Training in progress, step 8000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bcc177d71a1506005267437007f2de8d24b91691c13a50b732ea2bcf9fe2581f
+oid sha256:e4640cdfd5049fa55f638b28d54cd06e5001e7d6ec0e00957bc1d8edde2a495a
 size 847599616

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:acb87da978e2dad13433319692c27084c166b313f922269425e1c29be9ff602b
+oid sha256:64a56a8a2bc15f4d758b0eb9991e9912dcd8f3557ca8a359e579b699adf35587
 size 1386414411

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7578bdc324f9e75df38586eb586132281169978ccdfb68af7c97dfe72a3c31b4
+oid sha256:af7bb138aa5106db6be0a3fa0ca8b584ca968cb624961c21434f43b0f94a7f53
 size 14917

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6ffeaed815dfe24a3d4d30e6928de6e80034c94a6e68899611f9c7a0a75684d5
+oid sha256:53d5b14b9a348d0c0d9daa4cabaaf2992e8a1a55a0f722d47cf938ca82885ab9
 size 14917

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ead4bfc2097774b0bab8635cdb6eefc7edb67d58c75bf9f6514bf7d08d200842
+oid sha256:3dff6d0a97e77162b13dd1d33a894b8edaf14cd2511685238f1583301c4a8320
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.4,
+  "epoch": 0.5333333333333333,
   "eval_steps": 1000,
-  "global_step": 6000,
+  "global_step": 8000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -482,6 +482,164 @@
       "eval_samples_per_second": 7.619,
       "eval_steps_per_second": 0.381,
       "step": 6000
+    },
+    {
+      "epoch": 0.4066666666666667,
+      "grad_norm": 1.1989500522613525,
+      "learning_rate": 0.0002763227702858047,
+      "loss": 8.64713623046875,
+      "step": 6100
+    },
+    {
+      "epoch": 0.41333333333333333,
+      "grad_norm": 1.3336397409439087,
+      "learning_rate": 0.00027222898458590343,
+      "loss": 8.749034423828125,
+      "step": 6200
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 1.0257115364074707,
+      "learning_rate": 0.0002681000942935204,
+      "loss": 8.847327880859375,
+      "step": 6300
+    },
+    {
+      "epoch": 0.4266666666666667,
+      "grad_norm": 1.03694748878479,
+      "learning_rate": 0.0002639381061239921,
+      "loss": 8.704371337890626,
+      "step": 6400
+    },
+    {
+      "epoch": 0.43333333333333335,
+      "grad_norm": 0.9518272876739502,
+      "learning_rate": 0.00025974504287882194,
+      "loss": 8.664287109375,
+      "step": 6500
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 1.0655875205993652,
+      "learning_rate": 0.000255522942462562,
+      "loss": 8.662864990234375,
+      "step": 6600
+    },
+    {
+      "epoch": 0.44666666666666666,
+      "grad_norm": 1.0382905006408691,
+      "learning_rate": 0.00025127385689235426,
+      "loss": 8.6200341796875,
+      "step": 6700
+    },
+    {
+      "epoch": 0.4533333333333333,
+      "grad_norm": 1.3670793771743774,
+      "learning_rate": 0.00024699985130061374,
+      "loss": 8.617677612304687,
+      "step": 6800
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 1.2171597480773926,
+      "learning_rate": 0.0002427030029313362,
+      "loss": 8.605474853515625,
+      "step": 6900
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "grad_norm": 1.1724241971969604,
+      "learning_rate": 0.00023838540013052062,
+      "loss": 8.558206787109375,
+      "step": 7000
+    },
+    {
+      "epoch": 0.4666666666666667,
+      "eval_accuracy": 0.2943894324853229,
+      "eval_loss": 8.570584297180176,
+      "eval_runtime": 65.4101,
+      "eval_samples_per_second": 7.644,
+      "eval_steps_per_second": 0.382,
+      "step": 7000
+    },
+    {
+      "epoch": 0.47333333333333333,
+      "grad_norm": 1.0969085693359375,
+      "learning_rate": 0.00023404914133119486,
+      "loss": 8.515238037109375,
+      "step": 7100
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 1.4420628547668457,
+      "learning_rate": 0.00022969633403353913,
+      "loss": 8.62325927734375,
+      "step": 7200
+    },
+    {
+      "epoch": 0.4866666666666667,
+      "grad_norm": 1.0490083694458008,
+      "learning_rate": 0.0002253290937806034,
+      "loss": 8.469694213867188,
+      "step": 7300
+    },
+    {
+      "epoch": 0.49333333333333335,
+      "grad_norm": 1.3232401609420776,
+      "learning_rate": 0.00022094954313011468,
+      "loss": 8.4819580078125,
+      "step": 7400
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 1.0941740274429321,
+      "learning_rate": 0.0002165598106228758,
+      "loss": 8.58266357421875,
+      "step": 7500
+    },
+    {
+      "epoch": 0.5066666666666667,
+      "grad_norm": 1.1495511531829834,
+      "learning_rate": 0.00021216202974825614,
+      "loss": 8.533701171875,
+      "step": 7600
+    },
+    {
+      "epoch": 0.5133333333333333,
+      "grad_norm": 1.004111886024475,
+      "learning_rate": 0.00020775833790727695,
+      "loss": 8.46027099609375,
+      "step": 7700
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 1.4242042303085327,
+      "learning_rate": 0.0002033508753737963,
+      "loss": 8.484051513671876,
+      "step": 7800
+    },
+    {
+      "epoch": 0.5266666666666666,
+      "grad_norm": 1.1219866275787354,
+      "learning_rate": 0.00019894178425429674,
+      "loss": 8.434028930664063,
+      "step": 7900
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 1.009364128112793,
+      "learning_rate": 0.00019453320744678324,
+      "loss": 8.4456640625,
+      "step": 8000
+    },
+    {
+      "epoch": 0.5333333333333333,
+      "eval_accuracy": 0.29964774951076323,
+      "eval_loss": 8.433549880981445,
+      "eval_runtime": 65.652,
+      "eval_samples_per_second": 7.616,
+      "eval_steps_per_second": 0.381,
+      "step": 8000
     }
   ],
   "logging_steps": 100,