Training in progress, epoch 2, checkpoint

last-checkpoint/global_step624/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a550a8ce3c87016034f9768ceb86c6a54482d958de85b5c855ac78c4f22972f
+size 15231238316

last-checkpoint/global_step624/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:fa0a96b09bd2259d9a1a76627011bca6cea8c67951d7d013d836053f9ebc242b
+size 15231238316

last-checkpoint/global_step624/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f1f57185dc62af94883cc1db2f7b0c4741cc73ff80aa0f94994f6c1e9bd9d5e
+size 15231238316

last-checkpoint/global_step624/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47c0f7dfacda0e37bf6448e3697ec5d23db8e3c5c869038949aba69c395a9ab6
+size 15231238316

last-checkpoint/global_step624/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb0aa2f97095fad03f427c01efcfbd928e4d1f4294c8f749f1ad9a6a75a60684
+size 167957

last-checkpoint/global_step624/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35686ff96ce67e59c576b8e8b83c3df613780b38ba366171ddc6acffaf75b3ee
+size 167957

last-checkpoint/global_step624/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:daee16d4384d967a1e1613e826997d95b98aa4e484e54721edb88fd48b45bb0a
+size 167957

last-checkpoint/global_step624/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90b045825d1a4bea7e5b809f1d105ec1b341094894088a6663147bc0c19a87ed
+size 167957

last-checkpoint/latest

@@ -1 +1 @@
-global_step312
+global_step624

last-checkpoint/model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b930fd61091d332c38a54ddd0a1a92e8aea22dc60ba309062d8f2bec8a8e43be
+oid sha256:03666b341aea1e882baeabdf705b2bfc652d846442dc2942ffb8428fa4abb414
 size 4877660776

last-checkpoint/model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa7927ec913f0feb59b7ff721a583bf1892df8755cce3ed6d8ccc5135fb6e56e
+oid sha256:def41960df887f693f8e05b14a3ff8c152a1bf898211f3e272cf952d606b758b
 size 4932751008

last-checkpoint/model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb05f961c96d00c29a34add65afe730dc2cacc4dd838daa63b4a7cd4eeb14074
+oid sha256:d3116e484db3429bb3dcd133d60c3ef7735ce0e657a253966dd6473ecdf97dc2
 size 4330865200

last-checkpoint/model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:831c6fc2546570c295e81485edbf2470a9a29cd5b744cbfcf06ea949f58664bb
+oid sha256:6c04996a89919200924220b41f8d07d0cf34bb79035ebff8622bd16ed7cc0275
 size 1089994880

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a6ec98147904582f9073debdd2d46db8be81f1fae8b7e30a4fb33c0724bf01a
+oid sha256:c223685611eb8ca66172e98a5ae4d95e91d096b50c87d5ac51cf3a20f20698f0
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53345026ed0535abd8cf6a0e1ecbdbf9a4d5ef86d37a344cbca5ef2567e6cc18
+oid sha256:07ecb9f981bc1cea06513b87be9861bad9289cd33d5bc89c8c1d83bebf350e5a
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb45d6dbfeb6c6a340bea2c45d80cd63df0bd6b093ca44076be3a394a5390059
+oid sha256:6b78f34e8919a0955c52fa2192994cab8b958695edae102824ebf7fb17e0c11e
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:184eba8fad81b78f577f59c07102dc1253cd2cdbd814ae3d20c579d2332fc55c
+oid sha256:497903fb965e5819817c1bcae461581db9a95d72adb9e86baf55fc2dc3a38323
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:207b0fe8f5401e1ad9120824c6360cf156b298d29d9b08b3a31a3cca0a00431d
+oid sha256:2dbda5999b7120f8da97338911708f83a46f73f3dd722839e09cc5da29ffb94b
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.0,
+  "epoch": 2.0,
   "eval_steps": 500,
-  "global_step": 312,
+  "global_step": 624,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -503,6 +503,487 @@
       "eval_samples_per_second": 10.836,
       "eval_steps_per_second": 0.379,
       "step": 312
+    },
+    {
+      "epoch": 1.0256410256410255,
+      "grad_norm": 31.256761205730616,
+      "learning_rate": 3.6579572446555817e-07,
+      "logits/chosen": -0.353515625,
+      "logits/rejected": -0.349609375,
+      "logps/chosen": -286.0,
+      "logps/rejected": -268.0,
+      "loss": 0.3844,
+      "rewards/accuracies": 0.8374999761581421,
+      "rewards/chosen": 0.1279296875,
+      "rewards/margins": 0.984375,
+      "rewards/rejected": -0.859375,
+      "step": 320
+    },
+    {
+      "epoch": 1.0576923076923077,
+      "grad_norm": 29.12754445870708,
+      "learning_rate": 3.598574821852731e-07,
+      "logits/chosen": -0.451171875,
+      "logits/rejected": -0.47265625,
+      "logps/chosen": -380.0,
+      "logps/rejected": -334.0,
+      "loss": 0.293,
+      "rewards/accuracies": 0.8999999761581421,
+      "rewards/chosen": 0.3984375,
+      "rewards/margins": 1.53125,
+      "rewards/rejected": -1.1328125,
+      "step": 330
+    },
+    {
+      "epoch": 1.0897435897435896,
+      "grad_norm": 23.255524440901077,
+      "learning_rate": 3.5391923990498813e-07,
+      "logits/chosen": -0.388671875,
+      "logits/rejected": -0.296875,
+      "logps/chosen": -350.0,
+      "logps/rejected": -320.0,
+      "loss": 0.3295,
+      "rewards/accuracies": 0.8999999761581421,
+      "rewards/chosen": 0.369140625,
+      "rewards/margins": 1.5625,
+      "rewards/rejected": -1.1953125,
+      "step": 340
+    },
+    {
+      "epoch": 1.1217948717948718,
+      "grad_norm": 32.212012138764614,
+      "learning_rate": 3.479809976247031e-07,
+      "logits/chosen": -0.228515625,
+      "logits/rejected": -0.265625,
+      "logps/chosen": -255.0,
+      "logps/rejected": -268.0,
+      "loss": 0.3649,
+      "rewards/accuracies": 0.8125,
+      "rewards/chosen": 0.07763671875,
+      "rewards/margins": 1.375,
+      "rewards/rejected": -1.296875,
+      "step": 350
+    },
+    {
+      "epoch": 1.1538461538461537,
+      "grad_norm": 21.919087970652495,
+      "learning_rate": 3.42042755344418e-07,
+      "logits/chosen": -0.412109375,
+      "logits/rejected": -0.359375,
+      "logps/chosen": -382.0,
+      "logps/rejected": -326.0,
+      "loss": 0.3203,
+      "rewards/accuracies": 0.8999999761581421,
+      "rewards/chosen": 0.279296875,
+      "rewards/margins": 1.75,
+      "rewards/rejected": -1.46875,
+      "step": 360
+    },
+    {
+      "epoch": 1.185897435897436,
+      "grad_norm": 24.48490434607468,
+      "learning_rate": 3.36104513064133e-07,
+      "logits/chosen": -0.400390625,
+      "logits/rejected": -0.41015625,
+      "logps/chosen": -336.0,
+      "logps/rejected": -318.0,
+      "loss": 0.3216,
+      "rewards/accuracies": 0.8999999761581421,
+      "rewards/chosen": -0.04052734375,
+      "rewards/margins": 1.359375,
+      "rewards/rejected": -1.3984375,
+      "step": 370
+    },
+    {
+      "epoch": 1.217948717948718,
+      "grad_norm": 19.39368867421837,
+      "learning_rate": 3.3016627078384796e-07,
+      "logits/chosen": -0.33984375,
+      "logits/rejected": -0.359375,
+      "logps/chosen": -334.0,
+      "logps/rejected": -304.0,
+      "loss": 0.3092,
+      "rewards/accuracies": 0.800000011920929,
+      "rewards/chosen": 0.1484375,
+      "rewards/margins": 1.484375,
+      "rewards/rejected": -1.3359375,
+      "step": 380
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 31.36015324660143,
+      "learning_rate": 3.2422802850356297e-07,
+      "logits/chosen": -0.369140625,
+      "logits/rejected": -0.337890625,
+      "logps/chosen": -260.0,
+      "logps/rejected": -280.0,
+      "loss": 0.3556,
+      "rewards/accuracies": 0.875,
+      "rewards/chosen": -0.056884765625,
+      "rewards/margins": 1.2421875,
+      "rewards/rejected": -1.3046875,
+      "step": 390
+    },
+    {
+      "epoch": 1.282051282051282,
+      "grad_norm": 22.45941815431359,
+      "learning_rate": 3.182897862232779e-07,
+      "logits/chosen": -0.41796875,
+      "logits/rejected": -0.4765625,
+      "logps/chosen": -262.0,
+      "logps/rejected": -332.0,
+      "loss": 0.3101,
+      "rewards/accuracies": 0.862500011920929,
+      "rewards/chosen": 0.08056640625,
+      "rewards/margins": 1.4375,
+      "rewards/rejected": -1.3515625,
+      "step": 400
+    },
+    {
+      "epoch": 1.314102564102564,
+      "grad_norm": 25.29350114221551,
+      "learning_rate": 3.1235154394299283e-07,
+      "logits/chosen": -0.388671875,
+      "logits/rejected": -0.353515625,
+      "logps/chosen": -284.0,
+      "logps/rejected": -278.0,
+      "loss": 0.3135,
+      "rewards/accuracies": 0.8500000238418579,
+      "rewards/chosen": -0.07373046875,
+      "rewards/margins": 1.4375,
+      "rewards/rejected": -1.515625,
+      "step": 410
+    },
+    {
+      "epoch": 1.3461538461538463,
+      "grad_norm": 21.92318193084022,
+      "learning_rate": 3.0641330166270784e-07,
+      "logits/chosen": -0.384765625,
+      "logits/rejected": -0.462890625,
+      "logps/chosen": -390.0,
+      "logps/rejected": -312.0,
+      "loss": 0.2906,
+      "rewards/accuracies": 0.9375,
+      "rewards/chosen": 0.296875,
+      "rewards/margins": 1.8203125,
+      "rewards/rejected": -1.5234375,
+      "step": 420
+    },
+    {
+      "epoch": 1.3782051282051282,
+      "grad_norm": 23.48433321437472,
+      "learning_rate": 3.004750593824228e-07,
+      "logits/chosen": -0.39453125,
+      "logits/rejected": -0.41796875,
+      "logps/chosen": -304.0,
+      "logps/rejected": -302.0,
+      "loss": 0.3054,
+      "rewards/accuracies": 0.8125,
+      "rewards/chosen": 0.3203125,
+      "rewards/margins": 1.4921875,
+      "rewards/rejected": -1.171875,
+      "step": 430
+    },
+    {
+      "epoch": 1.4102564102564101,
+      "grad_norm": 26.26090008450148,
+      "learning_rate": 2.9453681710213776e-07,
+      "logits/chosen": -0.31640625,
+      "logits/rejected": -0.4375,
+      "logps/chosen": -342.0,
+      "logps/rejected": -300.0,
+      "loss": 0.2742,
+      "rewards/accuracies": 0.925000011920929,
+      "rewards/chosen": 0.298828125,
+      "rewards/margins": 1.734375,
+      "rewards/rejected": -1.4296875,
+      "step": 440
+    },
+    {
+      "epoch": 1.4423076923076923,
+      "grad_norm": 23.659041497927436,
+      "learning_rate": 2.885985748218527e-07,
+      "logits/chosen": -0.37109375,
+      "logits/rejected": -0.359375,
+      "logps/chosen": -384.0,
+      "logps/rejected": -350.0,
+      "loss": 0.2861,
+      "rewards/accuracies": 0.925000011920929,
+      "rewards/chosen": 0.416015625,
+      "rewards/margins": 1.8984375,
+      "rewards/rejected": -1.484375,
+      "step": 450
+    },
+    {
+      "epoch": 1.4743589743589745,
+      "grad_norm": 25.88824077819283,
+      "learning_rate": 2.8266033254156767e-07,
+      "logits/chosen": -0.32421875,
+      "logits/rejected": -0.400390625,
+      "logps/chosen": -306.0,
+      "logps/rejected": -308.0,
+      "loss": 0.2735,
+      "rewards/accuracies": 0.887499988079071,
+      "rewards/chosen": 0.26953125,
+      "rewards/margins": 1.6953125,
+      "rewards/rejected": -1.421875,
+      "step": 460
+    },
+    {
+      "epoch": 1.5064102564102564,
+      "grad_norm": 39.69692113209802,
+      "learning_rate": 2.7672209026128263e-07,
+      "logits/chosen": -0.3828125,
+      "logits/rejected": -0.39453125,
+      "logps/chosen": -332.0,
+      "logps/rejected": -326.0,
+      "loss": 0.2837,
+      "rewards/accuracies": 0.862500011920929,
+      "rewards/chosen": 0.39453125,
+      "rewards/margins": 1.75,
+      "rewards/rejected": -1.359375,
+      "step": 470
+    },
+    {
+      "epoch": 1.5384615384615383,
+      "grad_norm": 30.346903581562415,
+      "learning_rate": 2.7078384798099764e-07,
+      "logits/chosen": -0.34765625,
+      "logits/rejected": -0.447265625,
+      "logps/chosen": -286.0,
+      "logps/rejected": -290.0,
+      "loss": 0.2892,
+      "rewards/accuracies": 0.887499988079071,
+      "rewards/chosen": 0.1240234375,
+      "rewards/margins": 1.5,
+      "rewards/rejected": -1.3828125,
+      "step": 480
+    },
+    {
+      "epoch": 1.5705128205128205,
+      "grad_norm": 26.633915775199014,
+      "learning_rate": 2.648456057007126e-07,
+      "logits/chosen": -0.283203125,
+      "logits/rejected": -0.48828125,
+      "logps/chosen": -302.0,
+      "logps/rejected": -308.0,
+      "loss": 0.2848,
+      "rewards/accuracies": 0.8999999761581421,
+      "rewards/chosen": 0.359375,
+      "rewards/margins": 1.9609375,
+      "rewards/rejected": -1.6015625,
+      "step": 490
+    },
+    {
+      "epoch": 1.6025641025641026,
+      "grad_norm": 28.260367944812693,
+      "learning_rate": 2.589073634204275e-07,
+      "logits/chosen": -0.435546875,
+      "logits/rejected": -0.392578125,
+      "logps/chosen": -324.0,
+      "logps/rejected": -330.0,
+      "loss": 0.2791,
+      "rewards/accuracies": 0.875,
+      "rewards/chosen": 0.328125,
+      "rewards/margins": 1.609375,
+      "rewards/rejected": -1.28125,
+      "step": 500
+    },
+    {
+      "epoch": 1.6346153846153846,
+      "grad_norm": 18.511599498231647,
+      "learning_rate": 2.529691211401425e-07,
+      "logits/chosen": -0.376953125,
+      "logits/rejected": -0.33984375,
+      "logps/chosen": -294.0,
+      "logps/rejected": -272.0,
+      "loss": 0.2808,
+      "rewards/accuracies": 0.8999999761581421,
+      "rewards/chosen": 0.322265625,
+      "rewards/margins": 1.4609375,
+      "rewards/rejected": -1.140625,
+      "step": 510
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 22.844798241441286,
+      "learning_rate": 2.4703087885985747e-07,
+      "logits/chosen": -0.376953125,
+      "logits/rejected": -0.4296875,
+      "logps/chosen": -312.0,
+      "logps/rejected": -318.0,
+      "loss": 0.2799,
+      "rewards/accuracies": 0.887499988079071,
+      "rewards/chosen": 0.412109375,
+      "rewards/margins": 1.5859375,
+      "rewards/rejected": -1.171875,
+      "step": 520
+    },
+    {
+      "epoch": 1.6987179487179487,
+      "grad_norm": 22.977308639981327,
+      "learning_rate": 2.410926365795724e-07,
+      "logits/chosen": -0.369140625,
+      "logits/rejected": -0.2470703125,
+      "logps/chosen": -352.0,
+      "logps/rejected": -340.0,
+      "loss": 0.2788,
+      "rewards/accuracies": 0.8999999761581421,
+      "rewards/chosen": 0.37890625,
+      "rewards/margins": 1.7734375,
+      "rewards/rejected": -1.390625,
+      "step": 530
+    },
+    {
+      "epoch": 1.7307692307692308,
+      "grad_norm": 19.697815015591537,
+      "learning_rate": 2.351543942992874e-07,
+      "logits/chosen": -0.294921875,
+      "logits/rejected": -0.34375,
+      "logps/chosen": -234.0,
+      "logps/rejected": -282.0,
+      "loss": 0.3076,
+      "rewards/accuracies": 0.875,
+      "rewards/chosen": 0.0830078125,
+      "rewards/margins": 1.8671875,
+      "rewards/rejected": -1.78125,
+      "step": 540
+    },
+    {
+      "epoch": 1.7628205128205128,
+      "grad_norm": 16.42605287271066,
+      "learning_rate": 2.2921615201900234e-07,
+      "logits/chosen": -0.5625,
+      "logits/rejected": -0.41796875,
+      "logps/chosen": -350.0,
+      "logps/rejected": -344.0,
+      "loss": 0.2614,
+      "rewards/accuracies": 0.9125000238418579,
+      "rewards/chosen": 0.21875,
+      "rewards/margins": 1.890625,
+      "rewards/rejected": -1.671875,
+      "step": 550
+    },
+    {
+      "epoch": 1.7948717948717947,
+      "grad_norm": 21.70957132124816,
+      "learning_rate": 2.2327790973871732e-07,
+      "logits/chosen": -0.431640625,
+      "logits/rejected": -0.3984375,
+      "logps/chosen": -324.0,
+      "logps/rejected": -272.0,
+      "loss": 0.3165,
+      "rewards/accuracies": 0.862500011920929,
+      "rewards/chosen": 0.0252685546875,
+      "rewards/margins": 1.671875,
+      "rewards/rejected": -1.6484375,
+      "step": 560
+    },
+    {
+      "epoch": 1.8269230769230769,
+      "grad_norm": 22.45992885626303,
+      "learning_rate": 2.173396674584323e-07,
+      "logits/chosen": -0.29296875,
+      "logits/rejected": -0.359375,
+      "logps/chosen": -294.0,
+      "logps/rejected": -302.0,
+      "loss": 0.2871,
+      "rewards/accuracies": 0.862500011920929,
+      "rewards/chosen": 0.1748046875,
+      "rewards/margins": 1.7578125,
+      "rewards/rejected": -1.5859375,
+      "step": 570
+    },
+    {
+      "epoch": 1.858974358974359,
+      "grad_norm": 20.14427708154685,
+      "learning_rate": 2.1140142517814726e-07,
+      "logits/chosen": -0.359375,
+      "logits/rejected": -0.359375,
+      "logps/chosen": -324.0,
+      "logps/rejected": -308.0,
+      "loss": 0.2926,
+      "rewards/accuracies": 0.887499988079071,
+      "rewards/chosen": 0.443359375,
+      "rewards/margins": 1.6015625,
+      "rewards/rejected": -1.15625,
+      "step": 580
+    },
+    {
+      "epoch": 1.891025641025641,
+      "grad_norm": 19.79995273385685,
+      "learning_rate": 2.0546318289786222e-07,
+      "logits/chosen": -0.322265625,
+      "logits/rejected": -0.38671875,
+      "logps/chosen": -254.0,
+      "logps/rejected": -288.0,
+      "loss": 0.2864,
+      "rewards/accuracies": 0.8999999761581421,
+      "rewards/chosen": 0.419921875,
+      "rewards/margins": 1.703125,
+      "rewards/rejected": -1.28125,
+      "step": 590
+    },
+    {
+      "epoch": 1.9230769230769231,
+      "grad_norm": 23.566386363575898,
+      "learning_rate": 1.9952494061757718e-07,
+      "logits/chosen": -0.306640625,
+      "logits/rejected": -0.322265625,
+      "logps/chosen": -282.0,
+      "logps/rejected": -272.0,
+      "loss": 0.2718,
+      "rewards/accuracies": 0.887499988079071,
+      "rewards/chosen": 0.57421875,
+      "rewards/margins": 1.78125,
+      "rewards/rejected": -1.2109375,
+      "step": 600
+    },
+    {
+      "epoch": 1.9551282051282053,
+      "grad_norm": 29.152008466264334,
+      "learning_rate": 1.9358669833729216e-07,
+      "logits/chosen": -0.59375,
+      "logits/rejected": -0.447265625,
+      "logps/chosen": -308.0,
+      "logps/rejected": -282.0,
+      "loss": 0.2682,
+      "rewards/accuracies": 0.887499988079071,
+      "rewards/chosen": 0.234375,
+      "rewards/margins": 1.7421875,
+      "rewards/rejected": -1.5078125,
+      "step": 610
+    },
+    {
+      "epoch": 1.9871794871794872,
+      "grad_norm": 22.58745268335563,
+      "learning_rate": 1.876484560570071e-07,
+      "logits/chosen": -0.451171875,
+      "logits/rejected": -0.470703125,
+      "logps/chosen": -360.0,
+      "logps/rejected": -300.0,
+      "loss": 0.2825,
+      "rewards/accuracies": 0.875,
+      "rewards/chosen": 0.33984375,
+      "rewards/margins": 1.703125,
+      "rewards/rejected": -1.3671875,
+      "step": 620
+    },
+    {
+      "epoch": 2.0,
+      "eval_logits/chosen": -0.3828125,
+      "eval_logits/rejected": -0.375,
+      "eval_logps/chosen": -298.0,
+      "eval_logps/rejected": -336.0,
+      "eval_loss": 0.5634375214576721,
+      "eval_rewards/accuracies": 0.625,
+      "eval_rewards/chosen": -0.345703125,
+      "eval_rewards/margins": 0.625,
+      "eval_rewards/rejected": -0.96875,
+      "eval_runtime": 18.4069,
+      "eval_samples_per_second": 10.865,
+      "eval_steps_per_second": 0.38,
+      "step": 624
     }
   ],
   "logging_steps": 10,