2776,
+      "step": 39000
+    },
+    {
+      "epoch": 1.73,
+      "eval_LOC_f1": 0.8417204029165086,
+      "eval_ORG_f1": 0.7461313828771049,
+      "eval_PER_f1": 0.8621870343195805,
+      "eval_loss": 0.2955803871154785,
+      "eval_overall_accuracy": 0.9141780043391887,
+      "eval_overall_f1": 0.8213932893138981,
+      "eval_overall_precision": 0.8195021231836067,
+      "eval_overall_recall": 0.8232932041266809,
+      "eval_runtime": 949.1162,
+      "eval_samples_per_second": 69.222,
+      "eval_steps_per_second": 0.271,
+      "step": 39000
+    },
+    {
+      "epoch": 1.77,
+      "grad_norm": 14.713150024414062,
+      "learning_rate": 1.4940423486554176e-05,
+      "loss": 0.2616,
+      "step": 40000
+    },
+    {
+      "epoch": 1.77,
+      "eval_LOC_f1": 0.8312933303965682,
+      "eval_ORG_f1": 0.738299968952903,
+      "eval_PER_f1": 0.8513267743278481,
+      "eval_loss": 0.29292425513267517,
+      "eval_overall_accuracy": 0.9104176452208262,
+      "eval_overall_f1": 0.8118819476942669,
+      "eval_overall_precision": 0.796034420507883,
+      "eval_overall_recall": 0.8283732798973026,
+      "eval_runtime": 908.7427,
+      "eval_samples_per_second": 72.298,
+      "eval_steps_per_second": 0.283,
+      "step": 40000
+    },
+    {
+      "epoch": 1.82,
+      "grad_norm": 16.184900283813477,
+      "learning_rate": 1.481393407371803e-05,
+      "loss": 0.2701,
+      "step": 41000
+    },
+    {
+      "epoch": 1.82,
+      "eval_LOC_f1": 0.8392716598242965,
+      "eval_ORG_f1": 0.74401776384535,
+      "eval_PER_f1": 0.8639952804501724,
+      "eval_loss": 0.2767677903175354,
+      "eval_overall_accuracy": 0.9163774213073009,
+      "eval_overall_f1": 0.8195078963845922,
+      "eval_overall_precision": 0.8094401856885441,
+      "eval_overall_recall": 0.8298292029995147,
+      "eval_runtime": 885.7415,
+      "eval_samples_per_second": 74.175,
+      "eval_steps_per_second": 0.29,
+      "step": 41000
+    },
+    {
+      "epoch": 1.86,
+      "grad_norm": 16.436620712280273,
+      "learning_rate": 1.4687444660881885e-05,
+      "loss": 0.2669,
+      "step": 42000
+    },
+    {
+      "epoch": 1.86,
+      "eval_LOC_f1": 0.8361988121287902,
+      "eval_ORG_f1": 0.7500968409804315,
+      "eval_PER_f1": 0.8611851501962505,
+      "eval_loss": 0.29421770572662354,
+      "eval_overall_accuracy": 0.9147995344962793,
+      "eval_overall_f1": 0.819935938895562,
+      "eval_overall_precision": 0.8067546477976939,
+      "eval_overall_recall": 0.833555113734208,
+      "eval_runtime": 886.1039,
+      "eval_samples_per_second": 74.145,
+      "eval_steps_per_second": 0.29,
+      "step": 42000
+    },
+    {
+      "epoch": 1.9,
+      "grad_norm": 4.060434341430664,
+      "learning_rate": 1.456095524804574e-05,
+      "loss": 0.2422,
+      "step": 43000
+    },
+    {
+      "epoch": 1.9,
+      "eval_LOC_f1": 0.8396598172309967,
+      "eval_ORG_f1": 0.752934357339516,
+      "eval_PER_f1": 0.8587656968190062,
+      "eval_loss": 0.29513150453567505,
+      "eval_overall_accuracy": 0.9147182001245772,
+      "eval_overall_f1": 0.8206534155814486,
+      "eval_overall_precision": 0.8120531232517684,
+      "eval_overall_recall": 0.8294378258215007,
+      "eval_runtime": 885.9872,
+      "eval_samples_per_second": 74.155,
+      "eval_steps_per_second": 0.29,
+      "step": 43000
+    },
+    {
+      "epoch": 1.95,
+      "grad_norm": 4.084081172943115,
+      "learning_rate": 1.4434465835209595e-05,
+      "loss": 0.2616,
+      "step": 44000
+    },
+    {
+      "epoch": 1.95,
+      "eval_LOC_f1": 0.8452747626229368,
+      "eval_ORG_f1": 0.7507735621040889,
+      "eval_PER_f1": 0.8679754713527367,
+      "eval_loss": 0.29186713695526123,
+      "eval_overall_accuracy": 0.915251618045657,
+      "eval_overall_f1": 0.8252959748971241,
+      "eval_overall_precision": 0.820965230928905,
+      "eval_overall_recall": 0.8296726521283091,
+      "eval_runtime": 886.49,
+      "eval_samples_per_second": 74.113,
+      "eval_steps_per_second": 0.29,
+      "step": 44000
+    },
+    {
+      "epoch": 1.99,
+      "grad_norm": 12.051443099975586,
+      "learning_rate": 1.4307976422373449e-05,
+      "loss": 0.2449,
+      "step": 45000
+    },
+    {
+      "epoch": 1.99,
+      "eval_LOC_f1": 0.8420882739030321,
+      "eval_ORG_f1": 0.7511743283897188,
+      "eval_PER_f1": 0.8660442600276625,
+      "eval_loss": 0.28106340765953064,
+      "eval_overall_accuracy": 0.9165787238772637,
+      "eval_overall_f1": 0.8232545031821703,
+      "eval_overall_precision": 0.823325400056368,
+      "eval_overall_recall": 0.8231836185168371,
+      "eval_runtime": 931.6188,
+      "eval_samples_per_second": 70.522,
+      "eval_steps_per_second": 0.276,
+      "step": 45000
+    },
+    {
+      "epoch": 2.04,
+      "grad_norm": 16.095355987548828,
+      "learning_rate": 1.4181487009537302e-05,
+      "loss": 0.2379,
+      "step": 46000
+    },
+    {
+      "epoch": 2.04,
+      "eval_LOC_f1": 0.8334692878701362,
+      "eval_ORG_f1": 0.7499450670182377,
+      "eval_PER_f1": 0.8676157711285138,
+      "eval_loss": 0.2910194396972656,
+      "eval_overall_accuracy": 0.914796145564125,
+      "eval_overall_f1": 0.821648434727601,
+      "eval_overall_precision": 0.8099514821518198,
+      "eval_overall_recall": 0.8336881819747327,
+      "eval_runtime": 955.4465,
+      "eval_samples_per_second": 68.764,
+      "eval_steps_per_second": 0.269,
+      "step": 46000
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 6.1998419761657715,
+      "learning_rate": 1.4054997596701157e-05,
+      "loss": 0.2128,
+      "step": 47000
+    },
+    {
+      "epoch": 2.08,
+      "eval_LOC_f1": 0.8394146138221968,
+      "eval_ORG_f1": 0.7394133361546803,
+      "eval_PER_f1": 0.864682724271338,
+      "eval_loss": 0.30839666724205017,
+      "eval_overall_accuracy": 0.9148042790012952,
+      "eval_overall_f1": 0.8188789651986448,
+      "eval_overall_precision": 0.8056906504249807,
+      "eval_overall_recall": 0.8325062228971304,
+      "eval_runtime": 901.3734,
+      "eval_samples_per_second": 72.889,
+      "eval_steps_per_second": 0.285,
+      "step": 47000
+    },
+    {
+      "epoch": 2.13,
+      "grad_norm": 5.403193950653076,
+      "learning_rate": 1.3928508183865012e-05,
+      "loss": 0.2237,
+      "step": 48000
+    },
+    {
+      "epoch": 2.13,
+      "eval_LOC_f1": 0.8372631513660468,
+      "eval_ORG_f1": 0.7524156839779593,
+      "eval_PER_f1": 0.8648470673721019,
+      "eval_loss": 0.3043561279773712,
+      "eval_overall_accuracy": 0.9152800850757528,
+      "eval_overall_f1": 0.8220857007666829,
+      "eval_overall_precision": 0.8082709895080826,
+      "eval_overall_recall": 0.836380856959469,
+      "eval_runtime": 884.5053,
+      "eval_samples_per_second": 74.279,
+      "eval_steps_per_second": 0.291,
+      "step": 48000
+    },
+    {
+      "epoch": 2.17,
+      "grad_norm": 13.765303611755371,
+      "learning_rate": 1.3802018771028867e-05,
+      "loss": 0.2246,
+      "step": 49000
+    },
+    {
+      "epoch": 2.17,
+      "eval_LOC_f1": 0.8349439826902872,
+      "eval_ORG_f1": 0.7424747298710351,
+      "eval_PER_f1": 0.859314059653789,
+      "eval_loss": 0.28388652205467224,
+      "eval_overall_accuracy": 0.9144118406578324,
+      "eval_overall_f1": 0.817645207294658,
+      "eval_overall_precision": 0.8118686576378439,
+      "eval_overall_recall": 0.8235045478028086,
+      "eval_runtime": 886.1908,
+      "eval_samples_per_second": 74.138,
+      "eval_steps_per_second": 0.29,
+      "step": 49000
+    },
+    {
+      "epoch": 2.21,
+      "grad_norm": 26.609722137451172,
+      "learning_rate": 1.3675529358192721e-05,
+      "loss": 0.2231,
+      "step": 50000
+    },
+    {
+      "epoch": 2.21,
+      "eval_LOC_f1": 0.8453938301706774,
+      "eval_ORG_f1": 0.7530178399743618,
+      "eval_PER_f1": 0.8633811603243918,
+      "eval_loss": 0.30370599031448364,
+      "eval_overall_accuracy": 0.9166329467917318,
+      "eval_overall_f1": 0.8246776205110672,
+      "eval_overall_precision": 0.8187134051793966,
+      "eval_overall_recall": 0.8307293705089469,
+      "eval_runtime": 885.5464,
+      "eval_samples_per_second": 74.191,
+      "eval_steps_per_second": 0.29,
+      "step": 50000
+    },
+    {
+      "epoch": 2.26,
+      "grad_norm": 18.287857055664062,
+      "learning_rate": 1.3549039945356574e-05,
+      "loss": 0.2156,
+      "step": 51000
+    },
+    {
+      "epoch": 2.26,
+      "eval_LOC_f1": 0.8369090369642839,
+      "eval_ORG_f1": 0.7570827451034141,
+      "eval_PER_f1": 0.8699436414871374,
+      "eval_loss": 0.2922073304653168,
+      "eval_overall_accuracy": 0.916367932297269,
+      "eval_overall_f1": 0.8256348807545127,
+      "eval_overall_precision": 0.8155410977732979,
+      "eval_overall_recall": 0.8359816522378947,
+      "eval_runtime": 885.263,
+      "eval_samples_per_second": 74.215,
+      "eval_steps_per_second": 0.29,
+      "step": 51000
+    },
+    {
+      "epoch": 2.3,
+      "grad_norm": 35.76387405395508,
+      "learning_rate": 1.3422550532520428e-05,
+      "loss": 0.2279,
+      "step": 52000
+    },
+    {
+      "epoch": 2.3,
+      "eval_LOC_f1": 0.8493380871850663,
+      "eval_ORG_f1": 0.7652859960552268,
+      "eval_PER_f1": 0.8658015544747966,
+      "eval_loss": 0.30765289068222046,
+      "eval_overall_accuracy": 0.9169528619870936,
+      "eval_overall_f1": 0.8303520832274882,
+      "eval_overall_precision": 0.8291563575626546,
+      "eval_overall_recall": 0.8315512625827762,
+      "eval_runtime": 939.0823,
+      "eval_samples_per_second": 69.962,
+      "eval_steps_per_second": 0.274,
+      "step": 52000
+    },
+    {
+      "epoch": 2.35,
+      "grad_norm": 12.871335983276367,
+      "learning_rate": 1.3296061119684283e-05,
+      "loss": 0.2192,
+      "step": 53000
+    },
+    {
+      "epoch": 2.35,
+      "eval_LOC_f1": 0.8450333357909482,
+      "eval_ORG_f1": 0.7589152754918096,
+      "eval_PER_f1": 0.8706159740642501,
+      "eval_loss": 0.29916831851005554,
+      "eval_overall_accuracy": 0.9182454007107268,
+      "eval_overall_f1": 0.8283387559440156,
+      "eval_overall_precision": 0.8151347746682732,
+      "eval_overall_recall": 0.8419775506050691,
+      "eval_runtime": 927.2359,
+      "eval_samples_per_second": 70.856,
+      "eval_steps_per_second": 0.277,
+      "step": 53000
+    },
+    {
+      "epoch": 2.39,
+      "grad_norm": 12.074441909790039,
+      "learning_rate": 1.3169571706848138e-05,
+      "loss": 0.2199,
+      "step": 54000
+    },
+    {
+      "epoch": 2.39,
+      "eval_LOC_f1": 0.845725804758205,
+      "eval_ORG_f1": 0.7583743578767123,
+      "eval_PER_f1": 0.8723780235920504,
+      "eval_loss": 0.29886308312416077,
+      "eval_overall_accuracy": 0.9192946141056846,
+      "eval_overall_f1": 0.8293654188671028,
+      "eval_overall_precision": 0.8241158649684679,
+      "eval_overall_recall": 0.8346822800068883,
+      "eval_runtime": 890.1649,
+      "eval_samples_per_second": 73.807,
+      "eval_steps_per_second": 0.289,
+      "step": 54000
+    },
+    {
+      "epoch": 2.43,
+      "grad_norm": 23.149980545043945,
+      "learning_rate": 1.3043082294011993e-05,
+      "loss": 0.2255,
+      "step": 55000
+    },
+    {
+      "epoch": 2.43,
+      "eval_LOC_f1": 0.8466151994355207,
+      "eval_ORG_f1": 0.7544473410506125,
+      "eval_PER_f1": 0.870403734801872,
+      "eval_loss": 0.2841680943965912,
+      "eval_overall_accuracy": 0.9179688638469395,
+      "eval_overall_f1": 0.8283511691203761,
+      "eval_overall_precision": 0.8183908572825472,
+      "eval_overall_recall": 0.8385569140692268,
+      "eval_runtime": 885.5261,
+      "eval_samples_per_second": 74.193,
+      "eval_steps_per_second": 0.29,
+      "step": 55000
+    },
+    {
+      "epoch": 2.48,
+      "grad_norm": 9.740825653076172,
+      "learning_rate": 1.2916592881175847e-05,
+      "loss": 0.2166,
+      "step": 56000
+    },
+    {
+      "epoch": 2.48,
+      "eval_LOC_f1": 0.8564384031559538,
+      "eval_ORG_f1": 0.7616027673681177,
+      "eval_PER_f1": 0.8733214429549507,
+      "eval_loss": 0.2920551300048828,
+      "eval_overall_accuracy": 0.9202286038073975,
+      "eval_overall_f1": 0.8339421536254372,
+      "eval_overall_precision": 0.8307068573159461,
+      "eval_overall_recall": 0.8372027490332984,
+      "eval_runtime": 885.9705,
+      "eval_samples_per_second": 74.156,
+      "eval_steps_per_second": 0.29,
+      "step": 56000
+    },
+    {
+      "epoch": 2.52,
+      "grad_norm": 7.81465482711792,
+      "learning_rate": 1.27901034683397e-05,
+      "loss": 0.2195,
+      "step": 57000
+    },
+    {
+      "epoch": 2.52,
+      "eval_LOC_f1": 0.8524520572659642,
+      "eval_ORG_f1": 0.7654914529914529,
+      "eval_PER_f1": 0.8711133515111243,
+      "eval_loss": 0.2894265651702881,
+      "eval_overall_accuracy": 0.9196877302355783,
+      "eval_overall_f1": 0.8334080883643471,
+      "eval_overall_precision": 0.8305412821928031,
+      "eval_overall_recall": 0.836294753980306,
+      "eval_runtime": 886.104,
+      "eval_samples_per_second": 74.145,
+      "eval_steps_per_second": 0.29,
+      "step": 57000
+    },
+    {
+      "epoch": 2.57,
+      "grad_norm": 4.749297618865967,
+      "learning_rate": 1.2663614055503555e-05,
+      "loss": 0.2198,
+      "step": 58000
+    },
+    {
+      "epoch": 2.57,
+      "eval_LOC_f1": 0.8479476339833629,
+      "eval_ORG_f1": 0.7569071497897121,
+      "eval_PER_f1": 0.8715211159515157,
+      "eval_loss": 0.2978798449039459,
+      "eval_overall_accuracy": 0.917242276793067,
+      "eval_overall_f1": 0.8290111404616975,
+      "eval_overall_precision": 0.8113011029852536,
+      "eval_overall_recall": 0.847511623902187,
+      "eval_runtime": 912.438,
+      "eval_samples_per_second": 72.005,
+      "eval_steps_per_second": 0.282,
+      "step": 58000
+    },
+    {
+      "epoch": 2.61,
+      "grad_norm": 8.958308219909668,
+      "learning_rate": 1.2537124642667409e-05,
+      "loss": 0.2186,
+      "step": 59000
+    },
+    {
+      "epoch": 2.61,
+      "eval_LOC_f1": 0.8410615339749197,
+      "eval_ORG_f1": 0.7623071419893903,
+      "eval_PER_f1": 0.8698487455846391,
+      "eval_loss": 0.2916683554649353,
+      "eval_overall_accuracy": 0.9165563569250457,
+      "eval_overall_f1": 0.828324104278644,
+      "eval_overall_precision": 0.8190848632805027,
+      "eval_overall_recall": 0.8377741597131988,
+      "eval_runtime": 949.686,
+      "eval_samples_per_second": 69.181,
+      "eval_steps_per_second": 0.271,
+      "step": 59000
+    },
+    {
+      "epoch": 2.66,
+      "grad_norm": 2.5494885444641113,
+      "learning_rate": 1.2410635229831265e-05,
+      "loss": 0.2105,
+      "step": 60000
+    },
+    {
+      "epoch": 2.66,
+      "eval_LOC_f1": 0.8486577670408396,
+      "eval_ORG_f1": 0.7658020018726344,
+      "eval_PER_f1": 0.8764428548203543,
+      "eval_loss": 0.28897759318351746,
+      "eval_overall_accuracy": 0.9185185486423599,
+      "eval_overall_f1": 0.8335781872027352,
+      "eval_overall_precision": 0.8256759558603319,
+      "eval_overall_recall": 0.8416331386884168,
+      "eval_runtime": 906.7935,
+      "eval_samples_per_second": 72.453,
+      "eval_steps_per_second": 0.283,
+      "step": 60000
+    },
+    {
+      "epoch": 2.7,
+      "grad_norm": 2.0383992195129395,
+      "learning_rate": 1.2284145816995119e-05,
+      "loss": 0.2117,
+      "step": 61000
+    },
+    {
+      "epoch": 2.7,
+      "eval_LOC_f1": 0.8508689748097309,
+      "eval_ORG_f1": 0.7658278739306382,
+      "eval_PER_f1": 0.8731916232956014,
+      "eval_loss": 0.28091031312942505,
+      "eval_overall_accuracy": 0.9196301183889559,
+      "eval_overall_f1": 0.833681650059079,
+      "eval_overall_precision": 0.8200686015431561,
+      "eval_overall_recall": 0.8477542777525557,
+      "eval_runtime": 885.0192,
+      "eval_samples_per_second": 74.236,
+      "eval_steps_per_second": 0.29,
+      "step": 61000
+    },
+    {
+      "epoch": 2.74,
+      "grad_norm": 1.6501883268356323,
+      "learning_rate": 1.2157656404158972e-05,
+      "loss": 0.1994,
+      "step": 62000
+    },
+    {
+      "epoch": 2.74,
+      "eval_LOC_f1": 0.8548620423851409,
+      "eval_ORG_f1": 0.7666146057733736,
+      "eval_PER_f1": 0.8740409497434966,
+      "eval_loss": 0.267339825630188,
+      "eval_overall_accuracy": 0.9214282857900041,
+      "eval_overall_f1": 0.835940143844595,
+      "eval_overall_precision": 0.8308500027062763,
+      "eval_overall_recall": 0.8410930381827575,
+      "eval_runtime": 886.6936,
+      "eval_samples_per_second": 74.095,
+      "eval_steps_per_second": 0.29,
+      "step": 62000
+    },
+    {
+      "epoch": 2.79,
+      "grad_norm": 3.6416823863983154,
+      "learning_rate": 1.2031166991322827e-05,
+      "loss": 0.2075,
+      "step": 63000
+    },
+    {
+      "epoch": 2.79,
+      "eval_LOC_f1": 0.8514531524204939,
+      "eval_ORG_f1": 0.7741699019900634,
+      "eval_PER_f1": 0.8762775257778839,
+      "eval_loss": 0.2862880229949951,
+      "eval_overall_accuracy": 0.9219467924096053,
+      "eval_overall_f1": 0.8376946930582835,
+      "eval_overall_precision": 0.8390993269298734,
+      "eval_overall_recall": 0.836294753980306,
+      "eval_runtime": 883.1341,
+      "eval_samples_per_second": 74.394,
+      "eval_steps_per_second": 0.291,
+      "step": 63000
+    },
+    {
+      "epoch": 2.83,
+      "grad_norm": 3.767646312713623,
+      "learning_rate": 1.1904677578486681e-05,
+      "loss": 0.2144,
+      "step": 64000
+    },
+    {
+      "epoch": 2.83,
+      "eval_LOC_f1": 0.8542264412564663,
+      "eval_ORG_f1": 0.762882333169584,
+      "eval_PER_f1": 0.8737852991218755,
+      "eval_loss": 0.2919914424419403,
+      "eval_overall_accuracy": 0.9192736027263282,
+      "eval_overall_f1": 0.8335330637616842,
+      "eval_overall_precision": 0.8234296712697055,
+      "eval_overall_recall": 0.8438874712337774,
+      "eval_runtime": 882.9236,
+      "eval_samples_per_second": 74.412,
+      "eval_steps_per_second": 0.291,
+      "step": 64000
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 1.2373511791229248,
+      "learning_rate": 1.1778188165650538e-05,
+      "loss": 0.2107,
+      "step": 65000
+    },
+    {
+      "epoch": 2.88,
+      "eval_LOC_f1": 0.8536771728748805,
+      "eval_ORG_f1": 0.7691218130311614,
+      "eval_PER_f1": 0.8777075297286194,
+      "eval_loss": 0.2956686019897461,
+      "eval_overall_accuracy": 0.92124189452152,
+      "eval_overall_f1": 0.8368564609614728,
+      "eval_overall_precision": 0.8285232067510548,
+      "eval_overall_recall": 0.84535904942311,
+      "eval_runtime": 930.5121,
+      "eval_samples_per_second": 70.606,
+      "eval_steps_per_second": 0.276,
+      "step": 65000
+    },
+    {
+      "epoch": 2.92,
+      "grad_norm": 13.250840187072754,
+      "learning_rate": 1.1651698752814391e-05,
+      "loss": 0.2133,
+      "step": 66000
+    },
+    {
+      "epoch": 2.92,
+      "eval_LOC_f1": 0.8533000763334159,
+      "eval_ORG_f1": 0.7736413979491799,
+      "eval_PER_f1": 0.8740390436699679,
+      "eval_loss": 0.2793155908584595,
+      "eval_overall_accuracy": 0.9226618570941534,
+      "eval_overall_f1": 0.8369070216139791,
+      "eval_overall_precision": 0.8351781983723613,
+      "eval_overall_recall": 0.8386430170483898,
+      "eval_runtime": 953.5624,
+      "eval_samples_per_second": 68.9,
+      "eval_steps_per_second": 0.27,
+      "step": 66000
+    },
+    {
+      "epoch": 2.97,
+      "grad_norm": 25.819507598876953,
+      "learning_rate": 1.1525209339978245e-05,
+      "loss": 0.2112,
+      "step": 67000
+    },
+    {
+      "epoch": 2.97,
+      "eval_LOC_f1": 0.8548619072433559,
+      "eval_ORG_f1": 0.7661784507158363,
+      "eval_PER_f1": 0.8776364551402296,
+      "eval_loss": 0.2820794880390167,
+      "eval_overall_accuracy": 0.9220823496957755,
+      "eval_overall_f1": 0.8374375390381013,
+      "eval_overall_precision": 0.8353114340451381,
+      "eval_overall_recall": 0.8395744947320631,
+      "eval_runtime": 899.3376,
+      "eval_samples_per_second": 73.054,
+      "eval_steps_per_second": 0.286,
+      "step": 67000
+    },
+    {
+      "epoch": 3.01,
+      "grad_norm": 13.493629455566406,
+      "learning_rate": 1.13987199271421e-05,
+      "loss": 0.1983,
+      "step": 68000
+    },
+    {
+      "epoch": 3.01,
+      "eval_LOC_f1": 0.8558231253148143,
+      "eval_ORG_f1": 0.7679850431851696,
+      "eval_PER_f1": 0.8760919620026149,
+      "eval_loss": 0.2852949798107147,
+      "eval_overall_accuracy": 0.9224734324663767,
+      "eval_overall_f1": 0.8365820844153812,
+      "eval_overall_precision": 0.8359935591789517,
+      "eval_overall_recall": 0.8371714388590572,
+      "eval_runtime": 881.1865,
+      "eval_samples_per_second": 74.559,
+      "eval_steps_per_second": 0.292,
+      "step": 68000
     }
   ],
   "logging_steps": 1000,
@@ -344,7 +1510,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 7,
   "save_steps": 500,
-  "total_flos": 4767734068578096.0,
+  "total_flos": 2.1809439865622904e+16,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null