Training in progress, step 200, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f3ff6417f337444900c9d0ee3418cae9650f976bbe8cad9be7e0ebb946b00b8
-size 329100200
+oid sha256:9ce65bbed9331f4da673d60d1dc63ba5675640ffcc17c86a99dd8472de8ef429
+size 253577648

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c37d5c1333fa20846532da34840740f4df14612f69d38d91951e2a09c63b6c97
-size 188845393
+oid sha256:15cccd4cc3adad21271edce5de20158da491b15c14ccceddfd2093b1befc0139
+size 37798173

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f5656667be1478f776aa5cbc3870cc2b7f43ab2e34c2fe808cae05f89d24863b
+oid sha256:8a9d93ac0e7384563f970848a6d911064afa1c50720cde8b4ec828d00b5723a5
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17980325c36d869e063d721782688706e0e30902ea32568d55af1c62cc6c3e6f
+oid sha256:002599c9fd9c2e34b48da3b08b765f3b266ceba26e25343d6c7012afd5618305
 size 1465

last-checkpoint/trainer_state.json

@@ -2,7 +2,7 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.06666666666666667,
+  "epoch": 0.2,
   "eval_steps": 50,
   "global_step": 200,
   "is_hyper_param_search": false,
@@ -10,320 +10,320 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.0016666666666666668,
-      "grad_norm": 54.850189208984375,
-      "learning_rate": 0.0002,
-      "loss": 6.845032501220703,
+      "epoch": 0.005,
+      "grad_norm": 68.2828140258789,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 3.547993850708008,
       "step": 5
     },
     {
-      "epoch": 0.0033333333333333335,
-      "grad_norm": 55.685916900634766,
-      "learning_rate": 0.00045000000000000004,
-      "loss": 6.620301055908203,
+      "epoch": 0.01,
+      "grad_norm": 381.42022705078125,
+      "learning_rate": 7.2e-05,
+      "loss": 3.2959220886230467,
       "step": 10
     },
     {
-      "epoch": 0.005,
-      "grad_norm": 7.728530406951904,
-      "learning_rate": 0.0005,
-      "loss": 7.703640747070312,
+      "epoch": 0.015,
+      "grad_norm": 19640.955078125,
+      "learning_rate": 0.00011200000000000001,
+      "loss": 3.369282531738281,
       "step": 15
     },
     {
-      "epoch": 0.006666666666666667,
-      "grad_norm": 7.862086296081543,
-      "learning_rate": 0.0005,
-      "loss": 7.1933952331542965,
+      "epoch": 0.02,
+      "grad_norm": 205.24681091308594,
+      "learning_rate": 0.000152,
+      "loss": 3.256881332397461,
       "step": 20
     },
     {
-      "epoch": 0.008333333333333333,
-      "grad_norm": 41934556.0,
-      "learning_rate": 0.0005,
-      "loss": 7.290315246582031,
+      "epoch": 0.025,
+      "grad_norm": 259.6229248046875,
+      "learning_rate": 0.000192,
+      "loss": 3.3879322052001952,
       "step": 25
     },
     {
-      "epoch": 0.01,
-      "grad_norm": 32783185920.0,
-      "learning_rate": 0.0005,
-      "loss": 10.341773986816406,
+      "epoch": 0.03,
+      "grad_norm": 2096.502685546875,
+      "learning_rate": 0.0001999916943334945,
+      "loss": 3.4791305541992186,
       "step": 30
     },
     {
-      "epoch": 0.011666666666666667,
-      "grad_norm": 67137892352.0,
-      "learning_rate": 0.0005,
-      "loss": 16.722955322265626,
+      "epoch": 0.035,
+      "grad_norm": 29.51216697692871,
+      "learning_rate": 0.0001999579549278937,
+      "loss": 3.5544124603271485,
       "step": 35
     },
     {
-      "epoch": 0.013333333333333334,
-      "grad_norm": 107766048.0,
-      "learning_rate": 0.0005,
-      "loss": 18.834597778320312,
+      "epoch": 0.04,
+      "grad_norm": 2868.5859375,
+      "learning_rate": 0.00019989827142936862,
+      "loss": 3.4086448669433596,
       "step": 40
     },
     {
-      "epoch": 0.015,
-      "grad_norm": 750511.1875,
-      "learning_rate": 0.0005,
-      "loss": 16.470211791992188,
+      "epoch": 0.045,
+      "grad_norm": 116.66321563720703,
+      "learning_rate": 0.00019981265932877488,
+      "loss": 3.26219482421875,
       "step": 45
     },
     {
-      "epoch": 0.016666666666666666,
-      "grad_norm": 1690175.0,
-      "learning_rate": 0.0005,
-      "loss": 17.498741149902344,
+      "epoch": 0.05,
+      "grad_norm": 1199.3328857421875,
+      "learning_rate": 0.00019970114084673796,
+      "loss": 3.556040954589844,
       "step": 50
     },
     {
-      "epoch": 0.016666666666666666,
-      "eval_loss": 9.509819984436035,
-      "eval_runtime": 217.3232,
-      "eval_samples_per_second": 1.371,
-      "eval_steps_per_second": 0.686,
+      "epoch": 0.05,
+      "eval_loss": 3.7243571281433105,
+      "eval_runtime": 200.7945,
+      "eval_samples_per_second": 1.484,
+      "eval_steps_per_second": 0.299,
       "step": 50
     },
     {
-      "epoch": 0.018333333333333333,
-      "grad_norm": 44012.49609375,
-      "learning_rate": 0.0005,
-      "loss": 19.190980529785158,
+      "epoch": 0.055,
+      "grad_norm": 1653.0960693359375,
+      "learning_rate": 0.0001995637449278864,
+      "loss": 3.860651397705078,
       "step": 55
     },
     {
-      "epoch": 0.02,
-      "grad_norm": 1957.0230712890625,
-      "learning_rate": 0.0005,
-      "loss": 20.642401123046874,
+      "epoch": 0.06,
+      "grad_norm": 411.4178466796875,
+      "learning_rate": 0.00019940050723333866,
+      "loss": 3.699799728393555,
       "step": 60
     },
     {
-      "epoch": 0.021666666666666667,
-      "grad_norm": 238.84210205078125,
-      "learning_rate": 0.0005,
-      "loss": 21.76719970703125,
+      "epoch": 0.065,
+      "grad_norm": 2424.376708984375,
+      "learning_rate": 0.0001992114701314478,
+      "loss": 3.703348159790039,
       "step": 65
     },
     {
-      "epoch": 0.023333333333333334,
-      "grad_norm": 71.29987335205078,
-      "learning_rate": 0.0005,
-      "loss": 20.405929565429688,
+      "epoch": 0.07,
+      "grad_norm": 23206.84375,
+      "learning_rate": 0.0001989966826868044,
+      "loss": 3.6570384979248045,
       "step": 70
     },
     {
-      "epoch": 0.025,
-      "grad_norm": 105.72884368896484,
-      "learning_rate": 0.0005,
-      "loss": 19.047821044921875,
+      "epoch": 0.075,
+      "grad_norm": 395.9013977050781,
+      "learning_rate": 0.00019875620064750202,
+      "loss": 3.622447204589844,
       "step": 75
     },
     {
-      "epoch": 0.02666666666666667,
-      "grad_norm": 7008.24560546875,
-      "learning_rate": 0.0005,
-      "loss": 18.463946533203124,
+      "epoch": 0.08,
+      "grad_norm": 228293.453125,
+      "learning_rate": 0.00019849008643066772,
+      "loss": 3.773283767700195,
       "step": 80
     },
     {
-      "epoch": 0.028333333333333332,
-      "grad_norm": 171495.171875,
-      "learning_rate": 0.0005,
-      "loss": 18.29894561767578,
+      "epoch": 0.085,
+      "grad_norm": 52893.1875,
+      "learning_rate": 0.00019819840910626174,
+      "loss": 3.650778961181641,
       "step": 85
     },
     {
-      "epoch": 0.03,
-      "grad_norm": 122140.40625,
-      "learning_rate": 0.0005,
-      "loss": 18.62984619140625,
+      "epoch": 0.09,
+      "grad_norm": 12679.291015625,
+      "learning_rate": 0.0001978812443791503,
+      "loss": 3.9305484771728514,
       "step": 90
     },
     {
-      "epoch": 0.03166666666666667,
-      "grad_norm": 8799.4404296875,
-      "learning_rate": 0.0005,
-      "loss": 18.06275177001953,
+      "epoch": 0.095,
+      "grad_norm": 439037.4375,
+      "learning_rate": 0.0001975386745694565,
+      "loss": 3.8479537963867188,
       "step": 95
     },
     {
-      "epoch": 0.03333333333333333,
-      "grad_norm": 435225.90625,
-      "learning_rate": 0.0005,
-      "loss": 17.608197021484376,
+      "epoch": 0.1,
+      "grad_norm": 878.3585815429688,
+      "learning_rate": 0.0001971707885911941,
+      "loss": 3.90667724609375,
       "step": 100
     },
     {
-      "epoch": 0.03333333333333333,
-      "eval_loss": 8.762983322143555,
-      "eval_runtime": 218.6047,
-      "eval_samples_per_second": 1.363,
-      "eval_steps_per_second": 0.682,
+      "epoch": 0.1,
+      "eval_loss": 4.1160054206848145,
+      "eval_runtime": 201.0129,
+      "eval_samples_per_second": 1.482,
+      "eval_steps_per_second": 0.298,
       "step": 100
     },
     {
-      "epoch": 0.035,
-      "grad_norm": 10736.8583984375,
-      "learning_rate": 0.0005,
-      "loss": 17.602194213867186,
+      "epoch": 0.105,
+      "grad_norm": 2385.37744140625,
+      "learning_rate": 0.00019677768192918971,
+      "loss": 3.805155944824219,
       "step": 105
     },
     {
-      "epoch": 0.03666666666666667,
-      "grad_norm": 2109406.5,
-      "learning_rate": 0.0005,
-      "loss": 18.206829833984376,
+      "epoch": 0.11,
+      "grad_norm": 10.20463752746582,
+      "learning_rate": 0.00019635945661430006,
+      "loss": 3.822935104370117,
       "step": 110
     },
     {
-      "epoch": 0.03833333333333333,
-      "grad_norm": 311519.9375,
-      "learning_rate": 0.0005,
-      "loss": 18.991905212402344,
+      "epoch": 0.115,
+      "grad_norm": 7.702394962310791,
+      "learning_rate": 0.0001959162211969295,
+      "loss": 3.702631378173828,
       "step": 115
     },
     {
-      "epoch": 0.04,
-      "grad_norm": 295.97235107421875,
-      "learning_rate": 0.0005,
-      "loss": 19.67296447753906,
+      "epoch": 0.12,
+      "grad_norm": 2.8037490844726562,
+      "learning_rate": 0.00019544809071885604,
+      "loss": 3.6845611572265624,
       "step": 120
     },
     {
-      "epoch": 0.041666666666666664,
-      "grad_norm": 31418684.0,
-      "learning_rate": 0.0005,
-      "loss": 20.369107055664063,
+      "epoch": 0.125,
+      "grad_norm": 3.6743063926696777,
+      "learning_rate": 0.00019495518668337201,
+      "loss": 3.644290542602539,
       "step": 125
     },
     {
-      "epoch": 0.043333333333333335,
-      "grad_norm": 3918581.0,
-      "learning_rate": 0.0005,
-      "loss": 20.751130676269533,
+      "epoch": 0.13,
+      "grad_norm": 5.9780964851379395,
+      "learning_rate": 0.00019443763702374812,
+      "loss": 3.5586227416992187,
       "step": 130
     },
     {
-      "epoch": 0.045,
-      "grad_norm": 13102.0244140625,
-      "learning_rate": 0.0005,
-      "loss": 21.04936828613281,
+      "epoch": 0.135,
+      "grad_norm": 2.230492353439331,
+      "learning_rate": 0.00019389557607002805,
+      "loss": 3.4619365692138673,
       "step": 135
     },
     {
-      "epoch": 0.04666666666666667,
-      "grad_norm": 16994096.0,
-      "learning_rate": 0.0005,
-      "loss": 21.621359252929686,
+      "epoch": 0.14,
+      "grad_norm": 1.7121639251708984,
+      "learning_rate": 0.00019332914451416347,
+      "loss": 3.531117248535156,
       "step": 140
     },
     {
-      "epoch": 0.04833333333333333,
-      "grad_norm": 2272.0400390625,
-      "learning_rate": 0.0005,
-      "loss": 22.08568878173828,
+      "epoch": 0.145,
+      "grad_norm": 4.843437194824219,
+      "learning_rate": 0.0001927384893734971,
+      "loss": 3.5250553131103515,
       "step": 145
     },
     {
-      "epoch": 0.05,
-      "grad_norm": 4981.4296875,
-      "learning_rate": 0.0005,
-      "loss": 22.322303771972656,
+      "epoch": 0.15,
+      "grad_norm": 1.5022116899490356,
+      "learning_rate": 0.00019212376395260448,
+      "loss": 3.252106475830078,
       "step": 150
     },
     {
-      "epoch": 0.05,
-      "eval_loss": 11.168928146362305,
-      "eval_runtime": 217.379,
-      "eval_samples_per_second": 1.371,
-      "eval_steps_per_second": 0.685,
+      "epoch": 0.15,
+      "eval_loss": 3.604867935180664,
+      "eval_runtime": 200.7401,
+      "eval_samples_per_second": 1.485,
+      "eval_steps_per_second": 0.299,
       "step": 150
     },
     {
-      "epoch": 0.051666666666666666,
-      "grad_norm": 2926.708251953125,
-      "learning_rate": 0.0005,
-      "loss": 22.440846252441407,
+      "epoch": 0.155,
+      "grad_norm": 2.400775909423828,
+      "learning_rate": 0.00019148512780350384,
+      "loss": 3.291422653198242,
       "step": 155
     },
     {
-      "epoch": 0.05333333333333334,
-      "grad_norm": 37812.78515625,
-      "learning_rate": 0.0005,
-      "loss": 22.036326599121093,
+      "epoch": 0.16,
+      "grad_norm": 2.1534276008605957,
+      "learning_rate": 0.00019082274668424422,
+      "loss": 3.4382129669189454,
       "step": 160
     },
     {
-      "epoch": 0.055,
-      "grad_norm": 638.6507568359375,
-      "learning_rate": 0.0005,
-      "loss": 20.635609436035157,
+      "epoch": 0.165,
+      "grad_norm": 5.130342960357666,
+      "learning_rate": 0.00019013679251588303,
+      "loss": 3.4447181701660154,
       "step": 165
     },
     {
-      "epoch": 0.056666666666666664,
-      "grad_norm": 45.453582763671875,
-      "learning_rate": 0.0005,
-      "loss": 19.9438720703125,
+      "epoch": 0.17,
+      "grad_norm": 125.57601928710938,
+      "learning_rate": 0.00018942744333786397,
+      "loss": 3.4935020446777343,
       "step": 170
     },
     {
-      "epoch": 0.058333333333333334,
-      "grad_norm": 122177.8359375,
-      "learning_rate": 0.0005,
-      "loss": 19.473664855957033,
+      "epoch": 0.175,
+      "grad_norm": 7.2148356437683105,
+      "learning_rate": 0.00018869488326180679,
+      "loss": 3.4430862426757813,
       "step": 175
     },
     {
-      "epoch": 0.06,
-      "grad_norm": 1840.748779296875,
-      "learning_rate": 0.0005,
-      "loss": 19.456393432617187,
+      "epoch": 0.18,
+      "grad_norm": 47.17856979370117,
+      "learning_rate": 0.0001879393024237212,
+      "loss": 3.4849868774414063,
       "step": 180
     },
     {
-      "epoch": 0.06166666666666667,
-      "grad_norm": 130342.7265625,
-      "learning_rate": 0.0005,
-      "loss": 19.97138366699219,
+      "epoch": 0.185,
+      "grad_norm": 3.984619140625,
+      "learning_rate": 0.00018716089693465696,
+      "loss": 3.467825698852539,
       "step": 185
     },
     {
-      "epoch": 0.06333333333333334,
-      "grad_norm": 26071.556640625,
-      "learning_rate": 0.0005,
-      "loss": 20.022232055664062,
+      "epoch": 0.19,
+      "grad_norm": 1.7979289293289185,
+      "learning_rate": 0.00018635986882980325,
+      "loss": 3.2273773193359374,
       "step": 190
     },
     {
-      "epoch": 0.065,
-      "grad_norm": 6111.51513671875,
-      "learning_rate": 0.0005,
-      "loss": 20.719598388671876,
+      "epoch": 0.195,
+      "grad_norm": 49.847412109375,
+      "learning_rate": 0.00018553642601605068,
+      "loss": 3.51593017578125,
       "step": 195
     },
     {
-      "epoch": 0.06666666666666667,
-      "grad_norm": 387584.5,
-      "learning_rate": 0.0005,
-      "loss": 21.63736572265625,
+      "epoch": 0.2,
+      "grad_norm": 1.552674412727356,
+      "learning_rate": 0.0001846907822180286,
+      "loss": 3.5006553649902346,
       "step": 200
     },
     {
-      "epoch": 0.06666666666666667,
-      "eval_loss": 11.059845924377441,
-      "eval_runtime": 216.5693,
-      "eval_samples_per_second": 1.376,
-      "eval_steps_per_second": 0.688,
+      "epoch": 0.2,
+      "eval_loss": 3.619206190109253,
+      "eval_runtime": 201.0123,
+      "eval_samples_per_second": 1.482,
+      "eval_steps_per_second": 0.298,
       "step": 200
     }
   ],
   "logging_steps": 5,
-  "max_steps": 3000,
+  "max_steps": 1000,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 9223372036854775807,
   "save_steps": 200,
@@ -339,7 +339,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2276052487372800.0,
+  "total_flos": 952423258521600.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:56d4930d2f2e2d96034931eacf623f5cc07c816a8f7e9190dfff6ba278e2fb60
+oid sha256:832e2094ad3db4c13c39e196610b6d644659d4779665163c375adaff3acee4e4
 size 5201