Training in progress, step 2096, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6858b7cacf8da273466148c63ce37c609bc72964df895f364ba999052d67cb02
+oid sha256:c1a2ebef4c2fa62ef083acf0283325484beb150e65ae9598b4f8e6997ab30d55
 size 80792880

last-checkpoint/global_step2096/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab09b5694eefe7786e712dcf5ef338807f00cac67b1a8440d22511c7e4545758
+size 60558832

last-checkpoint/global_step2096/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca9160ed484219503a480434fbf574571e9d5f1d720b21c5e2a0b9dfb85dd935
+size 60558832

last-checkpoint/global_step2096/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5109c80b3b772cfe133a4e232d0b1f6517a21abd3ce07abeb99f9168a5fcbfc0
+size 60558832

last-checkpoint/global_step2096/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83f0652215971f9411505e7ff6276df7910126de6633c2aaf02ff448702d5509
+size 60558832

last-checkpoint/global_step2096/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05414cb558c1e89c7599d8b422c0e345864d88b77a36f695e9914d1d15c0df3b
+size 60558832

last-checkpoint/global_step2096/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31a7547d2d3244cb3bbc140314965514ad539d6f70c7faa1acb1900a4a9aac9d
+size 60558832

last-checkpoint/global_step2096/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f70b71f3f411d80968f7911ca95e802eea151f860d81e428b05f518e6e4adb9
+size 60558832

last-checkpoint/global_step2096/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fabbe1238e434b478f60b7683cb4ad2201bf72030aad2b56611a0d100ad4180f
+size 60558832

last-checkpoint/global_step2096/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c56be51a75f861a6c1061df855e1c773b43908a82058ceae617ede592760334b
+size 390038

last-checkpoint/global_step2096/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fba77dce12ffc36f83474659416adc00392fd3cc22aa4809c0fc78cd41a6ee70
+size 390038

last-checkpoint/global_step2096/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ebb449d60f3d1684c4ffe693dbb7576d2a14869d26db198000fa7cc2a248216
+size 390038

last-checkpoint/global_step2096/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eae9aea2a7e808c932e611a4588fb8843e462e2bf40d55f8736a53bb22223393
+size 390038

last-checkpoint/global_step2096/zero_pp_rank_4_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28c16c282fc532c92df24aae3e94d6a192f31d3d3a9dc48bcdff32f0b9c34e9d
+size 390038

last-checkpoint/global_step2096/zero_pp_rank_5_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fc633928076b503a85f121de6cc6f5cc0f95ff8e1b0a1d15ed02a5b870bed57
+size 390038

last-checkpoint/global_step2096/zero_pp_rank_6_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1b9d4e33cc40d6590ad04654584253598b7503359e7747e703703a00d5e1c57
+size 390038

last-checkpoint/global_step2096/zero_pp_rank_7_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5c6de32f80e510731cf621e564d4fbaee82e9d81f5a5e02e37e05a8a81c6fa6
+size 390038

last-checkpoint/latest

@@ -1 +1 @@
-global_step2000
+global_step2096

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0cc5c0f1dd8463e5c58f14317bdebb3c5833d534b938627ce207723dd0b7e7f8
+oid sha256:3db20ad821c0bd6ef3820f3ee9a15e66a9f5ac2f7ddd41a44dca6d3296d1ee8d
 size 15984

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:214889c1db734b7da2083b307acb5ad3efffdf3efd538ba8a9398c61c9d624a6
+oid sha256:d8b3f45ae1458240bdc90980406976b3d3415ee685fe0bbc0c9bd438b38d4cc3
 size 15984

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b3661b41e96ef0c059587fa2433b349cb391639e7cc40b210b54dd421754a42
+oid sha256:4468d58f1c2f91dc50e131aecf9444c837c55f89d91de410985021a52c7b9d3c
 size 15984

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a00a0cf3269503efee297bfaa4a0177630fe8d2ab36cdc010884706c6927e2b
+oid sha256:653ffb9d097b24cf1819a32c53fd01044f4b6df3a39d0d1a41f49d567234e621
 size 15984

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a077c3964abf442ddbb13bd61d8f22de1f48735ef155591994eb411ae01e53a
+oid sha256:37e56dc24ba4e2baf45c62e54eda0982ce52c5db9637972a21bf3ccebc89a7bf
 size 15984

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a15004e563ff52373d189f7cc7ee16f283f740197d25f3a08a8fc58e0691c661
+oid sha256:9307ebbf541138419e4ea948891c68bec3ee50264969db8d7ddd9e305b6fae68
 size 15984

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0acbe321c626ff5100b5ee5e3eb751e22685631c6f9939d317f95d717a87875e
+oid sha256:8056cfa8e1113bd740eada2f62b9f13713a8fcff8af18eff52163e7abdf8d26e
 size 15984

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:56e0664b3713d71bc66950e282d440ee23a20cdb6e0d052e4fedc92caf00a354
+oid sha256:4f63a80cbfc65d36cb0a6c908bb11d8b46ac29771d0b22925818162c7defb651
 size 15984

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f4da5e5d48f3657b3469ffe99496799aa1e0c96ebacbde9407ba10230a28ebf
+oid sha256:f43f1b551d131f2e15824dafdf308fab750fa72bbedcb53092f970029438269d
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.9539709038874314,
+  "epoch": 0.9997615072740281,
   "eval_steps": 5,
-  "global_step": 2000,
+  "global_step": 2096,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -20824,6 +20824,1001 @@
       "eval_samples_per_second": 3.243,
       "eval_steps_per_second": 0.147,
       "step": 2000
+    },
+    {
+      "epoch": 0.9544478893393752,
+      "grad_norm": 0.46363001931995473,
+      "learning_rate": 6.247368302631396e-07,
+      "loss": 0.1768,
+      "step": 2001
+    },
+    {
+      "epoch": 0.9549248747913188,
+      "grad_norm": 0.5202493640608246,
+      "learning_rate": 6.116804407097165e-07,
+      "loss": 0.2037,
+      "step": 2002
+    },
+    {
+      "epoch": 0.9554018602432626,
+      "grad_norm": 0.5187833918929314,
+      "learning_rate": 5.987610890004159e-07,
+      "loss": 0.2036,
+      "step": 2003
+    },
+    {
+      "epoch": 0.9558788456952063,
+      "grad_norm": 0.5920049724547812,
+      "learning_rate": 5.859788109825793e-07,
+      "loss": 0.2256,
+      "step": 2004
+    },
+    {
+      "epoch": 0.9563558311471501,
+      "grad_norm": 0.49628985797291797,
+      "learning_rate": 5.733336421232027e-07,
+      "loss": 0.1671,
+      "step": 2005
+    },
+    {
+      "epoch": 0.9563558311471501,
+      "eval_PRM Accuracy": 0.9117647058823529,
+      "eval_PRM F1": 0.9387755102040817,
+      "eval_PRM F1 AUC": 0.8863958431743032,
+      "eval_PRM F1 AUC (fixed)": 0.9740198393953707,
+      "eval_PRM F1 Neg": 0.8421052631578947,
+      "eval_PRM NPV": 0.8571428571428571,
+      "eval_PRM Precision": 0.9324324324324325,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8275862068965517,
+      "eval_loss": 0.1853582262992859,
+      "eval_runtime": 9.4911,
+      "eval_samples_per_second": 4.636,
+      "eval_steps_per_second": 0.211,
+      "step": 2005
+    },
+    {
+      "epoch": 0.9568328165990937,
+      "grad_norm": 0.46054756457888824,
+      "learning_rate": 5.608256175088699e-07,
+      "loss": 0.1388,
+      "step": 2006
+    },
+    {
+      "epoch": 0.9573098020510374,
+      "grad_norm": 0.6037004635730433,
+      "learning_rate": 5.484547718456024e-07,
+      "loss": 0.1962,
+      "step": 2007
+    },
+    {
+      "epoch": 0.9577867875029812,
+      "grad_norm": 0.7908605629178659,
+      "learning_rate": 5.362211394588201e-07,
+      "loss": 0.2143,
+      "step": 2008
+    },
+    {
+      "epoch": 0.9582637729549248,
+      "grad_norm": 1.2406054827420188,
+      "learning_rate": 5.241247542931815e-07,
+      "loss": 0.3142,
+      "step": 2009
+    },
+    {
+      "epoch": 0.9587407584068686,
+      "grad_norm": 0.4741604532893723,
+      "learning_rate": 5.121656499125604e-07,
+      "loss": 0.202,
+      "step": 2010
+    },
+    {
+      "epoch": 0.9587407584068686,
+      "eval_PRM Accuracy": 0.9215686274509803,
+      "eval_PRM F1": 0.9452054794520548,
+      "eval_PRM F1 AUC": 0.9036372224846481,
+      "eval_PRM F1 AUC (fixed)": 0.9730751062824752,
+      "eval_PRM F1 Neg": 0.8620689655172413,
+      "eval_PRM NPV": 0.8620689655172413,
+      "eval_PRM Precision": 0.9452054794520548,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8620689655172413,
+      "eval_loss": 0.1857965588569641,
+      "eval_runtime": 14.288,
+      "eval_samples_per_second": 3.08,
+      "eval_steps_per_second": 0.14,
+      "step": 2010
+    },
+    {
+      "epoch": 0.9592177438588123,
+      "grad_norm": 0.5934312404017598,
+      "learning_rate": 5.003438594998911e-07,
+      "loss": 0.2037,
+      "step": 2011
+    },
+    {
+      "epoch": 0.9596947293107561,
+      "grad_norm": 0.5667813315735417,
+      "learning_rate": 4.88659415857129e-07,
+      "loss": 0.2542,
+      "step": 2012
+    },
+    {
+      "epoch": 0.9601717147626997,
+      "grad_norm": 0.6066873845783456,
+      "learning_rate": 4.771123514050957e-07,
+      "loss": 0.1754,
+      "step": 2013
+    },
+    {
+      "epoch": 0.9606487002146434,
+      "grad_norm": 0.45844057179588416,
+      "learning_rate": 4.6570269818346224e-07,
+      "loss": 0.1702,
+      "step": 2014
+    },
+    {
+      "epoch": 0.9611256856665872,
+      "grad_norm": 0.8627638637774376,
+      "learning_rate": 4.5443048785061007e-07,
+      "loss": 0.1902,
+      "step": 2015
+    },
+    {
+      "epoch": 0.9611256856665872,
+      "eval_PRM Accuracy": 0.9117647058823529,
+      "eval_PRM F1": 0.9387755102040817,
+      "eval_PRM F1 AUC": 0.8863958431743032,
+      "eval_PRM F1 AUC (fixed)": 0.9742560226735947,
+      "eval_PRM F1 Neg": 0.8421052631578947,
+      "eval_PRM NPV": 0.8571428571428571,
+      "eval_PRM Precision": 0.9324324324324325,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8275862068965517,
+      "eval_loss": 0.1854802966117859,
+      "eval_runtime": 13.5861,
+      "eval_samples_per_second": 3.239,
+      "eval_steps_per_second": 0.147,
+      "step": 2015
+    },
+    {
+      "epoch": 0.9616026711185309,
+      "grad_norm": 0.5948796848880178,
+      "learning_rate": 4.432957516835534e-07,
+      "loss": 0.1965,
+      "step": 2016
+    },
+    {
+      "epoch": 0.9620796565704746,
+      "grad_norm": 0.6019788555452265,
+      "learning_rate": 4.3229852057785626e-07,
+      "loss": 0.2276,
+      "step": 2017
+    },
+    {
+      "epoch": 0.9625566420224183,
+      "grad_norm": 0.4742278466628085,
+      "learning_rate": 4.2143882504756537e-07,
+      "loss": 0.1158,
+      "step": 2018
+    },
+    {
+      "epoch": 0.963033627474362,
+      "grad_norm": 0.6747526262962028,
+      "learning_rate": 4.107166952250885e-07,
+      "loss": 0.1962,
+      "step": 2019
+    },
+    {
+      "epoch": 0.9635106129263058,
+      "grad_norm": 0.6246184175068561,
+      "learning_rate": 4.0013216086113305e-07,
+      "loss": 0.168,
+      "step": 2020
+    },
+    {
+      "epoch": 0.9635106129263058,
+      "eval_PRM Accuracy": 0.9117647058823529,
+      "eval_PRM F1": 0.9387755102040817,
+      "eval_PRM F1 AUC": 0.8863958431743032,
+      "eval_PRM F1 AUC (fixed)": 0.9744922059518186,
+      "eval_PRM F1 Neg": 0.8421052631578947,
+      "eval_PRM NPV": 0.8571428571428571,
+      "eval_PRM Precision": 0.9324324324324325,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8275862068965517,
+      "eval_loss": 0.18552468717098236,
+      "eval_runtime": 14.5016,
+      "eval_samples_per_second": 3.034,
+      "eval_steps_per_second": 0.138,
+      "step": 2020
+    },
+    {
+      "epoch": 0.9639875983782494,
+      "grad_norm": 0.47683381284661164,
+      "learning_rate": 3.8968525132462296e-07,
+      "loss": 0.157,
+      "step": 2021
+    },
+    {
+      "epoch": 0.9644645838301932,
+      "grad_norm": 0.6119940362045991,
+      "learning_rate": 3.793759956026099e-07,
+      "loss": 0.1978,
+      "step": 2022
+    },
+    {
+      "epoch": 0.9649415692821369,
+      "grad_norm": 0.6456948862949443,
+      "learning_rate": 3.692044223002178e-07,
+      "loss": 0.2491,
+      "step": 2023
+    },
+    {
+      "epoch": 0.9654185547340806,
+      "grad_norm": 0.4486105863495388,
+      "learning_rate": 3.5917055964050373e-07,
+      "loss": 0.1459,
+      "step": 2024
+    },
+    {
+      "epoch": 0.9658955401860243,
+      "grad_norm": 0.9606332285377831,
+      "learning_rate": 3.4927443546445283e-07,
+      "loss": 0.256,
+      "step": 2025
+    },
+    {
+      "epoch": 0.9658955401860243,
+      "eval_PRM Accuracy": 0.9117647058823529,
+      "eval_PRM F1": 0.9387755102040817,
+      "eval_PRM F1 AUC": 0.8863958431743032,
+      "eval_PRM F1 AUC (fixed)": 0.9749645725082665,
+      "eval_PRM F1 Neg": 0.8421052631578947,
+      "eval_PRM NPV": 0.8571428571428571,
+      "eval_PRM Precision": 0.9324324324324325,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8275862068965517,
+      "eval_loss": 0.1847589612007141,
+      "eval_runtime": 13.8736,
+      "eval_samples_per_second": 3.171,
+      "eval_steps_per_second": 0.144,
+      "step": 2025
+    },
+    {
+      "epoch": 0.966372525637968,
+      "grad_norm": 0.49145718044168607,
+      "learning_rate": 3.395160772308559e-07,
+      "loss": 0.1963,
+      "step": 2026
+    },
+    {
+      "epoch": 0.9668495110899118,
+      "grad_norm": 0.46304627488343814,
+      "learning_rate": 3.2989551201624835e-07,
+      "loss": 0.1743,
+      "step": 2027
+    },
+    {
+      "epoch": 0.9673264965418554,
+      "grad_norm": 0.4999044809958458,
+      "learning_rate": 3.20412766514816e-07,
+      "loss": 0.2058,
+      "step": 2028
+    },
+    {
+      "epoch": 0.9678034819937992,
+      "grad_norm": 0.6597864406972287,
+      "learning_rate": 3.11067867038356e-07,
+      "loss": 0.2058,
+      "step": 2029
+    },
+    {
+      "epoch": 0.9682804674457429,
+      "grad_norm": 0.7226947390131475,
+      "learning_rate": 3.0186083951616595e-07,
+      "loss": 0.2703,
+      "step": 2030
+    },
+    {
+      "epoch": 0.9682804674457429,
+      "eval_PRM Accuracy": 0.9215686274509803,
+      "eval_PRM F1": 0.9452054794520548,
+      "eval_PRM F1 AUC": 0.9036372224846481,
+      "eval_PRM F1 AUC (fixed)": 0.9737836561171469,
+      "eval_PRM F1 Neg": 0.8620689655172413,
+      "eval_PRM NPV": 0.8620689655172413,
+      "eval_PRM Precision": 0.9452054794520548,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8620689655172413,
+      "eval_loss": 0.1856023669242859,
+      "eval_runtime": 14.3153,
+      "eval_samples_per_second": 3.074,
+      "eval_steps_per_second": 0.14,
+      "step": 2030
+    },
+    {
+      "epoch": 0.9687574528976867,
+      "grad_norm": 0.7232729910764847,
+      "learning_rate": 2.927917094950106e-07,
+      "loss": 0.205,
+      "step": 2031
+    },
+    {
+      "epoch": 0.9692344383496303,
+      "grad_norm": 0.5531092451895251,
+      "learning_rate": 2.838605021390106e-07,
+      "loss": 0.1434,
+      "step": 2032
+    },
+    {
+      "epoch": 0.969711423801574,
+      "grad_norm": 0.4689768889718397,
+      "learning_rate": 2.7506724222959854e-07,
+      "loss": 0.1638,
+      "step": 2033
+    },
+    {
+      "epoch": 0.9701884092535178,
+      "grad_norm": 0.6236593320179458,
+      "learning_rate": 2.6641195416545196e-07,
+      "loss": 0.2312,
+      "step": 2034
+    },
+    {
+      "epoch": 0.9706653947054615,
+      "grad_norm": 0.7551093880824219,
+      "learning_rate": 2.5789466196241007e-07,
+      "loss": 0.2462,
+      "step": 2035
+    },
+    {
+      "epoch": 0.9706653947054615,
+      "eval_PRM Accuracy": 0.9117647058823529,
+      "eval_PRM F1": 0.9387755102040817,
+      "eval_PRM F1 AUC": 0.8863958431743032,
+      "eval_PRM F1 AUC (fixed)": 0.9728389230042513,
+      "eval_PRM F1 Neg": 0.8421052631578947,
+      "eval_PRM NPV": 0.8571428571428571,
+      "eval_PRM Precision": 0.9324324324324325,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8275862068965517,
+      "eval_loss": 0.18592418730258942,
+      "eval_runtime": 13.4713,
+      "eval_samples_per_second": 3.266,
+      "eval_steps_per_second": 0.148,
+      "step": 2035
+    },
+    {
+      "epoch": 0.9711423801574052,
+      "grad_norm": 0.39148638193092083,
+      "learning_rate": 2.49515389253413e-07,
+      "loss": 0.1402,
+      "step": 2036
+    },
+    {
+      "epoch": 0.9716193656093489,
+      "grad_norm": 0.6837049878509137,
+      "learning_rate": 2.41274159288446e-07,
+      "loss": 0.2142,
+      "step": 2037
+    },
+    {
+      "epoch": 0.9720963510612927,
+      "grad_norm": 0.6015805169989217,
+      "learning_rate": 2.3317099493445627e-07,
+      "loss": 0.2208,
+      "step": 2038
+    },
+    {
+      "epoch": 0.9725733365132363,
+      "grad_norm": 0.4847482121424171,
+      "learning_rate": 2.2520591867531415e-07,
+      "loss": 0.1811,
+      "step": 2039
+    },
+    {
+      "epoch": 0.97305032196518,
+      "grad_norm": 0.59574749022119,
+      "learning_rate": 2.173789526117187e-07,
+      "loss": 0.1843,
+      "step": 2040
+    },
+    {
+      "epoch": 0.97305032196518,
+      "eval_PRM Accuracy": 0.9215686274509803,
+      "eval_PRM F1": 0.9452054794520548,
+      "eval_PRM F1 AUC": 0.9036372224846481,
+      "eval_PRM F1 AUC (fixed)": 0.9749645725082665,
+      "eval_PRM F1 Neg": 0.8620689655172413,
+      "eval_PRM NPV": 0.8620689655172413,
+      "eval_PRM Precision": 0.9452054794520548,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8620689655172413,
+      "eval_loss": 0.18514181673526764,
+      "eval_runtime": 13.4517,
+      "eval_samples_per_second": 3.271,
+      "eval_steps_per_second": 0.149,
+      "step": 2040
+    },
+    {
+      "epoch": 0.9735273074171238,
+      "grad_norm": 0.5741048128836703,
+      "learning_rate": 2.0969011846116992e-07,
+      "loss": 0.2053,
+      "step": 2041
+    },
+    {
+      "epoch": 0.9740042928690675,
+      "grad_norm": 0.5915322378907251,
+      "learning_rate": 2.021394375578911e-07,
+      "loss": 0.208,
+      "step": 2042
+    },
+    {
+      "epoch": 0.9744812783210112,
+      "grad_norm": 0.49623775659482605,
+      "learning_rate": 1.9472693085276218e-07,
+      "loss": 0.1829,
+      "step": 2043
+    },
+    {
+      "epoch": 0.9749582637729549,
+      "grad_norm": 0.8456364278385442,
+      "learning_rate": 1.8745261891327526e-07,
+      "loss": 0.3239,
+      "step": 2044
+    },
+    {
+      "epoch": 0.9754352492248987,
+      "grad_norm": 0.5170849611642481,
+      "learning_rate": 1.8031652192347925e-07,
+      "loss": 0.2037,
+      "step": 2045
+    },
+    {
+      "epoch": 0.9754352492248987,
+      "eval_PRM Accuracy": 0.9117647058823529,
+      "eval_PRM F1": 0.9387755102040817,
+      "eval_PRM F1 AUC": 0.8863958431743032,
+      "eval_PRM F1 AUC (fixed)": 0.9747283892300425,
+      "eval_PRM F1 Neg": 0.8421052631578947,
+      "eval_PRM NPV": 0.8571428571428571,
+      "eval_PRM Precision": 0.9324324324324325,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8275862068965517,
+      "eval_loss": 0.18571333587169647,
+      "eval_runtime": 13.6831,
+      "eval_samples_per_second": 3.216,
+      "eval_steps_per_second": 0.146,
+      "step": 2045
+    },
+    {
+      "epoch": 0.9759122346768424,
+      "grad_norm": 0.6496734494944529,
+      "learning_rate": 1.7331865968391314e-07,
+      "loss": 0.1692,
+      "step": 2046
+    },
+    {
+      "epoch": 0.976389220128786,
+      "grad_norm": 0.5594479369809225,
+      "learning_rate": 1.6645905161155606e-07,
+      "loss": 0.1582,
+      "step": 2047
+    },
+    {
+      "epoch": 0.9768662055807298,
+      "grad_norm": 0.7823614945934644,
+      "learning_rate": 1.5973771673976623e-07,
+      "loss": 0.316,
+      "step": 2048
+    },
+    {
+      "epoch": 0.9773431910326735,
+      "grad_norm": 0.43263496968572235,
+      "learning_rate": 1.5315467371824765e-07,
+      "loss": 0.1619,
+      "step": 2049
+    },
+    {
+      "epoch": 0.9778201764846172,
+      "grad_norm": 0.5713807971055265,
+      "learning_rate": 1.4670994081297795e-07,
+      "loss": 0.2004,
+      "step": 2050
+    },
+    {
+      "epoch": 0.9778201764846172,
+      "eval_PRM Accuracy": 0.9117647058823529,
+      "eval_PRM F1": 0.9387755102040817,
+      "eval_PRM F1 AUC": 0.8863958431743032,
+      "eval_PRM F1 AUC (fixed)": 0.9737836561171469,
+      "eval_PRM F1 Neg": 0.8421052631578947,
+      "eval_PRM NPV": 0.8571428571428571,
+      "eval_PRM Precision": 0.9324324324324325,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8275862068965517,
+      "eval_loss": 0.18510852754116058,
+      "eval_runtime": 13.7396,
+      "eval_samples_per_second": 3.202,
+      "eval_steps_per_second": 0.146,
+      "step": 2050
+    },
+    {
+      "epoch": 0.9782971619365609,
+      "grad_norm": 0.6660396602843291,
+      "learning_rate": 1.4040353590616395e-07,
+      "loss": 0.2063,
+      "step": 2051
+    },
+    {
+      "epoch": 0.9787741473885047,
+      "grad_norm": 0.49585433430640596,
+      "learning_rate": 1.3423547649620838e-07,
+      "loss": 0.1868,
+      "step": 2052
+    },
+    {
+      "epoch": 0.9792511328404484,
+      "grad_norm": 0.555056426242968,
+      "learning_rate": 1.2820577969762104e-07,
+      "loss": 0.2221,
+      "step": 2053
+    },
+    {
+      "epoch": 0.979728118292392,
+      "grad_norm": 0.6521947149470015,
+      "learning_rate": 1.2231446224101327e-07,
+      "loss": 0.2179,
+      "step": 2054
+    },
+    {
+      "epoch": 0.9802051037443358,
+      "grad_norm": 0.4868305540978795,
+      "learning_rate": 1.1656154047303691e-07,
+      "loss": 0.1237,
+      "step": 2055
+    },
+    {
+      "epoch": 0.9802051037443358,
+      "eval_PRM Accuracy": 0.9215686274509803,
+      "eval_PRM F1": 0.9452054794520548,
+      "eval_PRM F1 AUC": 0.9036372224846481,
+      "eval_PRM F1 AUC (fixed)": 0.9737836561171469,
+      "eval_PRM F1 Neg": 0.8620689655172413,
+      "eval_PRM NPV": 0.8620689655172413,
+      "eval_PRM Precision": 0.9452054794520548,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8620689655172413,
+      "eval_loss": 0.18632923066616058,
+      "eval_runtime": 13.6312,
+      "eval_samples_per_second": 3.228,
+      "eval_steps_per_second": 0.147,
+      "step": 2055
+    },
+    {
+      "epoch": 0.9806820891962795,
+      "grad_norm": 0.5598366438944131,
+      "learning_rate": 1.1094703035633425e-07,
+      "loss": 0.2018,
+      "step": 2056
+    },
+    {
+      "epoch": 0.9811590746482233,
+      "grad_norm": 0.49904654459510944,
+      "learning_rate": 1.0547094746949371e-07,
+      "loss": 0.1783,
+      "step": 2057
+    },
+    {
+      "epoch": 0.9816360601001669,
+      "grad_norm": 0.5393112228505368,
+      "learning_rate": 1.001333070070054e-07,
+      "loss": 0.1675,
+      "step": 2058
+    },
+    {
+      "epoch": 0.9821130455521107,
+      "grad_norm": 0.578826032904967,
+      "learning_rate": 9.493412377923339e-08,
+      "loss": 0.1948,
+      "step": 2059
+    },
+    {
+      "epoch": 0.9825900310040544,
+      "grad_norm": 0.5392258489199684,
+      "learning_rate": 8.987341221235457e-08,
+      "loss": 0.1956,
+      "step": 2060
+    },
+    {
+      "epoch": 0.9825900310040544,
+      "eval_PRM Accuracy": 0.9215686274509803,
+      "eval_PRM F1": 0.9452054794520548,
+      "eval_PRM F1 AUC": 0.9036372224846481,
+      "eval_PRM F1 AUC (fixed)": 0.9749645725082663,
+      "eval_PRM F1 Neg": 0.8620689655172413,
+      "eval_PRM NPV": 0.8620689655172413,
+      "eval_PRM Precision": 0.9452054794520548,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8620689655172413,
+      "eval_loss": 0.18599630892276764,
+      "eval_runtime": 14.498,
+      "eval_samples_per_second": 3.035,
+      "eval_steps_per_second": 0.138,
+      "step": 2060
+    },
+    {
+      "epoch": 0.9830670164559981,
+      "grad_norm": 0.6616701345184373,
+      "learning_rate": 8.495118634833655e-08,
+      "loss": 0.1429,
+      "step": 2061
+    },
+    {
+      "epoch": 0.9835440019079418,
+      "grad_norm": 0.7130294708477759,
+      "learning_rate": 8.016745984488206e-08,
+      "loss": 0.1512,
+      "step": 2062
+    },
+    {
+      "epoch": 0.9840209873598855,
+      "grad_norm": 0.6976150540958784,
+      "learning_rate": 7.552224597541235e-08,
+      "loss": 0.23,
+      "step": 2063
+    },
+    {
+      "epoch": 0.9844979728118293,
+      "grad_norm": 0.5420611759772007,
+      "learning_rate": 7.101555762900614e-08,
+      "loss": 0.2171,
+      "step": 2064
+    },
+    {
+      "epoch": 0.9849749582637729,
+      "grad_norm": 0.46001325692877904,
+      "learning_rate": 6.664740731037733e-08,
+      "loss": 0.1264,
+      "step": 2065
+    },
+    {
+      "epoch": 0.9849749582637729,
+      "eval_PRM Accuracy": 0.9117647058823529,
+      "eval_PRM F1": 0.9387755102040817,
+      "eval_PRM F1 AUC": 0.8863958431743032,
+      "eval_PRM F1 AUC (fixed)": 0.9749645725082664,
+      "eval_PRM F1 Neg": 0.8421052631578947,
+      "eval_PRM NPV": 0.8571428571428571,
+      "eval_PRM Precision": 0.9324324324324325,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8275862068965517,
+      "eval_loss": 0.1859685778617859,
+      "eval_runtime": 14.424,
+      "eval_samples_per_second": 3.05,
+      "eval_steps_per_second": 0.139,
+      "step": 2065
+    },
+    {
+      "epoch": 0.9854519437157166,
+      "grad_norm": 0.6372943877733731,
+      "learning_rate": 6.241780713983624e-08,
+      "loss": 0.1407,
+      "step": 2066
+    },
+    {
+      "epoch": 0.9859289291676604,
+      "grad_norm": 0.8129639240032717,
+      "learning_rate": 5.8326768853261825e-08,
+      "loss": 0.1824,
+      "step": 2067
+    },
+    {
+      "epoch": 0.9864059146196041,
+      "grad_norm": 0.4987915008637717,
+      "learning_rate": 5.437430380206832e-08,
+      "loss": 0.1687,
+      "step": 2068
+    },
+    {
+      "epoch": 0.9868829000715478,
+      "grad_norm": 0.45563934893068897,
+      "learning_rate": 5.056042295316643e-08,
+      "loss": 0.1195,
+      "step": 2069
+    },
+    {
+      "epoch": 0.9873598855234915,
+      "grad_norm": 0.6072972937744073,
+      "learning_rate": 4.688513688893004e-08,
+      "loss": 0.2232,
+      "step": 2070
+    },
+    {
+      "epoch": 0.9873598855234915,
+      "eval_PRM Accuracy": 0.9117647058823529,
+      "eval_PRM F1": 0.9387755102040817,
+      "eval_PRM F1 AUC": 0.8863958431743032,
+      "eval_PRM F1 AUC (fixed)": 0.9749645725082665,
+      "eval_PRM F1 Neg": 0.8421052631578947,
+      "eval_PRM NPV": 0.8571428571428571,
+      "eval_PRM Precision": 0.9324324324324325,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8275862068965517,
+      "eval_loss": 0.185791015625,
+      "eval_runtime": 13.7396,
+      "eval_samples_per_second": 3.202,
+      "eval_steps_per_second": 0.146,
+      "step": 2070
+    },
+    {
+      "epoch": 0.9878368709754353,
+      "grad_norm": 0.7779254486583322,
+      "learning_rate": 4.334845580718505e-08,
+      "loss": 0.3258,
+      "step": 2071
+    },
+    {
+      "epoch": 0.988313856427379,
+      "grad_norm": 0.5548725465171946,
+      "learning_rate": 3.995038952117058e-08,
+      "loss": 0.2461,
+      "step": 2072
+    },
+    {
+      "epoch": 0.9887908418793226,
+      "grad_norm": 0.6449785112740309,
+      "learning_rate": 3.669094745950008e-08,
+      "loss": 0.1826,
+      "step": 2073
+    },
+    {
+      "epoch": 0.9892678273312664,
+      "grad_norm": 0.58356724744049,
+      "learning_rate": 3.357013866615022e-08,
+      "loss": 0.2221,
+      "step": 2074
+    },
+    {
+      "epoch": 0.9897448127832101,
+      "grad_norm": 0.3475024641711149,
+      "learning_rate": 3.058797180044426e-08,
+      "loss": 0.1303,
+      "step": 2075
+    },
+    {
+      "epoch": 0.9897448127832101,
+      "eval_PRM Accuracy": 0.9215686274509803,
+      "eval_PRM F1": 0.9452054794520548,
+      "eval_PRM F1 AUC": 0.9036372224846481,
+      "eval_PRM F1 AUC (fixed)": 0.9728389230042513,
+      "eval_PRM F1 Neg": 0.8620689655172413,
+      "eval_PRM NPV": 0.8620689655172413,
+      "eval_PRM Precision": 0.9452054794520548,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8620689655172413,
+      "eval_loss": 0.18559126555919647,
+      "eval_runtime": 13.3317,
+      "eval_samples_per_second": 3.3,
+      "eval_steps_per_second": 0.15,
+      "step": 2075
+    },
+    {
+      "epoch": 0.9902217982351539,
+      "grad_norm": 0.41991971109136683,
+      "learning_rate": 2.7744455136990976e-08,
+      "loss": 0.1326,
+      "step": 2076
+    },
+    {
+      "epoch": 0.9906987836870975,
+      "grad_norm": 0.6015953672635286,
+      "learning_rate": 2.503959656571242e-08,
+      "loss": 0.2018,
+      "step": 2077
+    },
+    {
+      "epoch": 0.9911757691390413,
+      "grad_norm": 0.6075786716834879,
+      "learning_rate": 2.2473403591777297e-08,
+      "loss": 0.2541,
+      "step": 2078
+    },
+    {
+      "epoch": 0.991652754590985,
+      "grad_norm": 0.49201559098251413,
+      "learning_rate": 2.004588333560653e-08,
+      "loss": 0.1541,
+      "step": 2079
+    },
+    {
+      "epoch": 0.9921297400429286,
+      "grad_norm": 0.6828738882424833,
+      "learning_rate": 1.7757042532845493e-08,
+      "loss": 0.161,
+      "step": 2080
+    },
+    {
+      "epoch": 0.9921297400429286,
+      "eval_PRM Accuracy": 0.9215686274509803,
+      "eval_PRM F1": 0.9452054794520548,
+      "eval_PRM F1 AUC": 0.9036372224846481,
+      "eval_PRM F1 AUC (fixed)": 0.9740198393953707,
+      "eval_PRM F1 Neg": 0.8620689655172413,
+      "eval_PRM NPV": 0.8620689655172413,
+      "eval_PRM Precision": 0.9452054794520548,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8620689655172413,
+      "eval_loss": 0.1857244372367859,
+      "eval_runtime": 13.6084,
+      "eval_samples_per_second": 3.233,
+      "eval_steps_per_second": 0.147,
+      "step": 2080
+    },
+    {
+      "epoch": 0.9926067254948724,
+      "grad_norm": 0.7375947051401394,
+      "learning_rate": 1.5606887534341807e-08,
+      "loss": 0.2017,
+      "step": 2081
+    },
+    {
+      "epoch": 0.9930837109468161,
+      "grad_norm": 0.6238346413953868,
+      "learning_rate": 1.3595424306139804e-08,
+      "loss": 0.257,
+      "step": 2082
+    },
+    {
+      "epoch": 0.9935606963987599,
+      "grad_norm": 0.4652925076099241,
+      "learning_rate": 1.1722658429441646e-08,
+      "loss": 0.1492,
+      "step": 2083
+    },
+    {
+      "epoch": 0.9940376818507035,
+      "grad_norm": 0.4562941808058073,
+      "learning_rate": 9.988595100612897e-09,
+      "loss": 0.175,
+      "step": 2084
+    },
+    {
+      "epoch": 0.9945146673026473,
+      "grad_norm": 0.3897288159212509,
+      "learning_rate": 8.393239131171404e-09,
+      "loss": 0.1457,
+      "step": 2085
+    },
+    {
+      "epoch": 0.9945146673026473,
+      "eval_PRM Accuracy": 0.9215686274509803,
+      "eval_PRM F1": 0.9452054794520548,
+      "eval_PRM F1 AUC": 0.9036372224846481,
+      "eval_PRM F1 AUC (fixed)": 0.9735474728389231,
+      "eval_PRM F1 Neg": 0.8620689655172413,
+      "eval_PRM NPV": 0.8620689655172413,
+      "eval_PRM Precision": 0.9452054794520548,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8620689655172413,
+      "eval_loss": 0.18607954680919647,
+      "eval_runtime": 14.1612,
+      "eval_samples_per_second": 3.107,
+      "eval_steps_per_second": 0.141,
+      "step": 2085
+    },
+    {
+      "epoch": 0.994991652754591,
+      "grad_norm": 0.7132931890723386,
+      "learning_rate": 6.936594947742902e-09,
+      "loss": 0.2117,
+      "step": 2086
+    },
+    {
+      "epoch": 0.9954686382065348,
+      "grad_norm": 0.6171413474024248,
+      "learning_rate": 5.618666592088761e-09,
+      "loss": 0.2049,
+      "step": 2087
+    },
+    {
+      "epoch": 0.9959456236584784,
+      "grad_norm": 0.3289005124173065,
+      "learning_rate": 4.43945772106158e-09,
+      "loss": 0.1137,
+      "step": 2088
+    },
+    {
+      "epoch": 0.9964226091104221,
+      "grad_norm": 0.694145334823455,
+      "learning_rate": 3.3989716066162856e-09,
+      "loss": 0.2319,
+      "step": 2089
+    },
+    {
+      "epoch": 0.9968995945623659,
+      "grad_norm": 0.5523620831764013,
+      "learning_rate": 2.497211135787936e-09,
+      "loss": 0.1545,
+      "step": 2090
+    },
+    {
+      "epoch": 0.9968995945623659,
+      "eval_PRM Accuracy": 0.9117647058823529,
+      "eval_PRM F1": 0.9387755102040817,
+      "eval_PRM F1 AUC": 0.8863958431743032,
+      "eval_PRM F1 AUC (fixed)": 0.9744922059518186,
+      "eval_PRM F1 Neg": 0.8421052631578947,
+      "eval_PRM NPV": 0.8571428571428571,
+      "eval_PRM Precision": 0.9324324324324325,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8275862068965517,
+      "eval_loss": 0.18575216829776764,
+      "eval_runtime": 14.1793,
+      "eval_samples_per_second": 3.103,
+      "eval_steps_per_second": 0.141,
+      "step": 2090
+    },
+    {
+      "epoch": 0.9973765800143095,
+      "grad_norm": 0.5435729439876591,
+      "learning_rate": 1.7341788106972624e-09,
+      "loss": 0.1771,
+      "step": 2091
+    },
+    {
+      "epoch": 0.9978535654662533,
+      "grad_norm": 0.5245664971355852,
+      "learning_rate": 1.1098767485284712e-09,
+      "loss": 0.1802,
+      "step": 2092
+    },
+    {
+      "epoch": 0.998330550918197,
+      "grad_norm": 0.5755856092820003,
+      "learning_rate": 6.243066815347919e-10,
+      "loss": 0.17,
+      "step": 2093
+    },
+    {
+      "epoch": 0.9988075363701407,
+      "grad_norm": 0.5415094526687191,
+      "learning_rate": 2.7746995702737644e-10,
+      "loss": 0.1711,
+      "step": 2094
+    },
+    {
+      "epoch": 0.9992845218220844,
+      "grad_norm": 0.5985415814340851,
+      "learning_rate": 6.936753737529777e-11,
+      "loss": 0.2475,
+      "step": 2095
+    },
+    {
+      "epoch": 0.9992845218220844,
+      "eval_PRM Accuracy": 0.9117647058823529,
+      "eval_PRM F1": 0.9387755102040817,
+      "eval_PRM F1 AUC": 0.8863958431743032,
+      "eval_PRM F1 AUC (fixed)": 0.9737836561171469,
+      "eval_PRM F1 Neg": 0.8421052631578947,
+      "eval_PRM NPV": 0.8571428571428571,
+      "eval_PRM Precision": 0.9324324324324325,
+      "eval_PRM Recall": 0.9452054794520548,
+      "eval_PRM Specificty": 0.8275862068965517,
+      "eval_loss": 0.18576326966285706,
+      "eval_runtime": 14.0597,
+      "eval_samples_per_second": 3.13,
+      "eval_steps_per_second": 0.142,
+      "step": 2095
+    },
+    {
+      "epoch": 0.9997615072740281,
+      "grad_norm": 0.6252395976260101,
+      "learning_rate": 0.0,
+      "loss": 0.2175,
+      "step": 2096
     }
   ],
   "logging_steps": 1,
@@ -20838,12 +21833,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 2290427323154432.0,
+  "total_flos": 2401999876456448.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null