Training in progress, epoch 2, checkpoint

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:005461edf1879aad10a0e74fa8063b899d4dc153c4c4ee2ee870ec6ab02d0724
+oid sha256:3da40a5e2cd69e35db1c4bfd34661796a25ac71a510cbccbfb7b36b7caf9025a
 size 4903351912

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c242ad1a6b91bc1ab29f05b1aa380973a69a8aa79ef7deed8c29db48896f89f5
+oid sha256:d3d4d922ecb0fbc025a2f903f492c8baa33ea8cceb3dc068450e7e0b6518684e
 size 4947570872

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:30f6cf7f219f0d8b2be6e68425eaf8b45e404706a2998972e580bfbce5114648
+oid sha256:2c5d33e285bcbf98218ba55bce7efbdd306b8046089737b99aa4fe5c5e7ebea5
 size 4962221464

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3de7dc324edd38a81977296cbbf841bca1256ee41679022983816e95b863604d
+oid sha256:261c82a71bbdefce5c6b82990717f18e2f9cb6c02b8e8a09188d4f81d1e757ea
 size 3670322200

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b3f2db12fa449ee60e14ed19e4869728b4fe4cc9bf5a3ba552cf15c1e8a0861
+oid sha256:ca3ba1546a764910d034f0ce3e93370a84a16db802bc17eb73ebc48babbb424c
 size 36967230034

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:955b895101e13c81ba510512313a06782795770a0bf998c90f718166d25f1664
+oid sha256:74386f26f36ed67f56395205881e5db2d0c28ffcbeed50dd95b28771d2dac588
 size 15984

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:096e351ace65ff91008be171a45173ed6369cc639fce73a288f671041e24b0ec
+oid sha256:41c88f9de084200454883a13c3717941ea3fd433e2f8735507fc30611f9c5501
 size 15984

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f21c61b1a7e793bbdec183de3b52da90042305234bc7e5887986655cd3fc2192
+oid sha256:965b00d4cb4710ebab57c8787b9925bb3f77b8eeba94a186ec4bc1c2f326ef3f
 size 15984

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:348742498d81780506d9760c655a7a7555185b5fbd70a7ae296d88fd9aeecd84
+oid sha256:d5dc374b8b9a4c45c950f9d136feab85a767081fa59f0c7d68ed3a62060c4949
 size 15984

last-checkpoint/rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:399c4700dab9ae7b754110ce307fb7e26e22cb49b5bb233c435f0f12b77c202f
+oid sha256:5c7c212fb779217f1edac0baf44f67b608eefc1e0e4e3f5a9dd7eb557032c1bc
 size 15984

last-checkpoint/rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:864ea2379cc907eb4189c52706cb978150d9c26e18abf74679590729a8f0c8e8
+oid sha256:86e1effd626ce1e95dd68a0c8089fe19218f2b24dfe9e45ed2cab1c0ebc10ba1
 size 15984

last-checkpoint/rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25755ba07299ee6ff45936dd04df329596319c9f8095af71e6f3a219e7543e26
+oid sha256:799cc83f60dfc1c4243cfd6403592112414a2eec494e6832f10221c96ff62c20
 size 15984

last-checkpoint/rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:908f018cd701ed629c41299726da4a25f202f20a1d4bc2075a2266ed4013db3a
+oid sha256:586777c398770c3255d3a1f48c7fef44ea9d89117c627c9ea490e16bfd9a49ba
 size 15984

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f709d9064ea54be0c4294c2790b947f9f308ff4a7e82014074b3bee9d133837
+oid sha256:5713040e0c397c4c3f4a5dd5b82184ec7e9ce43fb4fc625e2c829d6f4609bb2d
 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": 300,
-  "global_step": 273,
+  "global_step": 546,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -392,6 +392,399 @@
       "learning_rate": 1.5121454937319975e-06,
       "loss": 4.9708,
       "step": 270
+    },
+    {
+      "epoch": 1.0073260073260073,
+      "grad_norm": 3.9375,
+      "learning_rate": 1.4955587465513128e-06,
+      "loss": 4.9494,
+      "step": 275
+    },
+    {
+      "epoch": 1.0256410256410255,
+      "grad_norm": 3.703125,
+      "learning_rate": 1.4787892672278555e-06,
+      "loss": 4.9172,
+      "step": 280
+    },
+    {
+      "epoch": 1.043956043956044,
+      "grad_norm": 3.515625,
+      "learning_rate": 1.461843239333021e-06,
+      "loss": 4.9262,
+      "step": 285
+    },
+    {
+      "epoch": 1.0622710622710623,
+      "grad_norm": 3.75,
+      "learning_rate": 1.444726911538657e-06,
+      "loss": 4.9115,
+      "step": 290
+    },
+    {
+      "epoch": 1.0805860805860805,
+      "grad_norm": 3.671875,
+      "learning_rate": 1.4274465953129325e-06,
+      "loss": 4.8926,
+      "step": 295
+    },
+    {
+      "epoch": 1.098901098901099,
+      "grad_norm": 3.625,
+      "learning_rate": 1.4100086625930462e-06,
+      "loss": 4.8845,
+      "step": 300
+    },
+    {
+      "epoch": 1.098901098901099,
+      "eval_loss": 4.895308017730713,
+      "eval_runtime": 14.5245,
+      "eval_samples_per_second": 36.421,
+      "eval_steps_per_second": 1.17,
+      "step": 300
+    },
+    {
+      "epoch": 1.1172161172161172,
+      "grad_norm": 3.5,
+      "learning_rate": 1.3924195434356441e-06,
+      "loss": 4.8754,
+      "step": 305
+    },
+    {
+      "epoch": 1.1355311355311355,
+      "grad_norm": 3.484375,
+      "learning_rate": 1.3746857236458005e-06,
+      "loss": 4.8566,
+      "step": 310
+    },
+    {
+      "epoch": 1.1538461538461537,
+      "grad_norm": 3.875,
+      "learning_rate": 1.3568137423854457e-06,
+      "loss": 4.873,
+      "step": 315
+    },
+    {
+      "epoch": 1.1721611721611722,
+      "grad_norm": 3.515625,
+      "learning_rate": 1.3388101897621183e-06,
+      "loss": 4.8476,
+      "step": 320
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 4.0,
+      "learning_rate": 1.32068170439893e-06,
+      "loss": 4.8846,
+      "step": 325
+    },
+    {
+      "epoch": 1.2087912087912087,
+      "grad_norm": 3.453125,
+      "learning_rate": 1.3024349709866448e-06,
+      "loss": 4.864,
+      "step": 330
+    },
+    {
+      "epoch": 1.2271062271062272,
+      "grad_norm": 5.125,
+      "learning_rate": 1.2840767178187654e-06,
+      "loss": 4.8457,
+      "step": 335
+    },
+    {
+      "epoch": 1.2454212454212454,
+      "grad_norm": 3.828125,
+      "learning_rate": 1.265613714310548e-06,
+      "loss": 4.8469,
+      "step": 340
+    },
+    {
+      "epoch": 1.2637362637362637,
+      "grad_norm": 3.59375,
+      "learning_rate": 1.2470527685028482e-06,
+      "loss": 4.8406,
+      "step": 345
+    },
+    {
+      "epoch": 1.282051282051282,
+      "grad_norm": 3.609375,
+      "learning_rate": 1.228400724551728e-06,
+      "loss": 4.8545,
+      "step": 350
+    },
+    {
+      "epoch": 1.3003663003663004,
+      "grad_norm": 3.578125,
+      "learning_rate": 1.2096644602047445e-06,
+      "loss": 4.8236,
+      "step": 355
+    },
+    {
+      "epoch": 1.3186813186813187,
+      "grad_norm": 3.515625,
+      "learning_rate": 1.1908508842648505e-06,
+      "loss": 4.8126,
+      "step": 360
+    },
+    {
+      "epoch": 1.3369963369963371,
+      "grad_norm": 3.453125,
+      "learning_rate": 1.171966934042847e-06,
+      "loss": 4.8038,
+      "step": 365
+    },
+    {
+      "epoch": 1.3553113553113554,
+      "grad_norm": 3.484375,
+      "learning_rate": 1.1530195727993199e-06,
+      "loss": 4.8278,
+      "step": 370
+    },
+    {
+      "epoch": 1.3736263736263736,
+      "grad_norm": 3.53125,
+      "learning_rate": 1.1340157871770115e-06,
+      "loss": 4.8217,
+      "step": 375
+    },
+    {
+      "epoch": 1.3919413919413919,
+      "grad_norm": 3.6875,
+      "learning_rate": 1.1149625846245681e-06,
+      "loss": 4.7912,
+      "step": 380
+    },
+    {
+      "epoch": 1.4102564102564101,
+      "grad_norm": 3.515625,
+      "learning_rate": 1.095866990812615e-06,
+      "loss": 4.8004,
+      "step": 385
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 3.5625,
+      "learning_rate": 1.0767360470431157e-06,
+      "loss": 4.7896,
+      "step": 390
+    },
+    {
+      "epoch": 1.4468864468864469,
+      "grad_norm": 3.625,
+      "learning_rate": 1.0575768076529625e-06,
+      "loss": 4.7788,
+      "step": 395
+    },
+    {
+      "epoch": 1.4652014652014653,
+      "grad_norm": 3.640625,
+      "learning_rate": 1.0383963374127645e-06,
+      "loss": 4.8106,
+      "step": 400
+    },
+    {
+      "epoch": 1.4835164835164836,
+      "grad_norm": 3.71875,
+      "learning_rate": 1.0192017089217861e-06,
+      "loss": 4.7749,
+      "step": 405
+    },
+    {
+      "epoch": 1.5018315018315018,
+      "grad_norm": 3.484375,
+      "learning_rate": 1e-06,
+      "loss": 4.7864,
+      "step": 410
+    },
+    {
+      "epoch": 1.52014652014652,
+      "grad_norm": 3.59375,
+      "learning_rate": 9.80798291078214e-07,
+      "loss": 4.7869,
+      "step": 415
+    },
+    {
+      "epoch": 1.5384615384615383,
+      "grad_norm": 3.625,
+      "learning_rate": 9.616036625872356e-07,
+      "loss": 4.8314,
+      "step": 420
+    },
+    {
+      "epoch": 1.5567765567765568,
+      "grad_norm": 3.734375,
+      "learning_rate": 9.424231923470376e-07,
+      "loss": 4.7831,
+      "step": 425
+    },
+    {
+      "epoch": 1.575091575091575,
+      "grad_norm": 3.578125,
+      "learning_rate": 9.232639529568842e-07,
+      "loss": 4.7636,
+      "step": 430
+    },
+    {
+      "epoch": 1.5934065934065935,
+      "grad_norm": 3.71875,
+      "learning_rate": 9.041330091873851e-07,
+      "loss": 4.7465,
+      "step": 435
+    },
+    {
+      "epoch": 1.6117216117216118,
+      "grad_norm": 3.546875,
+      "learning_rate": 8.850374153754321e-07,
+      "loss": 4.7616,
+      "step": 440
+    },
+    {
+      "epoch": 1.63003663003663,
+      "grad_norm": 3.46875,
+      "learning_rate": 8.659842128229886e-07,
+      "loss": 4.7542,
+      "step": 445
+    },
+    {
+      "epoch": 1.6483516483516483,
+      "grad_norm": 3.5,
+      "learning_rate": 8.4698042720068e-07,
+      "loss": 4.745,
+      "step": 450
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 3.921875,
+      "learning_rate": 8.280330659571531e-07,
+      "loss": 4.738,
+      "step": 455
+    },
+    {
+      "epoch": 1.684981684981685,
+      "grad_norm": 3.75,
+      "learning_rate": 8.091491157351493e-07,
+      "loss": 4.7335,
+      "step": 460
+    },
+    {
+      "epoch": 1.7032967032967035,
+      "grad_norm": 3.546875,
+      "learning_rate": 7.903355397952556e-07,
+      "loss": 4.7792,
+      "step": 465
+    },
+    {
+      "epoch": 1.7216117216117217,
+      "grad_norm": 3.546875,
+      "learning_rate": 7.715992754482718e-07,
+      "loss": 4.7288,
+      "step": 470
+    },
+    {
+      "epoch": 1.73992673992674,
+      "grad_norm": 3.515625,
+      "learning_rate": 7.529472314971522e-07,
+      "loss": 4.7489,
+      "step": 475
+    },
+    {
+      "epoch": 1.7582417582417582,
+      "grad_norm": 3.484375,
+      "learning_rate": 7.34386285689452e-07,
+      "loss": 4.7163,
+      "step": 480
+    },
+    {
+      "epoch": 1.7765567765567765,
+      "grad_norm": 3.828125,
+      "learning_rate": 7.159232821812347e-07,
+      "loss": 4.7201,
+      "step": 485
+    },
+    {
+      "epoch": 1.7948717948717947,
+      "grad_norm": 3.5625,
+      "learning_rate": 6.975650290133554e-07,
+      "loss": 4.7334,
+      "step": 490
+    },
+    {
+      "epoch": 1.8131868131868132,
+      "grad_norm": 3.609375,
+      "learning_rate": 6.793182956010699e-07,
+      "loss": 4.7339,
+      "step": 495
+    },
+    {
+      "epoch": 1.8315018315018317,
+      "grad_norm": 3.484375,
+      "learning_rate": 6.611898102378818e-07,
+      "loss": 4.7384,
+      "step": 500
+    },
+    {
+      "epoch": 1.84981684981685,
+      "grad_norm": 3.546875,
+      "learning_rate": 6.431862576145544e-07,
+      "loss": 4.7402,
+      "step": 505
+    },
+    {
+      "epoch": 1.8681318681318682,
+      "grad_norm": 3.703125,
+      "learning_rate": 6.253142763541995e-07,
+      "loss": 4.723,
+      "step": 510
+    },
+    {
+      "epoch": 1.8864468864468864,
+      "grad_norm": 3.5,
+      "learning_rate": 6.075804565643561e-07,
+      "loss": 4.7457,
+      "step": 515
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 3.5625,
+      "learning_rate": 5.899913374069538e-07,
+      "loss": 4.712,
+      "step": 520
+    },
+    {
+      "epoch": 1.9230769230769231,
+      "grad_norm": 3.703125,
+      "learning_rate": 5.725534046870677e-07,
+      "loss": 4.7273,
+      "step": 525
+    },
+    {
+      "epoch": 1.9413919413919414,
+      "grad_norm": 3.5625,
+      "learning_rate": 5.552730884613428e-07,
+      "loss": 4.6981,
+      "step": 530
+    },
+    {
+      "epoch": 1.9597069597069599,
+      "grad_norm": 3.609375,
+      "learning_rate": 5.381567606669793e-07,
+      "loss": 4.7482,
+      "step": 535
+    },
+    {
+      "epoch": 1.978021978021978,
+      "grad_norm": 3.765625,
+      "learning_rate": 5.212107327721445e-07,
+      "loss": 4.718,
+      "step": 540
+    },
+    {
+      "epoch": 1.9963369963369964,
+      "grad_norm": 3.6875,
+      "learning_rate": 5.044412534486873e-07,
+      "loss": 4.7148,
+      "step": 545
     }
   ],
   "logging_steps": 5,
@@ -411,7 +804,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.7871723276837847e+18,
+  "total_flos": 3.574344656441311e+18,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null