Training in progress, step 24500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a49f5cb3d3ccac7817610fa964bf361289eb0cfe2f4eb026b3d7dec09b14c94e
+oid sha256:864e393a661e445ab791d8ddc8820ce0124af5e25c563025e5274fb46871c164
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4c4dc22fe695886da65efe80a9483d46131feb5c3951ee4e1441cd0c20f8ca7b
+oid sha256:69b19c7b3ab8d24c43cb3d97f2b5942e5102d4a46144bd92361e16d8ebb3d307
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7232363cf74531ce8b422db0be4a1c52fb65a8350caa9a3033db48afd581b28e
+oid sha256:ae59df9b85342e49ccc0406baf5315bd2fd8f6e695f8cf5a8b8edc0181516fc6
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:599e4a0a057d34a3454abdec04ea1df53e6a5a47a3aac39f21ff339cea978175
-size 14244
+oid sha256:741e3f6978068fd80c0a2938c07ae3df768f971aa6e2756e07360dec11f377ea
+size 14180

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25acdcc8c2ddbe44392c8342cb938e96014287130fce2f529b63734cdda74387
+oid sha256:8d31530c22f7fe8bcf00bdc50aaced04b559ddaa1caf679a3f57a45170ec249b
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac4e57d345b8d2e1b9cf78f859eede2962dbe55a1ef61e82edae2f8653a05425
+oid sha256:96928cc818781495e8d68be480c829211a700dd8771b47547694b6ed3ba139b1
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 24000,
-  "best_metric": 0.7566245198249817,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-24000",
-  "epoch": 1.8460118452426735,
+  "best_global_step": 24500,
+  "best_metric": 0.7536496520042419,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-24500",
+  "epoch": 1.884470425351896,
   "eval_steps": 250,
-  "global_step": 24000,
+  "global_step": 24500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4136,6 +4136,92 @@
       "eval_samples_per_second": 55.849,
       "eval_steps_per_second": 13.962,
       "step": 24000
+    },
+    {
+      "epoch": 1.8498577032535959,
+      "grad_norm": 1.1524229049682617,
+      "learning_rate": 7.771861932836403e-05,
+      "loss": 0.7732,
+      "step": 24050
+    },
+    {
+      "epoch": 1.853703561264518,
+      "grad_norm": 0.4653956890106201,
+      "learning_rate": 7.745889930654755e-05,
+      "loss": 0.7961,
+      "step": 24100
+    },
+    {
+      "epoch": 1.8575494192754403,
+      "grad_norm": 0.8423280119895935,
+      "learning_rate": 7.719917928473105e-05,
+      "loss": 0.7963,
+      "step": 24150
+    },
+    {
+      "epoch": 1.8613952772863627,
+      "grad_norm": 0.6979435086250305,
+      "learning_rate": 7.693945926291458e-05,
+      "loss": 0.7816,
+      "step": 24200
+    },
+    {
+      "epoch": 1.8652411352972849,
+      "grad_norm": 0.7800914645195007,
+      "learning_rate": 7.66797392410981e-05,
+      "loss": 0.7563,
+      "step": 24250
+    },
+    {
+      "epoch": 1.8652411352972849,
+      "eval_loss": 0.7574715614318848,
+      "eval_runtime": 17.9043,
+      "eval_samples_per_second": 55.852,
+      "eval_steps_per_second": 13.963,
+      "step": 24250
+    },
+    {
+      "epoch": 1.869086993308207,
+      "grad_norm": 0.9678017497062683,
+      "learning_rate": 7.642001921928162e-05,
+      "loss": 0.7689,
+      "step": 24300
+    },
+    {
+      "epoch": 1.8729328513191295,
+      "grad_norm": 0.454647421836853,
+      "learning_rate": 7.616029919746513e-05,
+      "loss": 0.7231,
+      "step": 24350
+    },
+    {
+      "epoch": 1.8767787093300514,
+      "grad_norm": 0.7899460792541504,
+      "learning_rate": 7.590057917564866e-05,
+      "loss": 0.7471,
+      "step": 24400
+    },
+    {
+      "epoch": 1.8806245673409738,
+      "grad_norm": 1.1373926401138306,
+      "learning_rate": 7.564085915383217e-05,
+      "loss": 0.8342,
+      "step": 24450
+    },
+    {
+      "epoch": 1.884470425351896,
+      "grad_norm": 1.4272133111953735,
+      "learning_rate": 7.538113913201568e-05,
+      "loss": 0.7585,
+      "step": 24500
+    },
+    {
+      "epoch": 1.884470425351896,
+      "eval_loss": 0.7536496520042419,
+      "eval_runtime": 17.9401,
+      "eval_samples_per_second": 55.741,
+      "eval_steps_per_second": 13.935,
+      "step": 24500
     }
   ],
   "logging_steps": 50,