Add files using upload-large-folder tool

global_step4403/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17a6a648da445af15cd5bea321fc11a5c23fdd269644dbce9ca75f1656dc2830
+size 3760226304

latest

@@ -1 +1 @@
-global_step2995
+global_step4403

model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55aeaab7a90a7d2a1a4866d187f47e67153d275ac8524a3f6723acf38c0cc7aa
+oid sha256:806858e608db4065644f4bd3251d595f67c672bfe354e29d0a3c63a5315f59fb
 size 4968243304

model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f251e4436393cbc69c5d59ce60af68995d48904e2551f59703dbe5d0e4ffb0c
+oid sha256:bb2d943f5670b6d38e1105bcd1b47158bacb3676741c7fb1a6ead2c9bd91f046
 size 4991495816

model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bea4aa4a2bcb6fe68408853132fadbdd37b13fa8c5b108c675136d3c2f97b64d
+oid sha256:5567189897093f0418f25ab4c3dd52a386cdb413f1c3a5ce228732fd7bd2a851
 size 4932751040

model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e72e729811d53e78a6f34e75fcaadc2ca0160800c769b6823646601422e049f4
+oid sha256:942ed51be07c0f5e4cbe7a38ce3ae081879e5b33d6b626bc70f69e8b38b2fec0
 size 1691924384

rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7d7e02ffb4d440dce7ab4ce0b5617578ec9ce3672acee7434ed6f1153f1ae0c
+oid sha256:fb1165242405b17b3d6a8186ae61b13dcb1faa5a54320bebd74ef8d71b964bf7
 size 15984

rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c5b40ca759e432b2688f021b81291d74a40f56a205e9842119f7e772275eebd3
+oid sha256:562c262916c9997ec644c42fed9655ab28706b74fca20290ca921c4761d6a4b0
 size 15984

rng_state_10.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d77ef8825b0e27c584c2d3eaeebfd82ebe38732eae91ff9f69c9a874632730c8
+oid sha256:f6fd4911141e2842a3d884ce4d2109eaeeedefd435098cadf9b564c65967a6eb
 size 15997

rng_state_11.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32f9f9d83cb982f1d825d89bde80de05cf37ecaa787ad08de64e0c8df94e6115
+oid sha256:8ba6ece259230c4e521fc019ff78563c612add590722906bc4725a8f02ad5376
 size 15997

rng_state_12.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:77270e25477a4a35eaceb98ecff060e192a6357bdc322e5d19aa92ecacd4e7e3
+oid sha256:f1bd528dc10399e3df15ae65f3098a87d293e8d4777932e1875d6dadae5f2760
 size 15997

rng_state_13.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9eab08108477ca9ce657269520a7cfc34510e7812e8499eb4c1a918805c8fb0
+oid sha256:8ddc21f5027e4c32e3cbc9daa1863acde639199470b640785b2bf069b1758850
 size 15997

rng_state_14.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:162d71274c75bc8f1dcbf21836ecdbc2a8c2756b4971b76847083de39d3d4098
+oid sha256:15c7d46874297ff7a0b9cc4607aac3d077700d8f60db95b27325fad44d0ed974
 size 15997

rng_state_15.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8517010852772c3221be3f5fe18254baa1d12a58a99ef4cabc7765112ce8a7ea
+oid sha256:6a20510fa92135806b278f3c0f3a327dcede9674b4eb3de8360d7643b2a73f1a
 size 15997

rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cdaef955ddd36d6bc1c40584113dd6205483e2aa85b02439b8b27e82e02a8359
+oid sha256:a8d40f8118f513299624ded0a9bcf09778b961635615090409394d4f96f928f6
 size 15984

rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:10b14ae5db356e6512538751d6b386c190754e307cc99cd652d5c6dd891e1f82
+oid sha256:b4391f924238a4cb855c4cbdc6d1a14954f785431c75997d05c7a4ee6615dae7
 size 15984

rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f26e28be26826eeeed244b77185c67b443ac185175f8d4bf5ba94caa8b271bc5
+oid sha256:be7b19bb9543a16bf9f4cd96466ac581436f63070f5815f3a7ba57980608994f
 size 15984

rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:847cedc1d6ca26f299a132c2ade9754887374acb9d98f26594a85d4c7742d474
+oid sha256:97da4a1ede0a3e0f96411cacd5bfdf84d9355198f7aadc9bcb8be41122043f63
 size 15984

rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bcd043d1690ae0ff6991b03322799a0b28f021427b15fd9f1e5ed8b9905d9307
+oid sha256:544cb6421b975bd5d2b2360a4e666003794e6197ae654d2ad963cd6572a86ede
 size 15984

rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:772190f7e6667c865d25fc72da7bdd1b5d39f46fe03bb5c2d754aee1ad3c99c7
+oid sha256:f8d6eb32a23f3bef6262bbcb2eda724b2fd6f5e579969aa27c71a5971331722b
 size 15984

rng_state_8.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6632cfcdbb30b4b8f6c2437ffa4a4f76a2a7c729fefd331ecb675d2f63f96bf5
+oid sha256:29dee3c545638083bb257f6e165d367893ac8d5dbb07203c355bd5914e9e47a2
 size 15984

rng_state_9.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e6c87b73b3f0787cc3976a6cbd289ee87123c4cb74f388630459c010fa9fbeb
+oid sha256:6805b7c016228fa319f57f490b2d47f3b38a21aecf78e7cb6654a92340347d3b
 size 15984

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d561373aa5d93d08e27cff4bdee13071587b60cd3f13e8fe91e5701f1a9258dc
+oid sha256:c89398e01b5829487b6d28ed0235d051c0f71e779ce1d7835c279e8dbddad4e7
 size 1064

trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 6.788668555240793,
+  "epoch": 9.979036827195468,
   "eval_steps": 500,
-  "global_step": 3000,
+  "global_step": 4410,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -427,6 +427,202 @@
       "learning_rate": 6.394557823129253e-06,
       "loss": 0.0002,
       "step": 3000
+    },
+    {
+      "epoch": 6.9019830028328615,
+      "grad_norm": 0.014847405573922268,
+      "learning_rate": 6.16780045351474e-06,
+      "loss": 0.0002,
+      "step": 3050
+    },
+    {
+      "epoch": 7.013597733711048,
+      "grad_norm": 0.018305480145040905,
+      "learning_rate": 5.9410430839002275e-06,
+      "loss": 0.0002,
+      "step": 3100
+    },
+    {
+      "epoch": 7.126912181303116,
+      "grad_norm": 0.012819862819578573,
+      "learning_rate": 5.7142857142857145e-06,
+      "loss": 0.0001,
+      "step": 3150
+    },
+    {
+      "epoch": 7.240226628895184,
+      "grad_norm": 0.013371528432039266,
+      "learning_rate": 5.487528344671202e-06,
+      "loss": 0.0001,
+      "step": 3200
+    },
+    {
+      "epoch": 7.353541076487252,
+      "grad_norm": 0.005633164169599245,
+      "learning_rate": 5.260770975056689e-06,
+      "loss": 0.0001,
+      "step": 3250
+    },
+    {
+      "epoch": 7.46685552407932,
+      "grad_norm": 0.01061442292135964,
+      "learning_rate": 5.034013605442177e-06,
+      "loss": 0.0001,
+      "step": 3300
+    },
+    {
+      "epoch": 7.580169971671388,
+      "grad_norm": 0.00907172882592498,
+      "learning_rate": 4.807256235827665e-06,
+      "loss": 0.0001,
+      "step": 3350
+    },
+    {
+      "epoch": 7.693484419263456,
+      "grad_norm": 0.010227726380358893,
+      "learning_rate": 4.580498866213152e-06,
+      "loss": 0.0001,
+      "step": 3400
+    },
+    {
+      "epoch": 7.806798866855524,
+      "grad_norm": 0.008230239860332859,
+      "learning_rate": 4.35374149659864e-06,
+      "loss": 0.0001,
+      "step": 3450
+    },
+    {
+      "epoch": 7.920113314447592,
+      "grad_norm": 0.013256768674645692,
+      "learning_rate": 4.126984126984127e-06,
+      "loss": 0.0001,
+      "step": 3500
+    },
+    {
+      "epoch": 8.03172804532578,
+      "grad_norm": 0.01006097891003743,
+      "learning_rate": 3.9002267573696154e-06,
+      "loss": 0.0001,
+      "step": 3550
+    },
+    {
+      "epoch": 8.145042492917847,
+      "grad_norm": 0.009602466515273548,
+      "learning_rate": 3.6734693877551024e-06,
+      "loss": 0.0001,
+      "step": 3600
+    },
+    {
+      "epoch": 8.258356940509914,
+      "grad_norm": 0.005795654820959573,
+      "learning_rate": 3.44671201814059e-06,
+      "loss": 0.0001,
+      "step": 3650
+    },
+    {
+      "epoch": 8.371671388101984,
+      "grad_norm": 0.0034753960503702283,
+      "learning_rate": 3.2199546485260772e-06,
+      "loss": 0.0001,
+      "step": 3700
+    },
+    {
+      "epoch": 8.48498583569405,
+      "grad_norm": 0.006716712372637811,
+      "learning_rate": 2.993197278911565e-06,
+      "loss": 0.0001,
+      "step": 3750
+    },
+    {
+      "epoch": 8.598300283286118,
+      "grad_norm": 0.005691017397956627,
+      "learning_rate": 2.7664399092970525e-06,
+      "loss": 0.0001,
+      "step": 3800
+    },
+    {
+      "epoch": 8.711614730878187,
+      "grad_norm": 0.005512521725963195,
+      "learning_rate": 2.53968253968254e-06,
+      "loss": 0.0001,
+      "step": 3850
+    },
+    {
+      "epoch": 8.824929178470255,
+      "grad_norm": 0.004574835289146862,
+      "learning_rate": 2.3129251700680273e-06,
+      "loss": 0.0001,
+      "step": 3900
+    },
+    {
+      "epoch": 8.938243626062324,
+      "grad_norm": 0.005810520530240234,
+      "learning_rate": 2.086167800453515e-06,
+      "loss": 0.0001,
+      "step": 3950
+    },
+    {
+      "epoch": 9.04985835694051,
+      "grad_norm": 0.0032799015697558206,
+      "learning_rate": 1.8594104308390023e-06,
+      "loss": 0.0001,
+      "step": 4000
+    },
+    {
+      "epoch": 9.163172804532579,
+      "grad_norm": 0.0021797625948262924,
+      "learning_rate": 1.6326530612244897e-06,
+      "loss": 0.0,
+      "step": 4050
+    },
+    {
+      "epoch": 9.276487252124646,
+      "grad_norm": 0.003847390080167819,
+      "learning_rate": 1.4058956916099775e-06,
+      "loss": 0.0,
+      "step": 4100
+    },
+    {
+      "epoch": 9.389801699716713,
+      "grad_norm": 0.0025440319025627875,
+      "learning_rate": 1.179138321995465e-06,
+      "loss": 0.0,
+      "step": 4150
+    },
+    {
+      "epoch": 9.503116147308782,
+      "grad_norm": 0.002651261609298388,
+      "learning_rate": 9.523809523809525e-07,
+      "loss": 0.0,
+      "step": 4200
+    },
+    {
+      "epoch": 9.61643059490085,
+      "grad_norm": 0.0033509868369888817,
+      "learning_rate": 7.2562358276644e-07,
+      "loss": 0.0,
+      "step": 4250
+    },
+    {
+      "epoch": 9.729745042492917,
+      "grad_norm": 0.011727033108113469,
+      "learning_rate": 4.988662131519275e-07,
+      "loss": 0.0,
+      "step": 4300
+    },
+    {
+      "epoch": 9.843059490084986,
+      "grad_norm": 0.005342769601421854,
+      "learning_rate": 2.72108843537415e-07,
+      "loss": 0.0,
+      "step": 4350
+    },
+    {
+      "epoch": 9.956373937677053,
+      "grad_norm": 0.004383322417463679,
+      "learning_rate": 4.53514739229025e-08,
+      "loss": 0.0,
+      "step": 4400
     }
   ],
   "logging_steps": 50,
@@ -441,12 +637,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 1.6645832262549504e+16,
+  "total_flos": 2.446988850023629e+16,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null