Training in progress, step 30500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b53972236677d35c7627b344577268d19eab8e86becdb001ad3d20c33562409f
+oid sha256:fdc0ad979a12ba1d5693d6f619dbaea0818f22e8b6c5ce88c754da865d18dae9
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9ddb71187cf05b2d152abcaf17e6a1a66bbaf60093ba03650f5df78e882a41ae
+oid sha256:87d705db2757234d5d8a502c3a722562455d35ceffb96acc7f8a84d9a82b57f2
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0caa256a01d13ade6a76529796c9eea8cb685776ddc6a24c1e4a28d00ce8ca63
+oid sha256:a17843cbe8069abb8746e8e871bb3e80d6841c1b2726ca6959bb29928e72639f
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd6900b77a3aaa53dbd9ff02530a8bdbceb59bb2b9ea31ba74caa08f3dcac033
-size 14308
+oid sha256:5baf28c8bc0abeb08100e2c5cfe54b52c8077e77b1cb64b5b332c05fac995071
+size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8c04c082df1347d4a5bd7568aa1ea9e76f0f21cd9fd31e93e6a215b3ffd653d
+oid sha256:8eef9d19aff91ed1822b958b8f57eac8d029c3900de24c8d1baed5c3a38e5ec6
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7472cb4cae0bcbbe91407a8ab253d568120f4cb09f676c536e7c370fc9153e84
+oid sha256:6b26116b7f3dd831125aeebdfb3d11e2bdb55abc7c8e404210b0ff9bfd497ccc
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 30000,
-  "best_metric": 0.7264938354492188,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-30000",
-  "epoch": 2.307514806553342,
+  "best_global_step": 30500,
+  "best_metric": 0.7244414687156677,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-30500",
+  "epoch": 2.3459733866625645,
   "eval_steps": 250,
-  "global_step": 30000,
+  "global_step": 30500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -5168,6 +5168,92 @@
       "eval_samples_per_second": 56.228,
       "eval_steps_per_second": 14.057,
       "step": 30000
+    },
+    {
+      "epoch": 2.311360664564264,
+      "grad_norm": 0.861571729183197,
+      "learning_rate": 4.6567799911695194e-05,
+      "loss": 0.7052,
+      "step": 30050
+    },
+    {
+      "epoch": 2.3152065225751866,
+      "grad_norm": 0.9279738068580627,
+      "learning_rate": 4.6308079889878714e-05,
+      "loss": 0.715,
+      "step": 30100
+    },
+    {
+      "epoch": 2.3190523805861085,
+      "grad_norm": 0.8576169013977051,
+      "learning_rate": 4.6048359868062226e-05,
+      "loss": 0.7125,
+      "step": 30150
+    },
+    {
+      "epoch": 2.322898238597031,
+      "grad_norm": 1.3127994537353516,
+      "learning_rate": 4.578863984624575e-05,
+      "loss": 0.6703,
+      "step": 30200
+    },
+    {
+      "epoch": 2.3267440966079533,
+      "grad_norm": 0.5930036306381226,
+      "learning_rate": 4.5528919824429266e-05,
+      "loss": 0.6927,
+      "step": 30250
+    },
+    {
+      "epoch": 2.3267440966079533,
+      "eval_loss": 0.7252874970436096,
+      "eval_runtime": 17.8506,
+      "eval_samples_per_second": 56.021,
+      "eval_steps_per_second": 14.005,
+      "step": 30250
+    },
+    {
+      "epoch": 2.3305899546188753,
+      "grad_norm": 0.6445633769035339,
+      "learning_rate": 4.5269199802612785e-05,
+      "loss": 0.6804,
+      "step": 30300
+    },
+    {
+      "epoch": 2.3344358126297977,
+      "grad_norm": 0.9251648783683777,
+      "learning_rate": 4.5009479780796305e-05,
+      "loss": 0.7226,
+      "step": 30350
+    },
+    {
+      "epoch": 2.33828167064072,
+      "grad_norm": 1.2103322744369507,
+      "learning_rate": 4.474975975897982e-05,
+      "loss": 0.7179,
+      "step": 30400
+    },
+    {
+      "epoch": 2.342127528651642,
+      "grad_norm": 1.0718954801559448,
+      "learning_rate": 4.4490039737163344e-05,
+      "loss": 0.7638,
+      "step": 30450
+    },
+    {
+      "epoch": 2.3459733866625645,
+      "grad_norm": 0.8156006336212158,
+      "learning_rate": 4.423031971534686e-05,
+      "loss": 0.6733,
+      "step": 30500
+    },
+    {
+      "epoch": 2.3459733866625645,
+      "eval_loss": 0.7244414687156677,
+      "eval_runtime": 17.9373,
+      "eval_samples_per_second": 55.75,
+      "eval_steps_per_second": 13.937,
+      "step": 30500
     }
   ],
   "logging_steps": 50,