Training in progress, step 6000, checkpoint

checkpoint-6000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6201eca2fdba1c31fa3f35028224e3c090f9b15101bb73605ce3f209ec6b5402
+oid sha256:302c59683a0837b636a18c7d18c19c60469ae89154c37444246269a31473c4eb
 size 14809584

checkpoint-6000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:237383b37a8d99426fae3f72a15064ec5f948f034faf7a72ab520673fa012426
+oid sha256:f40b7ad6e040fc91f1fc002a3e014c85887fa2e0422a7a296026b85c58a44d02
 size 29680378

checkpoint-6000/trainer_state.json

@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.003616819658138206,
-      "grad_norm": 0.14744244515895844,
+      "grad_norm": 0.14744240045547485,
       "learning_rate": 0.0001,
       "loss": 0.7697,
       "step": 2000
@@ -18,14 +18,14 @@
     {
       "epoch": 0.003616819658138206,
       "eval_loss": 0.19154727458953857,
-      "eval_runtime": 110.4273,
-      "eval_samples_per_second": 330.851,
-      "eval_steps_per_second": 20.683,
+      "eval_runtime": 91.812,
+      "eval_samples_per_second": 397.933,
+      "eval_steps_per_second": 24.877,
       "step": 2000
     },
     {
       "epoch": 0.007233639316276412,
-      "grad_norm": 0.1623348891735077,
+      "grad_norm": 0.1623348593711853,
       "learning_rate": 0.0001,
       "loss": 0.1884,
       "step": 4000
@@ -33,14 +33,14 @@
     {
       "epoch": 0.007233639316276412,
       "eval_loss": 0.17771275341510773,
-      "eval_runtime": 90.2021,
-      "eval_samples_per_second": 405.035,
-      "eval_steps_per_second": 25.321,
+      "eval_runtime": 99.0721,
+      "eval_samples_per_second": 368.772,
+      "eval_steps_per_second": 23.054,
       "step": 4000
     },
     {
       "epoch": 0.010850458974414617,
-      "grad_norm": 0.13657262921333313,
+      "grad_norm": 0.13657265901565552,
       "learning_rate": 0.0001,
       "loss": 0.178,
       "step": 6000
@@ -48,9 +48,9 @@
     {
       "epoch": 0.010850458974414617,
       "eval_loss": 0.16984443366527557,
-      "eval_runtime": 99.6503,
-      "eval_samples_per_second": 366.632,
-      "eval_steps_per_second": 22.92,
+      "eval_runtime": 90.3285,
+      "eval_samples_per_second": 404.468,
+      "eval_steps_per_second": 25.285,
       "step": 6000
     }
   ],

checkpoint-6000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66a3b67ff96aad171c21e5c77947a3759ef9af49c075132a399cacc48fed219d
+oid sha256:f5e294196a968439ae6368c7b1ef6eb6e2124702ff760349c5ba2ea15623271d
 size 5496