Training in progress, step 38000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d197cd9a3404a8c8f67910780423aecd5fdc3add3f182e0d82ab6682395c0006
+oid sha256:dd7bc3b6641670f074b55a5b4f90528448bcea90e89d866eadc79e22cadf1bd2
 size 529337896

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:419fd2c3e0ba5d1952d3f93465eeb172ba6dac85f67c82c9eeb58a6bc214cded
+oid sha256:805511bda687dc757ba933396fffe6806155d85d1eb99922300896d858f2b615
 size 871247243

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a202a832e7dde4f4f295f57293a1d0bb6e0f939cebe50d0b66fc7a8608bef7a
+oid sha256:4fffdc0b55d45a254da995950f6ade5109e1182c4a9a3b190cf11bc8c48d0b82
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d525574f54acfd6c1c0d9e8a984a9fe4782c94abdc8722296a2554b91af7e0e9
+oid sha256:40c934a53b7d741f620d2fd5ca3781d46f7df109d0ca475956382429898d18f8
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.72,
+  "epoch": 0.76,
   "eval_steps": 1000,
-  "global_step": 36000,
+  "global_step": 38000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -2852,6 +2852,164 @@
       "eval_samples_per_second": 85.261,
       "eval_steps_per_second": 0.682,
       "step": 36000
+    },
+    {
+      "epoch": 0.722,
+      "grad_norm": 0.14137360453605652,
+      "learning_rate": 0.0003936988995130556,
+      "loss": 3.254,
+      "step": 36100
+    },
+    {
+      "epoch": 0.724,
+      "grad_norm": 0.1477884203195572,
+      "learning_rate": 0.0003884526092395261,
+      "loss": 3.25,
+      "step": 36200
+    },
+    {
+      "epoch": 0.726,
+      "grad_norm": 0.14579065144062042,
+      "learning_rate": 0.00038323307003000693,
+      "loss": 3.2324,
+      "step": 36300
+    },
+    {
+      "epoch": 0.728,
+      "grad_norm": 0.18951904773712158,
+      "learning_rate": 0.0003780405102040524,
+      "loss": 3.2393,
+      "step": 36400
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 0.1553814709186554,
+      "learning_rate": 0.0003728751569010509,
+      "loss": 3.2329,
+      "step": 36500
+    },
+    {
+      "epoch": 0.732,
+      "grad_norm": 0.15148834884166718,
+      "learning_rate": 0.00036773723607028965,
+      "loss": 3.2493,
+      "step": 36600
+    },
+    {
+      "epoch": 0.734,
+      "grad_norm": 0.14959374070167542,
+      "learning_rate": 0.00036262697246107,
+      "loss": 3.2412,
+      "step": 36700
+    },
+    {
+      "epoch": 0.736,
+      "grad_norm": 0.15129899978637695,
+      "learning_rate": 0.0003575445896128768,
+      "loss": 3.2299,
+      "step": 36800
+    },
+    {
+      "epoch": 0.738,
+      "grad_norm": 0.14516687393188477,
+      "learning_rate": 0.0003524903098456013,
+      "loss": 3.2402,
+      "step": 36900
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.14438898861408234,
+      "learning_rate": 0.00034746435424981315,
+      "loss": 3.2538,
+      "step": 37000
+    },
+    {
+      "epoch": 0.74,
+      "eval_accuracy": 0.3904481409001957,
+      "eval_loss": 3.250405788421631,
+      "eval_runtime": 11.8201,
+      "eval_samples_per_second": 84.602,
+      "eval_steps_per_second": 0.677,
+      "step": 37000
+    },
+    {
+      "epoch": 0.742,
+      "grad_norm": 0.21630696952342987,
+      "learning_rate": 0.00034246694267709257,
+      "loss": 3.2171,
+      "step": 37100
+    },
+    {
+      "epoch": 0.744,
+      "grad_norm": 0.14735347032546997,
+      "learning_rate": 0.0003374982937304113,
+      "loss": 3.2346,
+      "step": 37200
+    },
+    {
+      "epoch": 0.746,
+      "grad_norm": 0.15532496571540833,
+      "learning_rate": 0.0003325586247545698,
+      "loss": 3.2403,
+      "step": 37300
+    },
+    {
+      "epoch": 0.748,
+      "grad_norm": 0.17086553573608398,
+      "learning_rate": 0.00032764815182669205,
+      "loss": 3.2336,
+      "step": 37400
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.14083310961723328,
+      "learning_rate": 0.00032276708974677117,
+      "loss": 3.2287,
+      "step": 37500
+    },
+    {
+      "epoch": 0.752,
+      "grad_norm": 0.13698738813400269,
+      "learning_rate": 0.00031791565202827533,
+      "loss": 3.2262,
+      "step": 37600
+    },
+    {
+      "epoch": 0.754,
+      "grad_norm": 0.1423436999320984,
+      "learning_rate": 0.0003130940508888063,
+      "loss": 3.2448,
+      "step": 37700
+    },
+    {
+      "epoch": 0.756,
+      "grad_norm": 0.2180064171552658,
+      "learning_rate": 0.00030830249724081817,
+      "loss": 3.2235,
+      "step": 37800
+    },
+    {
+      "epoch": 0.758,
+      "grad_norm": 0.16625255346298218,
+      "learning_rate": 0.0003035412006823901,
+      "loss": 3.2224,
+      "step": 37900
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.14719747006893158,
+      "learning_rate": 0.0002988103694880576,
+      "loss": 3.2353,
+      "step": 38000
+    },
+    {
+      "epoch": 0.76,
+      "eval_accuracy": 0.39081996086105675,
+      "eval_loss": 3.2475223541259766,
+      "eval_runtime": 11.4579,
+      "eval_samples_per_second": 87.276,
+      "eval_steps_per_second": 0.698,
+      "step": 38000
     }
   ],
   "logging_steps": 100,