Training in progress, epoch 3, checkpoint

checkpoint-21/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ad72e757370b0af83a5a02b9d8c803f27d8f9c346657d034b0c1299a2e67cb6
+oid sha256:cc6c9bd36ff079ebad2560732c6bc6c0258a8ff8a82bfaba52f5ff7f59ab91ce
 size 3086640776

checkpoint-21/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f242db142262926b014c870ddab2206224f7f30dcaf88592583be1f598a13adc
+oid sha256:5148e1fdd912b47a99272cd2070fbbabb063dc847534d7b1aec51eee199f5070
 size 6056331

checkpoint-21/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed69d3ae118d797e33700fb938c2b3082fd9d43cc793829f446fd120f845714b
+oid sha256:aa0f95228341434b158308966669a313617960955e886ba19e972d72b58a1fc0
 size 15365

checkpoint-21/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:86accf27064cdd503053e90476a6bd10de333d4ff0594535ad55ea13a473c91d
+oid sha256:8fb125336725f7741cb4daa1e3d06e225bbacfde8d41c4dcabb6762c222e62c6
 size 15429

checkpoint-21/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:18ca8d714ef40be035404c1957b5a4dee84e1f43980408393f8aa710552ee6f6
+oid sha256:153c865f77c7129ba565bded50f334683d51c80f20e3cfec39e62f8737b86f0d
 size 15429

checkpoint-21/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2cfdebe99e40accc9c9d8f09c63136a14abda997d9b501969ec8e16e9d183179
+oid sha256:d017ce00fcebac7edc058ddd138f194eb0340f2d8ad0879bdab08f922ed0846e
 size 15429

checkpoint-21/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44beb602faaa18f6149ebc281880bf809c7972ab1d9b2e5653a5962d3a3930c7
+oid sha256:e6353f71b0e21a3df58576659e6881af379408a02aab091dafb80509d771fb79
 size 1465

checkpoint-21/trainer_state.json

@@ -2,7 +2,7 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.0,
+  "epoch": 3.0,
   "eval_steps": 500,
   "global_step": 21,
   "is_hyper_param_search": false,
@@ -10,155 +10,155 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.04878048780487805,
-      "grad_norm": 119.5,
+      "epoch": 0.15384615384615385,
+      "grad_norm": 103.5,
       "learning_rate": 0.0,
-      "loss": 6.2132,
+      "loss": 5.2404,
       "step": 1
     },
     {
-      "epoch": 0.0975609756097561,
-      "grad_norm": 132.0,
-      "learning_rate": 7.142857142857143e-06,
-      "loss": 6.6559,
+      "epoch": 0.3076923076923077,
+      "grad_norm": 104.5,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 5.287,
       "step": 2
     },
     {
-      "epoch": 0.14634146341463414,
-      "grad_norm": 128.0,
-      "learning_rate": 1.4285714285714285e-05,
-      "loss": 6.6652,
+      "epoch": 0.46153846153846156,
+      "grad_norm": 83.5,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 4.4505,
       "step": 3
     },
     {
-      "epoch": 0.1951219512195122,
-      "grad_norm": 129.0,
-      "learning_rate": 2.1428571428571428e-05,
-      "loss": 6.1414,
+      "epoch": 0.6153846153846154,
+      "grad_norm": 414.0,
+      "learning_rate": 5e-05,
+      "loss": 3.7876,
       "step": 4
     },
     {
-      "epoch": 0.24390243902439024,
-      "grad_norm": 348.0,
-      "learning_rate": 2.857142857142857e-05,
-      "loss": 5.1096,
+      "epoch": 0.7692307692307693,
+      "grad_norm": 47.5,
+      "learning_rate": 4.962019382530521e-05,
+      "loss": 4.1607,
       "step": 5
     },
     {
-      "epoch": 0.2926829268292683,
-      "grad_norm": 165.0,
-      "learning_rate": 3.571428571428572e-05,
-      "loss": 5.2497,
+      "epoch": 0.9230769230769231,
+      "grad_norm": 69.0,
+      "learning_rate": 4.849231551964771e-05,
+      "loss": 3.5029,
       "step": 6
     },
     {
-      "epoch": 0.34146341463414637,
-      "grad_norm": 136.0,
-      "learning_rate": 4.2857142857142856e-05,
-      "loss": 4.3006,
+      "epoch": 1.0,
+      "grad_norm": 64.5,
+      "learning_rate": 4.665063509461097e-05,
+      "loss": 3.8481,
       "step": 7
     },
     {
-      "epoch": 0.3902439024390244,
-      "grad_norm": 150.0,
-      "learning_rate": 5e-05,
-      "loss": 5.1854,
+      "epoch": 1.1538461538461537,
+      "grad_norm": 113.5,
+      "learning_rate": 4.415111107797445e-05,
+      "loss": 5.4532,
       "step": 8
     },
     {
-      "epoch": 0.43902439024390244,
-      "grad_norm": 213.0,
-      "learning_rate": 4.996067037544542e-05,
-      "loss": 8.1666,
+      "epoch": 1.3076923076923077,
+      "grad_norm": 36.25,
+      "learning_rate": 4.1069690242163484e-05,
+      "loss": 2.7612,
       "step": 9
     },
     {
-      "epoch": 0.4878048780487805,
-      "grad_norm": 137.0,
-      "learning_rate": 4.984280524733107e-05,
-      "loss": 9.5804,
+      "epoch": 1.4615384615384617,
+      "grad_norm": 35.75,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 2.5879,
       "step": 10
     },
     {
-      "epoch": 0.5365853658536586,
-      "grad_norm": 65.0,
-      "learning_rate": 4.96467754629559e-05,
-      "loss": 5.2593,
+      "epoch": 1.6153846153846154,
+      "grad_norm": 26.75,
+      "learning_rate": 3.355050358314172e-05,
+      "loss": 2.1463,
       "step": 11
     },
     {
-      "epoch": 0.5853658536585366,
-      "grad_norm": 40.25,
-      "learning_rate": 4.937319780454559e-05,
-      "loss": 3.9807,
+      "epoch": 1.7692307692307692,
+      "grad_norm": 28.375,
+      "learning_rate": 2.9341204441673266e-05,
+      "loss": 2.1599,
       "step": 12
     },
     {
-      "epoch": 0.6341463414634146,
-      "grad_norm": 39.75,
-      "learning_rate": 4.9022933048627496e-05,
-      "loss": 3.7199,
+      "epoch": 1.9230769230769231,
+      "grad_norm": 31.5,
+      "learning_rate": 2.5e-05,
+      "loss": 2.0294,
       "step": 13
     },
     {
-      "epoch": 0.6829268292682927,
-      "grad_norm": 50.0,
-      "learning_rate": 4.8597083257709194e-05,
-      "loss": 3.3626,
+      "epoch": 2.0,
+      "grad_norm": 32.0,
+      "learning_rate": 2.0658795558326743e-05,
+      "loss": 1.8506,
       "step": 14
     },
     {
-      "epoch": 0.7317073170731707,
-      "grad_norm": 44.5,
-      "learning_rate": 4.8096988312782174e-05,
-      "loss": 3.156,
+      "epoch": 2.1538461538461537,
+      "grad_norm": 32.0,
+      "learning_rate": 1.6449496416858284e-05,
+      "loss": 1.8264,
       "step": 15
     },
     {
-      "epoch": 0.7804878048780488,
-      "grad_norm": 43.5,
-      "learning_rate": 4.752422169756048e-05,
-      "loss": 2.62,
+      "epoch": 2.3076923076923075,
+      "grad_norm": 30.125,
+      "learning_rate": 1.2500000000000006e-05,
+      "loss": 1.6821,
       "step": 16
     },
     {
-      "epoch": 0.8292682926829268,
-      "grad_norm": 47.5,
-      "learning_rate": 4.6880585547718845e-05,
-      "loss": 2.4872,
+      "epoch": 2.4615384615384617,
+      "grad_norm": 30.625,
+      "learning_rate": 8.930309757836517e-06,
+      "loss": 1.6852,
       "step": 17
     },
     {
-      "epoch": 0.8780487804878049,
-      "grad_norm": 44.5,
-      "learning_rate": 4.6168104980707107e-05,
-      "loss": 2.0436,
+      "epoch": 2.6153846153846154,
+      "grad_norm": 32.5,
+      "learning_rate": 5.848888922025553e-06,
+      "loss": 1.7827,
       "step": 18
     },
     {
-      "epoch": 0.926829268292683,
-      "grad_norm": 50.5,
-      "learning_rate": 4.538902172398151e-05,
-      "loss": 1.9184,
+      "epoch": 2.769230769230769,
+      "grad_norm": 28.125,
+      "learning_rate": 3.3493649053890326e-06,
+      "loss": 1.6068,
       "step": 19
     },
     {
-      "epoch": 0.975609756097561,
-      "grad_norm": 55.5,
-      "learning_rate": 4.454578706170075e-05,
-      "loss": 1.8052,
+      "epoch": 2.9230769230769234,
+      "grad_norm": 35.25,
+      "learning_rate": 1.5076844803522922e-06,
+      "loss": 1.826,
       "step": 20
     },
     {
-      "epoch": 1.0,
-      "grad_norm": 41.75,
-      "learning_rate": 4.364105412207914e-05,
-      "loss": 1.4898,
+      "epoch": 3.0,
+      "grad_norm": 32.25,
+      "learning_rate": 3.7980617469479953e-07,
+      "loss": 1.6832,
       "step": 21
     }
   ],
   "logging_steps": 1,
-  "max_steps": 63,
+  "max_steps": 21,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
@@ -169,12 +169,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 1.0562561296564224e+16,
+  "total_flos": 1.0047314404048896e+16,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

checkpoint-21/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:841e48b365d43c1bb533dc5dd460b128556952c5da32bc814c2b5072cc6bb501
+oid sha256:3aefd857be144033d541a39da4375d1889c176cdc01d7c75e9761285d8084e1f
 size 5969