Training in progress, step 3000, checkpoint

checkpoint-3000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b8bd3b91ae3ba947cddaba7de105929b6f1db7173482bf6b62ded43242a3499
+oid sha256:5f6f0b06050ca66df522498b9c41c5ff75413b569b8f2adff757ac7bab1c6ae7
 size 435544704

checkpoint-3000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d4a752b252488217e59292e116dc56e1740cf1173059f96f1097021dd34a1e41
+oid sha256:b6e1678fea31c33d3302a26a5f09db7dc51b36d947dc4c4b63fcbc6685259742
 size 871183627

checkpoint-3000/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b948af50f62279d14e14a4bdf3caf2f2e59738c0f88063451274a1bc74dd6429
+oid sha256:4e4811ba1845931e6e23c54c8ffbeaf5f2f52f4b91869dc9236128514d0a0fa7
 size 14645

checkpoint-3000/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 2793,
-  "best_metric": 4.5619025230407715,
+  "best_metric": 4.560876846313477,
   "best_model_checkpoint": null,
   "epoch": 7.518796992481203,
   "eval_steps": 500,
@@ -11,114 +11,114 @@
   "log_history": [
     {
       "epoch": 0.002506265664160401,
-      "grad_norm": 15.276525497436523,
+      "grad_norm": 15.287896156311035,
       "learning_rate": 0.0,
-      "loss": 10.3822,
+      "loss": 10.3873,
       "step": 1
     },
     {
       "epoch": 1.0,
-      "grad_norm": 1.2039393186569214,
+      "grad_norm": 1.349755048751831,
       "learning_rate": 7.960000000000001e-05,
-      "loss": 6.9837,
+      "loss": 6.9819,
       "step": 399
     },
     {
       "epoch": 1.0,
-      "eval_loss": 5.983016014099121,
-      "eval_runtime": 15.8842,
-      "eval_samples_per_second": 1611.54,
-      "eval_steps_per_second": 6.296,
+      "eval_loss": 5.9839253425598145,
+      "eval_runtime": 40.2685,
+      "eval_samples_per_second": 635.682,
+      "eval_steps_per_second": 2.483,
       "step": 399
     },
     {
       "epoch": 2.0,
-      "grad_norm": 1.1639982461929321,
+      "grad_norm": 1.2756340503692627,
       "learning_rate": 9.601876675603218e-05,
-      "loss": 5.6173,
+      "loss": 5.6158,
       "step": 798
     },
     {
       "epoch": 2.0,
-      "eval_loss": 5.357014179229736,
-      "eval_runtime": 20.233,
-      "eval_samples_per_second": 1265.164,
-      "eval_steps_per_second": 4.942,
+      "eval_loss": 5.355185031890869,
+      "eval_runtime": 40.338,
+      "eval_samples_per_second": 634.588,
+      "eval_steps_per_second": 2.479,
       "step": 798
     },
     {
       "epoch": 3.0,
-      "grad_norm": 1.0922253131866455,
+      "grad_norm": 1.1477612257003784,
       "learning_rate": 9.067024128686328e-05,
-      "loss": 5.1654,
+      "loss": 5.164,
       "step": 1197
     },
     {
       "epoch": 3.0,
-      "eval_loss": 5.061666488647461,
-      "eval_runtime": 21.5628,
-      "eval_samples_per_second": 1187.138,
-      "eval_steps_per_second": 4.638,
+      "eval_loss": 5.061825275421143,
+      "eval_runtime": 40.9407,
+      "eval_samples_per_second": 625.245,
+      "eval_steps_per_second": 2.443,
       "step": 1197
     },
     {
       "epoch": 4.0,
-      "grad_norm": 1.1415351629257202,
+      "grad_norm": 1.1439207792282104,
       "learning_rate": 8.532171581769438e-05,
-      "loss": 4.9033,
+      "loss": 4.902,
       "step": 1596
     },
     {
       "epoch": 4.0,
-      "eval_loss": 4.871504306793213,
-      "eval_runtime": 21.5888,
-      "eval_samples_per_second": 1185.71,
-      "eval_steps_per_second": 4.632,
+      "eval_loss": 4.870055675506592,
+      "eval_runtime": 40.5657,
+      "eval_samples_per_second": 631.026,
+      "eval_steps_per_second": 2.465,
       "step": 1596
     },
     {
       "epoch": 5.0,
-      "grad_norm": 1.2201212644577026,
+      "grad_norm": 1.2400050163269043,
       "learning_rate": 7.997319034852548e-05,
-      "loss": 4.7113,
+      "loss": 4.7103,
       "step": 1995
     },
     {
       "epoch": 5.0,
-      "eval_loss": 4.739400386810303,
-      "eval_runtime": 21.6571,
-      "eval_samples_per_second": 1181.971,
-      "eval_steps_per_second": 4.617,
+      "eval_loss": 4.737980842590332,
+      "eval_runtime": 40.84,
+      "eval_samples_per_second": 626.787,
+      "eval_steps_per_second": 2.449,
       "step": 1995
     },
     {
       "epoch": 6.0,
-      "grad_norm": 1.3053981065750122,
+      "grad_norm": 1.3254481554031372,
       "learning_rate": 7.462466487935658e-05,
-      "loss": 4.5573,
+      "loss": 4.5566,
       "step": 2394
     },
     {
       "epoch": 6.0,
-      "eval_loss": 4.6380228996276855,
-      "eval_runtime": 20.1975,
-      "eval_samples_per_second": 1267.387,
-      "eval_steps_per_second": 4.951,
+      "eval_loss": 4.6372785568237305,
+      "eval_runtime": 40.6525,
+      "eval_samples_per_second": 629.678,
+      "eval_steps_per_second": 2.46,
       "step": 2394
     },
     {
       "epoch": 7.0,
-      "grad_norm": 1.306992769241333,
+      "grad_norm": 1.3006181716918945,
       "learning_rate": 6.927613941018766e-05,
-      "loss": 4.4278,
+      "loss": 4.4277,
       "step": 2793
     },
     {
       "epoch": 7.0,
-      "eval_loss": 4.5619025230407715,
-      "eval_runtime": 20.2525,
-      "eval_samples_per_second": 1263.94,
-      "eval_steps_per_second": 4.938,
+      "eval_loss": 4.560876846313477,
+      "eval_runtime": 40.4604,
+      "eval_samples_per_second": 632.667,
+      "eval_steps_per_second": 2.472,
       "step": 2793
     }
   ],

checkpoint-3000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3d99a586d2a38e89becddbee0adce6692398fb4ff8b6a71d5f8e545b03eeef6
+oid sha256:c4d8e95126c79934c43d708d7208a8515199b20fd582881a681e358089ed0c56
 size 5905