Training in progress, step 16000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9f1ed52cf80a5ab8d5ed12c9555137201579cc86f9a2f068ecfc94b91627d37
+oid sha256:43a22733b29e992c0bd766025c6983872cb4acdb51a99cfbd6ae33f04e90e488
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5311da3aa8361f2e32c9e1dadd3f99fb776577e3edbc936e8d5e4a247f82b48
+oid sha256:75a8f8b3d0832f2fc4424ae0d890bc11c09d5e5d890a55794f03850f8038b621
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ac8a732d897ca3b87406efba9af34062b6403d9fcd06247903138f1a20ce830
+oid sha256:64a4f0cd681e844471044bb218c832bcc1142d56051b8bf40fe1abbce916dd45
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e443e4b8c749db54f7826bd4573803e0f05009235443f881fa9932376e1aba8
-size 14244
+oid sha256:2268c750118e17cd7866635da6e44831e071d618455f938ab68da6b4daa1e485
+size 14180

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce992341f57d419e4ec2c7637893018b15690261bc216bb5e0545131a4a5ab1d
+oid sha256:53cd8dea1c9f30f479c9d6030e3863aab75d4d5334d9c298aeeb7854ecce09dd
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b94bf7f68d57ae943a99758bd0c736315f81c408fbb84a10b3ea5a5a1956ed5
+oid sha256:92a6c2c1a93051773698e8cc5d95d9193f3be9f07990ac4d13439261ba391175
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 15500,
-  "best_metric": 1.5013692378997803,
+  "best_global_step": 15750,
+  "best_metric": 1.4955236911773682,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-15500",
-  "epoch": 1.1922159833858934,
+  "epoch": 1.2306745634951157,
   "eval_steps": 250,
-  "global_step": 15500,
+  "global_step": 16000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -2674,6 +2674,92 @@
       "eval_samples_per_second": 55.539,
       "eval_steps_per_second": 13.885,
       "step": 15500
+    },
+    {
+      "epoch": 1.1960618413968156,
+      "grad_norm": 3.367638349533081,
+      "learning_rate": 0.00012187621743760226,
+      "loss": 1.4261,
+      "step": 15550
+    },
+    {
+      "epoch": 1.1999076994077378,
+      "grad_norm": 1.6329169273376465,
+      "learning_rate": 0.00012161649741578578,
+      "loss": 1.4475,
+      "step": 15600
+    },
+    {
+      "epoch": 1.2037535574186602,
+      "grad_norm": 1.397910475730896,
+      "learning_rate": 0.00012135677739396932,
+      "loss": 1.4747,
+      "step": 15650
+    },
+    {
+      "epoch": 1.2075994154295824,
+      "grad_norm": 1.7463736534118652,
+      "learning_rate": 0.00012109705737215283,
+      "loss": 1.4745,
+      "step": 15700
+    },
+    {
+      "epoch": 1.2114452734405046,
+      "grad_norm": 1.8097542524337769,
+      "learning_rate": 0.00012083733735033633,
+      "loss": 1.5276,
+      "step": 15750
+    },
+    {
+      "epoch": 1.2114452734405046,
+      "eval_loss": 1.4955236911773682,
+      "eval_runtime": 17.9054,
+      "eval_samples_per_second": 55.849,
+      "eval_steps_per_second": 13.962,
+      "step": 15750
+    },
+    {
+      "epoch": 1.2152911314514268,
+      "grad_norm": 1.745730996131897,
+      "learning_rate": 0.00012057761732851987,
+      "loss": 1.5282,
+      "step": 15800
+    },
+    {
+      "epoch": 1.2191369894623492,
+      "grad_norm": 1.3859128952026367,
+      "learning_rate": 0.00012031789730670338,
+      "loss": 1.4795,
+      "step": 15850
+    },
+    {
+      "epoch": 1.2229828474732714,
+      "grad_norm": 1.5086127519607544,
+      "learning_rate": 0.0001200581772848869,
+      "loss": 1.437,
+      "step": 15900
+    },
+    {
+      "epoch": 1.2268287054841935,
+      "grad_norm": 2.088292121887207,
+      "learning_rate": 0.00011979845726307042,
+      "loss": 1.5011,
+      "step": 15950
+    },
+    {
+      "epoch": 1.2306745634951157,
+      "grad_norm": 1.2746011018753052,
+      "learning_rate": 0.00011953873724125393,
+      "loss": 1.4596,
+      "step": 16000
+    },
+    {
+      "epoch": 1.2306745634951157,
+      "eval_loss": 1.4984314441680908,
+      "eval_runtime": 17.8066,
+      "eval_samples_per_second": 56.159,
+      "eval_steps_per_second": 14.04,
+      "step": 16000
     }
   ],
   "logging_steps": 50,