Training in progress, epoch 24, checkpoint

last-checkpoint/global_step4928/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a57a9a23ef7f096514a7c23cd3b6dd9320c85781d700c34b4b22f2c8df92a75
+size 2506176112

last-checkpoint/global_step4928/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:021f57078c9044b2309021b4fc20bdc862c548034ebebb7a1e2c77c8eaae5d09
+size 2506176112

last-checkpoint/global_step4928/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19eaeba11c0ceb2532ddfd732a74b89769b0567f5c2480d566b19b72c9fd6a80
+size 2506176112

last-checkpoint/global_step4928/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:ee5c93da09fbfeb06fc19b3cef49ab91db0e725035444d6cdb96809b8b9eadcb
+size 2506176112

last-checkpoint/global_step4928/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:c00ffef8eefa8f8358ed1c6f61d654b20c27eddbfe0dd82830a9f20e16757f85
+size 2506176112

last-checkpoint/global_step4928/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:c10d54de0cd548e613485ee0e69da7d813ee80b2317569160958e2e4574aa72b
+size 2506176112

last-checkpoint/global_step4928/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:626e1b55e0f1d6f3ce34dcade96d7fba40f0099032be2cd622d3062d9fca2e0e
+size 2506176112

last-checkpoint/global_step4928/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:eb896c66c8c21d39353a007d4c2ac348fc8f8b10134c68b5313b9e83069fd695
+size 2506176112

last-checkpoint/global_step4928/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84194c554e1ba6e218f3d9df508f289e525895d9d44dbdbafac31ef0e28aa543
+size 85570

last-checkpoint/global_step4928/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca9ecac1033adac02f1e567dee0188e38f517d7c2675ace09591182ee269512c
+size 85506

last-checkpoint/global_step4928/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77c04da8f7351dc76a8b4e781d08009e5e26ec0026e8092fb4066e4ee4c0c017
+size 85506

last-checkpoint/global_step4928/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52caac51b5ab59ce3fdf3092538f84272c220a80081cc21f22f230699d5152d1
+size 85506

last-checkpoint/global_step4928/zero_pp_rank_4_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15beab398fd58f2a629c4c60d768da129efbcfc290c667ba33bdf6fd1245d530
+size 85506

last-checkpoint/global_step4928/zero_pp_rank_5_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc28cb444e8b65d936deba8cb0b52f072349842d1308d11e82c4aed0769e1934
+size 85506

last-checkpoint/global_step4928/zero_pp_rank_6_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:350b51963f22fad0c545de65665496d3aeea7062c1ab553cc246600f06963c6a
+size 85506

last-checkpoint/global_step4928/zero_pp_rank_7_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76b41ed59377277c212831b6ecb28e620914d57efeefc205f64c9c84a7c32c26
+size 85506

last-checkpoint/latest

@@ -1 +1 @@
-global_step4731
+global_step4928

last-checkpoint/model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c0fa0a03d21a0ff6b27eafad6ffcf212788b879afb1bb9961a5a085f6a29e4a
+oid sha256:b9da2b7dcfa00a88a71ba39f69087d6c106b9fc3502f080f4fdf0c95ba7f0232
 size 4945242264

last-checkpoint/model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8b10af69a716125a68d0b4044a64aaba3019edc45d92d331681815aa21d3cbf
+oid sha256:66272c39e510a03fae4a8b8c2051165d69a38abcede58508cbcdf95984247125
 size 67121608

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de1e7081b4b7ecbe80a0244c117feb164a6edd889266c562a7b550e4bdb35eb2
+oid sha256:6118faeaee4fdfaa13ac70fa72cfa116964f5a3b96fac7723d08df58a38b397c
 size 15984

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:abeabad207497496517e104a739bcf618c55f0f5f24ab84280d26e5a138129b9
+oid sha256:4dfa684751cc5b0d0c6c20837671c3f615d1dea15a14f377cd3f328d98685be6
 size 15984

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cbfeffde7de6926c029ae5ce615a2206c5f1f7b6d3ba91813e5f083de4672d94
+oid sha256:f6288d8200e463b5ae8ad06a62ffa8cc379f48c86fd01515a13181f976efa71a
 size 15984

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9fe308403a8a8a99e9433e74b57e10647861653f3902c0d37d8ffd57267868ef
+oid sha256:bd3d0484a9201a8efbd51326113ef6b44a0b272cfa40529d39c56a9d67deac8e
 size 15984

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5559dcd2db39330772a1e035b47ab90a38c1fad876c0cc77298cb150d7c792b0
+oid sha256:426c6fc3fcaae6fad4fdb83d268d91ed94e3e0e88bed483e185a78e86a9692b3
 size 15984

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5e601f488ab4088a4b5f2b2eaf592f26d9277d818c0ccce4b09ffc93af9b671
+oid sha256:7d089430782ecc3b7446f527601b569a4990fdd9f8d96c18c87dc8d503cbdb70
 size 15984

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:948cff0d258b1f24171ea15c9fd1db43d638e0bf5726b3462453de79d891e501
+oid sha256:b61f73069ccc96351c7253c2334d20c170e92750457b685c9d79286d288292e9
 size 15984

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e1437c26fc61851ea1c51c8dbde816a8f21ba5abf80812e00135e1e418db89c
+oid sha256:e70ddcc8f9ba6c0be1ac8d1ba2f06e8a4253e0a843f3a263a85c073416ca948d
 size 15984

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7afe10276133c7ad9d69e4ce5223f876b6c12764d68dc657b0c96f5afe271edc
+oid sha256:cc0dea9c7426022052c6904d669ed6537aacbe6f69c00de710577f64daa74c6b
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 24.0,
+  "epoch": 24.99746835443038,
   "eval_steps": 100,
-  "global_step": 4731,
+  "global_step": 4928,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -7117,6 +7117,291 @@
       "rewards/margins": 0.5577437877655029,
       "rewards/rejected": -0.3512403070926666,
       "step": 4730
+    },
+    {
+      "epoch": 24.045569620253165,
+      "grad_norm": 627241.9523500776,
+      "learning_rate": 1.8426825446568473e-07,
+      "logits/chosen": -0.28766584396362305,
+      "logits/rejected": -0.6269916296005249,
+      "logps/chosen": -44.65516662597656,
+      "logps/rejected": -577.6954345703125,
+      "loss": 12950.232,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.20441746711730957,
+      "rewards/margins": 0.5374675393104553,
+      "rewards/rejected": -0.33305004239082336,
+      "step": 4740
+    },
+    {
+      "epoch": 24.09620253164557,
+      "grad_norm": 347781.48168387014,
+      "learning_rate": 1.8348480100282043e-07,
+      "logits/chosen": -0.7881828546524048,
+      "logits/rejected": 0.06337795406579971,
+      "logps/chosen": -27.82383155822754,
+      "logps/rejected": -579.8468017578125,
+      "loss": 12986.6266,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.2079242467880249,
+      "rewards/margins": 0.5530039668083191,
+      "rewards/rejected": -0.3450797498226166,
+      "step": 4750
+    },
+    {
+      "epoch": 24.146835443037975,
+      "grad_norm": 263636.0742414822,
+      "learning_rate": 1.827013475399561e-07,
+      "logits/chosen": -1.2877452373504639,
+      "logits/rejected": -0.24622194468975067,
+      "logps/chosen": -30.940113067626953,
+      "logps/rejected": -566.517333984375,
+      "loss": 12716.3281,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": 0.20519797503948212,
+      "rewards/margins": 0.5418139696121216,
+      "rewards/rejected": -0.33661606907844543,
+      "step": 4760
+    },
+    {
+      "epoch": 24.19746835443038,
+      "grad_norm": 343217.0970891512,
+      "learning_rate": 1.8191789407709182e-07,
+      "logits/chosen": -1.0294177532196045,
+      "logits/rejected": -0.4815802574157715,
+      "logps/chosen": -31.3253231048584,
+      "logps/rejected": -560.1129150390625,
+      "loss": 13408.0094,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/chosen": 0.19244466722011566,
+      "rewards/margins": 0.5293484926223755,
+      "rewards/rejected": -0.3369038701057434,
+      "step": 4770
+    },
+    {
+      "epoch": 24.248101265822786,
+      "grad_norm": 504214.9301429895,
+      "learning_rate": 1.811344406142275e-07,
+      "logits/chosen": -0.604789137840271,
+      "logits/rejected": 0.22590136528015137,
+      "logps/chosen": -41.899864196777344,
+      "logps/rejected": -555.2015380859375,
+      "loss": 13531.9625,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/chosen": 0.190764918923378,
+      "rewards/margins": 0.5197011232376099,
+      "rewards/rejected": -0.32893624901771545,
+      "step": 4780
+    },
+    {
+      "epoch": 24.29873417721519,
+      "grad_norm": 376293.61129873747,
+      "learning_rate": 1.803509871513632e-07,
+      "logits/chosen": -0.3547658324241638,
+      "logits/rejected": -0.13969659805297852,
+      "logps/chosen": -28.85129737854004,
+      "logps/rejected": -577.5880126953125,
+      "loss": 12898.8125,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.19991345703601837,
+      "rewards/margins": 0.5475735664367676,
+      "rewards/rejected": -0.3476601243019104,
+      "step": 4790
+    },
+    {
+      "epoch": 24.349367088607593,
+      "grad_norm": 414135.66871189536,
+      "learning_rate": 1.7956753368849888e-07,
+      "logits/chosen": 0.09798486530780792,
+      "logits/rejected": 0.8055311441421509,
+      "logps/chosen": -34.26675796508789,
+      "logps/rejected": -584.9435424804688,
+      "loss": 12507.7328,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.20852184295654297,
+      "rewards/margins": 0.5590152144432068,
+      "rewards/rejected": -0.3504934012889862,
+      "step": 4800
+    },
+    {
+      "epoch": 24.4,
+      "grad_norm": 343912.9213203651,
+      "learning_rate": 1.787840802256346e-07,
+      "logits/chosen": -0.12874791026115417,
+      "logits/rejected": 0.15529172122478485,
+      "logps/chosen": -32.1009635925293,
+      "logps/rejected": -574.7623291015625,
+      "loss": 13372.5594,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": 0.1985681653022766,
+      "rewards/margins": 0.546955943107605,
+      "rewards/rejected": -0.34838777780532837,
+      "step": 4810
+    },
+    {
+      "epoch": 24.450632911392404,
+      "grad_norm": 697134.5088322331,
+      "learning_rate": 1.7800062676277027e-07,
+      "logits/chosen": 1.3246450424194336,
+      "logits/rejected": 1.760595679283142,
+      "logps/chosen": -32.29949188232422,
+      "logps/rejected": -575.516357421875,
+      "loss": 12415.9672,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": 0.19931714236736298,
+      "rewards/margins": 0.5466721057891846,
+      "rewards/rejected": -0.3473549485206604,
+      "step": 4820
+    },
+    {
+      "epoch": 24.50126582278481,
+      "grad_norm": 389676.90431780973,
+      "learning_rate": 1.7721717329990597e-07,
+      "logits/chosen": -1.3381322622299194,
+      "logits/rejected": -0.6404735445976257,
+      "logps/chosen": -34.635719299316406,
+      "logps/rejected": -587.3080444335938,
+      "loss": 13101.6227,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.20362886786460876,
+      "rewards/margins": 0.5546956658363342,
+      "rewards/rejected": -0.3510667383670807,
+      "step": 4830
+    },
+    {
+      "epoch": 24.551898734177215,
+      "grad_norm": 314600.5637340995,
+      "learning_rate": 1.7643371983704165e-07,
+      "logits/chosen": 0.9953921437263489,
+      "logits/rejected": 0.9643779993057251,
+      "logps/chosen": -30.646331787109375,
+      "logps/rejected": -570.855712890625,
+      "loss": 12974.9633,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/chosen": 0.19645583629608154,
+      "rewards/margins": 0.5343093872070312,
+      "rewards/rejected": -0.3378535211086273,
+      "step": 4840
+    },
+    {
+      "epoch": 24.60253164556962,
+      "grad_norm": 327013.6839426029,
+      "learning_rate": 1.7565026637417739e-07,
+      "logits/chosen": -0.7217426300048828,
+      "logits/rejected": -0.7290517091751099,
+      "logps/chosen": -37.666648864746094,
+      "logps/rejected": -563.5865478515625,
+      "loss": 13273.0266,
+      "rewards/accuracies": 0.9624999761581421,
+      "rewards/chosen": 0.197954460978508,
+      "rewards/margins": 0.5282526612281799,
+      "rewards/rejected": -0.33029812574386597,
+      "step": 4850
+    },
+    {
+      "epoch": 24.653164556962025,
+      "grad_norm": 425662.97201424866,
+      "learning_rate": 1.7486681291131307e-07,
+      "logits/chosen": -0.32900291681289673,
+      "logits/rejected": 0.18864622712135315,
+      "logps/chosen": -32.36582565307617,
+      "logps/rejected": -566.5547485351562,
+      "loss": 13350.5594,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/chosen": 0.19610336422920227,
+      "rewards/margins": 0.5321984887123108,
+      "rewards/rejected": -0.3360951244831085,
+      "step": 4860
+    },
+    {
+      "epoch": 24.70379746835443,
+      "grad_norm": 402351.7657170625,
+      "learning_rate": 1.7408335944844877e-07,
+      "logits/chosen": -2.112635850906372,
+      "logits/rejected": -1.4337832927703857,
+      "logps/chosen": -36.27765655517578,
+      "logps/rejected": -588.708740234375,
+      "loss": 13690.0375,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.20721419155597687,
+      "rewards/margins": 0.5532296299934387,
+      "rewards/rejected": -0.34601545333862305,
+      "step": 4870
+    },
+    {
+      "epoch": 24.754430379746836,
+      "grad_norm": 357871.7969812476,
+      "learning_rate": 1.7329990598558445e-07,
+      "logits/chosen": -0.34599849581718445,
+      "logits/rejected": 0.0005200624582357705,
+      "logps/chosen": -28.50592041015625,
+      "logps/rejected": -560.6458740234375,
+      "loss": 13133.4,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": 0.1971847414970398,
+      "rewards/margins": 0.5326961278915405,
+      "rewards/rejected": -0.33551135659217834,
+      "step": 4880
+    },
+    {
+      "epoch": 24.80506329113924,
+      "grad_norm": 610016.4055058825,
+      "learning_rate": 1.7251645252272016e-07,
+      "logits/chosen": -1.225339651107788,
+      "logits/rejected": -0.8243592977523804,
+      "logps/chosen": -34.17732620239258,
+      "logps/rejected": -581.1436767578125,
+      "loss": 12571.1,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.20419716835021973,
+      "rewards/margins": 0.5520066022872925,
+      "rewards/rejected": -0.347809374332428,
+      "step": 4890
+    },
+    {
+      "epoch": 24.855696202531647,
+      "grad_norm": 379732.159808034,
+      "learning_rate": 1.7173299905985584e-07,
+      "logits/chosen": -1.817439317703247,
+      "logits/rejected": -1.4895861148834229,
+      "logps/chosen": -30.775598526000977,
+      "logps/rejected": -570.484130859375,
+      "loss": 13212.3781,
+      "rewards/accuracies": 0.9750000238418579,
+      "rewards/chosen": 0.20220620930194855,
+      "rewards/margins": 0.5392990112304688,
+      "rewards/rejected": -0.3370928466320038,
+      "step": 4900
+    },
+    {
+      "epoch": 24.906329113924052,
+      "grad_norm": 324902.93726753094,
+      "learning_rate": 1.7094954559699154e-07,
+      "logits/chosen": -0.6947388648986816,
+      "logits/rejected": -0.4560522437095642,
+      "logps/chosen": -40.327796936035156,
+      "logps/rejected": -581.7532348632812,
+      "loss": 12891.9016,
+      "rewards/accuracies": 0.987500011920929,
+      "rewards/chosen": 0.20357458293437958,
+      "rewards/margins": 0.5383836030960083,
+      "rewards/rejected": -0.33480900526046753,
+      "step": 4910
+    },
+    {
+      "epoch": 24.956962025316457,
+      "grad_norm": 294429.7500390618,
+      "learning_rate": 1.7016609213412722e-07,
+      "logits/chosen": -0.9121583104133606,
+      "logits/rejected": 0.40684938430786133,
+      "logps/chosen": -28.22664451599121,
+      "logps/rejected": -580.9795532226562,
+      "loss": 13184.5812,
+      "rewards/accuracies": 1.0,
+      "rewards/chosen": 0.20078356564044952,
+      "rewards/margins": 0.5547462701797485,
+      "rewards/rejected": -0.3539626896381378,
+      "step": 4920
     }
   ],
   "logging_steps": 10,