Training in progress, step 2500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed248223fb076f4557238ee5eec9add808ee83dacf213c5b42e548a3180d84ba
+oid sha256:1615a27846e2fc7e29e90ffc163230511731ab45b760a836e71cf204236e624d
 size 36730224

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca187211ec5ca22e22494af8fae9ec9214346dc466486151b4b1df2b8f4958fe
+oid sha256:6df96e6756fb0b0f050ff10208655f563c96fccd6a701602a5e90ccbd1e235e0
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2512ba1bd332fd1303d75224b66504a865cf77c5c151bdcdfcda52c23dbe2db6
+oid sha256:74818f2f3daacf3c0fca3dde38a211872e713dacdefaa18d7435559b30dc197d
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8fdbdc59fd9e02ea3e69d676110893ccd87d22af41e77bc0d26c87371fa56871
+oid sha256:bc76fea5e634e306eb2d4aa830b8059981049eb83bd4681631ccd02732a47898
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e942a77a12aac95a1f43f3b73ecd979987695991d41d93e1b6557c90b404597e
+oid sha256:0d8c94e41b1ac1e8ac4ea58294a516d2e00679d859d84c40c165bc9d685fc309
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 2000,
-  "best_metric": 4.620576858520508,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints/checkpoint-2000",
-  "epoch": 0.15383432043688947,
+  "best_global_step": 2500,
+  "best_metric": 4.558788776397705,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints/checkpoint-2500",
+  "epoch": 0.19229290054611184,
   "eval_steps": 250,
-  "global_step": 2000,
+  "global_step": 2500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -352,6 +352,92 @@
       "eval_samples_per_second": 53.279,
       "eval_steps_per_second": 13.32,
       "step": 2000
+    },
+    {
+      "epoch": 0.1576801784478117,
+      "grad_norm": 3.5292210578918457,
+      "learning_rate": 0.00012999448549685672,
+      "loss": 4.7815,
+      "step": 2050
+    },
+    {
+      "epoch": 0.16152603645873395,
+      "grad_norm": 4.665738105773926,
+      "learning_rate": 0.0001297371420168376,
+      "loss": 4.6789,
+      "step": 2100
+    },
+    {
+      "epoch": 0.16537189446965617,
+      "grad_norm": 4.332949161529541,
+      "learning_rate": 0.00012947979853681848,
+      "loss": 4.5991,
+      "step": 2150
+    },
+    {
+      "epoch": 0.1692177524805784,
+      "grad_norm": 3.8279120922088623,
+      "learning_rate": 0.00012922245505679936,
+      "loss": 4.5791,
+      "step": 2200
+    },
+    {
+      "epoch": 0.17306361049150065,
+      "grad_norm": 1.9522042274475098,
+      "learning_rate": 0.00012896511157678027,
+      "loss": 4.5643,
+      "step": 2250
+    },
+    {
+      "epoch": 0.17306361049150065,
+      "eval_loss": 4.609655857086182,
+      "eval_runtime": 18.946,
+      "eval_samples_per_second": 52.782,
+      "eval_steps_per_second": 13.195,
+      "step": 2250
+    },
+    {
+      "epoch": 0.1769094685024229,
+      "grad_norm": 4.264033794403076,
+      "learning_rate": 0.00012870776809676115,
+      "loss": 4.6666,
+      "step": 2300
+    },
+    {
+      "epoch": 0.18075532651334514,
+      "grad_norm": 4.572433948516846,
+      "learning_rate": 0.000128450424616742,
+      "loss": 4.6096,
+      "step": 2350
+    },
+    {
+      "epoch": 0.18460118452426735,
+      "grad_norm": 3.8559391498565674,
+      "learning_rate": 0.0001281930811367229,
+      "loss": 4.6425,
+      "step": 2400
+    },
+    {
+      "epoch": 0.1884470425351896,
+      "grad_norm": 2.9414010047912598,
+      "learning_rate": 0.0001279357376567038,
+      "loss": 4.6336,
+      "step": 2450
+    },
+    {
+      "epoch": 0.19229290054611184,
+      "grad_norm": 4.745160102844238,
+      "learning_rate": 0.00012767839417668467,
+      "loss": 4.6792,
+      "step": 2500
+    },
+    {
+      "epoch": 0.19229290054611184,
+      "eval_loss": 4.558788776397705,
+      "eval_runtime": 18.9882,
+      "eval_samples_per_second": 52.664,
+      "eval_steps_per_second": 13.166,
+      "step": 2500
     }
   ],
   "logging_steps": 50,