Training in progress, step 186, checkpoint

last-checkpoint/global_step186/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2aaeb41c5665004a47a0e58988e98070c92b9773f3c2a6506921c45b1babd846
+size 13476835648

last-checkpoint/global_step186/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f675a0302b331fdd45744c30f6f0ed81b7395f64ea102eb758e6a125d333ff7
+size 13476835648

last-checkpoint/global_step186/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:a5ef3873e9941ca5f0097ae20188c5e1184da4d166f11ccf47da8f4f629aabfb
+size 13476835648

last-checkpoint/global_step186/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21a018209b152d9cec098fe56d766d1120fa5dc43784e496041d2b97ea000790
+size 13476835648

last-checkpoint/global_step186/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f30787e32a351f2fafa763bc3112306c77dd360bd809c192f5a312a49f34940c
+size 150693

last-checkpoint/global_step186/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3505ab73f642650cd46241792677675a984a76b2c3fdebac40e6b8d0af9cd51d
+size 150693

last-checkpoint/global_step186/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:830251cfc2c437586c0ae49bf1b66313be2c453ea7011d96d958fc6e9e364ac1
+size 150693

last-checkpoint/global_step186/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9aa8517485b38137ae2838bcc85f98965bb72cb092a20364df05e9ef39b1b5f
+size 150693

last-checkpoint/latest

@@ -1 +1 @@
-global_step62
+global_step186

last-checkpoint/model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7119c04b975a058ab7fbdb25e33f2fe499bbc653045f0751f1967de06bc059f4
+oid sha256:313d5167651e3868509dd3619ee1c2d07051385b98649993872036ea95d4aba0
 size 4938985352

last-checkpoint/model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fcf1cb7f139afc7498f2627b6355a7a650682f1a1ff022a1b690a7b2ebda85e4
+oid sha256:845c39a62c9cb44d23ae87ad70b7d3a9dc92f88b09fb653b4375a2ca9ac136c3
 size 4947390880

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0511ed38446e52ff4c30cc18a9793d536a2cd844005faafdbac6f3cb431e8edc
+oid sha256:61a944543e23aed815241148692fce42317e0a4e455f648feb2320a8353247ad
 size 3590488816

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4a1dc8b9f546123ce9b44c28828ae71d9d0ecd6f59c6b7b8337c73cb2e24c49
+oid sha256:50ca27603f7c4a4d9a980223e1a102922ac31561acfa7da4f28042e4e8b3528b
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a8b225b0323772b66570a16b5c7e6446dc3c9ee076597906a2c90e3c0a383bf
+oid sha256:c57f658c3d9b6de611da50021823095fc4067036fb9a4b34655cc2e34fbf5c2b
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c036a8d528905f7018266618c8bebe4add10c2778ef432fd0116369fa622d815
+oid sha256:109da2c363a1e09432fa2ea2b7dceb6f67d1f91975069fc669e71dd536d654d8
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b93d88429d97654e76d10b1ba4c36f29a47e210c65a03efdc136f04af8282797
+oid sha256:bda65e490958bfe3fe01157ed131eb2354f8fc19536842d27c998ecf29d48991
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5645f0c0c551384f392d5d4b3ba0069b3e59c71dc60c4952144026a9bc6705f2
+oid sha256:a4f13df3470221c87dbd73f49aa8200fcb840f387f1d9666c0df7b34d2fa99d1
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.1984,
+  "epoch": 0.5952,
   "eval_steps": 62,
-  "global_step": 62,
+  "global_step": 186,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -128,6 +128,218 @@
       "eval_samples_per_second": 7.308,
       "eval_steps_per_second": 0.475,
       "step": 62
+    },
+    {
+      "epoch": 0.224,
+      "grad_norm": 0.9090466290588308,
+      "learning_rate": 3.7234042553191484e-07,
+      "logits/generated": -0.7592553496360779,
+      "logits/real": -0.5271034240722656,
+      "logps/generated": -341.49017333984375,
+      "logps/real": -148.29876708984375,
+      "loss": 0.1549,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -4.969511985778809,
+      "rewards/margins": 10.437345504760742,
+      "rewards/real": 5.467833042144775,
+      "step": 70
+    },
+    {
+      "epoch": 0.256,
+      "grad_norm": 0.5548494311175409,
+      "learning_rate": 4.25531914893617e-07,
+      "logits/generated": -0.7383196949958801,
+      "logits/real": -0.8009077906608582,
+      "logps/generated": -321.6851806640625,
+      "logps/real": -123.8510971069336,
+      "loss": 0.1358,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -3.824869155883789,
+      "rewards/margins": 10.858232498168945,
+      "rewards/real": 7.033363342285156,
+      "step": 80
+    },
+    {
+      "epoch": 0.288,
+      "grad_norm": 0.42292249756861683,
+      "learning_rate": 4.787234042553192e-07,
+      "logits/generated": -0.8187915682792664,
+      "logits/real": -1.0330232381820679,
+      "logps/generated": -318.3337707519531,
+      "logps/real": -120.26998138427734,
+      "loss": 0.1213,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -3.1653218269348145,
+      "rewards/margins": 11.506210327148438,
+      "rewards/real": 8.340888977050781,
+      "step": 90
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.41767196123848027,
+      "learning_rate": 4.96437054631829e-07,
+      "logits/generated": -0.8708025217056274,
+      "logits/real": -1.1798157691955566,
+      "logps/generated": -309.2374572753906,
+      "logps/real": -115.8353042602539,
+      "loss": 0.1134,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -3.7763671875,
+      "rewards/margins": 12.563192367553711,
+      "rewards/real": 8.786824226379395,
+      "step": 100
+    },
+    {
+      "epoch": 0.352,
+      "grad_norm": 0.4453179950421757,
+      "learning_rate": 4.904988123515439e-07,
+      "logits/generated": -0.8516510725021362,
+      "logits/real": -1.1489499807357788,
+      "logps/generated": -326.2418518066406,
+      "logps/real": -125.87767028808594,
+      "loss": 0.1137,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -4.001976013183594,
+      "rewards/margins": 13.618423461914062,
+      "rewards/real": 9.616447448730469,
+      "step": 110
+    },
+    {
+      "epoch": 0.384,
+      "grad_norm": 0.39875861828610537,
+      "learning_rate": 4.845605700712589e-07,
+      "logits/generated": -0.7948419451713562,
+      "logits/real": -1.1261526346206665,
+      "logps/generated": -331.29888916015625,
+      "logps/real": -118.65771484375,
+      "loss": 0.1204,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -4.266651153564453,
+      "rewards/margins": 13.141134262084961,
+      "rewards/real": 8.874483108520508,
+      "step": 120
+    },
+    {
+      "epoch": 0.3968,
+      "eval_logits/generated": -0.8262335658073425,
+      "eval_logits/real": -1.1627280712127686,
+      "eval_logps/generated": -327.9040832519531,
+      "eval_logps/real": -113.16130065917969,
+      "eval_loss": 0.11531589180231094,
+      "eval_rewards/accuracies": 1.0,
+      "eval_rewards/generated": -4.523497581481934,
+      "eval_rewards/margins": 13.89319896697998,
+      "eval_rewards/real": 9.369701385498047,
+      "eval_runtime": 31.0275,
+      "eval_samples_per_second": 6.446,
+      "eval_steps_per_second": 0.419,
+      "step": 124
+    },
+    {
+      "epoch": 0.416,
+      "grad_norm": 0.35711714781872617,
+      "learning_rate": 4.786223277909738e-07,
+      "logits/generated": -0.8075119853019714,
+      "logits/real": -1.1974109411239624,
+      "logps/generated": -326.8999938964844,
+      "logps/real": -121.57466125488281,
+      "loss": 0.1108,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -4.337772369384766,
+      "rewards/margins": 13.636154174804688,
+      "rewards/real": 9.298380851745605,
+      "step": 130
+    },
+    {
+      "epoch": 0.448,
+      "grad_norm": 0.3901220032340114,
+      "learning_rate": 4.7268408551068883e-07,
+      "logits/generated": -0.8377895355224609,
+      "logits/real": -1.1887013912200928,
+      "logps/generated": -325.7310791015625,
+      "logps/real": -123.4073715209961,
+      "loss": 0.1129,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -4.29754114151001,
+      "rewards/margins": 13.601455688476562,
+      "rewards/real": 9.303915023803711,
+      "step": 140
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.3874502118142099,
+      "learning_rate": 4.667458432304038e-07,
+      "logits/generated": -0.8494235277175903,
+      "logits/real": -1.280548334121704,
+      "logps/generated": -323.9864807128906,
+      "logps/real": -115.0036849975586,
+      "loss": 0.1119,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -4.178435325622559,
+      "rewards/margins": 13.569964408874512,
+      "rewards/real": 9.391529083251953,
+      "step": 150
+    },
+    {
+      "epoch": 0.512,
+      "grad_norm": 0.408082769634744,
+      "learning_rate": 4.6080760095011875e-07,
+      "logits/generated": -0.8601406812667847,
+      "logits/real": -1.2258448600769043,
+      "logps/generated": -328.79156494140625,
+      "logps/real": -124.53865814208984,
+      "loss": 0.117,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -4.461178779602051,
+      "rewards/margins": 14.086248397827148,
+      "rewards/real": 9.625069618225098,
+      "step": 160
+    },
+    {
+      "epoch": 0.544,
+      "grad_norm": 0.36331526860516933,
+      "learning_rate": 4.548693586698337e-07,
+      "logits/generated": -0.8272736668586731,
+      "logits/real": -1.278106451034546,
+      "logps/generated": -331.06683349609375,
+      "logps/real": -119.9251480102539,
+      "loss": 0.1135,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -4.8740129470825195,
+      "rewards/margins": 14.372535705566406,
+      "rewards/real": 9.498522758483887,
+      "step": 170
+    },
+    {
+      "epoch": 0.576,
+      "grad_norm": 0.445865795650114,
+      "learning_rate": 4.4893111638954866e-07,
+      "logits/generated": -0.8073589205741882,
+      "logits/real": -1.2638658285140991,
+      "logps/generated": -342.365966796875,
+      "logps/real": -106.32196044921875,
+      "loss": 0.1114,
+      "rewards/accuracies": 1.0,
+      "rewards/generated": -5.440592288970947,
+      "rewards/margins": 14.317230224609375,
+      "rewards/real": 8.876636505126953,
+      "step": 180
+    },
+    {
+      "epoch": 0.5952,
+      "eval_logits/generated": -0.8445993065834045,
+      "eval_logits/real": -1.2393124103546143,
+      "eval_logps/generated": -335.83538818359375,
+      "eval_logps/real": -110.11853790283203,
+      "eval_loss": 0.11254960298538208,
+      "eval_rewards/accuracies": 1.0,
+      "eval_rewards/generated": -5.31662654876709,
+      "eval_rewards/margins": 14.990604400634766,
+      "eval_rewards/real": 9.673977851867676,
+      "eval_runtime": 25.7798,
+      "eval_samples_per_second": 7.758,
+      "eval_steps_per_second": 0.504,
+      "step": 186
     }
   ],
   "logging_steps": 10,