Training in progress, step 32500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5592650a84fed0e53ff657190c7416551adf3f79e86876a254dd6c273b0bccdc
+oid sha256:27524d688ba5a4ae0495e37fb470fa799513485230c4286aa92c730ea8ff1720
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32d27943f70edaf81f6f4c1789ed21c721bcc0f5827fef85bdd08cb9092a87bd
+oid sha256:563f5a064c4b39c0836e86f00dee2d8c60f00f322e1f0bd5e436a0d8b5d55205
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c3d55653bad9ee7c1db558a9b36a74c2617a2024075b12282a849a401a70d6bf
+oid sha256:9efa462e1ab2e4553d732eb9e42592eb13a5529b70ac4d5fb44f739d27fa34a0
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1f687f79c35c0845312b8176cbb48a154ba5067da083afab2edee21bf2e08ac
+oid sha256:e29738d10f11c84e844c29337308ab16f3f9558edfaa8334115526af83073d59
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66e856574eb8a1f2a02d63513aff48279087ce622f3e8211116a357d66952614
+oid sha256:bf573fdaab0c323d3840f8b395f4756e638c7ff2bf57a412d9956a284d789a29
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a9ed56cf408beb5289ea76a3c22da837df362efd2635153f560e4bbed768681d
+oid sha256:87064723657331f8e12b7e151a915fe1a65dd2d8be728ee5859becab87fef8c0
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 31000,
   "best_metric": 0.7226839661598206,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-31000",
-  "epoch": 2.4613491269902315,
+  "epoch": 2.4998077070994538,
   "eval_steps": 250,
-  "global_step": 32000,
+  "global_step": 32500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -5512,6 +5512,92 @@
       "eval_samples_per_second": 55.549,
       "eval_steps_per_second": 13.887,
       "step": 32000
+    },
+    {
+      "epoch": 2.465194985001154,
+      "grad_norm": 0.8729520440101624,
+      "learning_rate": 3.6178999039035924e-05,
+      "loss": 0.7121,
+      "step": 32050
+    },
+    {
+      "epoch": 2.469040843012076,
+      "grad_norm": 1.2588157653808594,
+      "learning_rate": 3.591927901721944e-05,
+      "loss": 0.7145,
+      "step": 32100
+    },
+    {
+      "epoch": 2.4728867010229982,
+      "grad_norm": 1.0234293937683105,
+      "learning_rate": 3.565955899540296e-05,
+      "loss": 0.7173,
+      "step": 32150
+    },
+    {
+      "epoch": 2.4767325590339206,
+      "grad_norm": 0.6210401058197021,
+      "learning_rate": 3.5399838973586476e-05,
+      "loss": 0.7404,
+      "step": 32200
+    },
+    {
+      "epoch": 2.4805784170448426,
+      "grad_norm": 1.0649775266647339,
+      "learning_rate": 3.5140118951769996e-05,
+      "loss": 0.713,
+      "step": 32250
+    },
+    {
+      "epoch": 2.4805784170448426,
+      "eval_loss": 0.7857936024665833,
+      "eval_runtime": 17.9415,
+      "eval_samples_per_second": 55.737,
+      "eval_steps_per_second": 13.934,
+      "step": 32250
+    },
+    {
+      "epoch": 2.484424275055765,
+      "grad_norm": 0.6743142604827881,
+      "learning_rate": 3.488039892995351e-05,
+      "loss": 0.7213,
+      "step": 32300
+    },
+    {
+      "epoch": 2.488270133066687,
+      "grad_norm": 0.7584249377250671,
+      "learning_rate": 3.462067890813703e-05,
+      "loss": 0.7175,
+      "step": 32350
+    },
+    {
+      "epoch": 2.4921159910776094,
+      "grad_norm": 1.7324374914169312,
+      "learning_rate": 3.436095888632055e-05,
+      "loss": 0.7255,
+      "step": 32400
+    },
+    {
+      "epoch": 2.495961849088532,
+      "grad_norm": 1.0071933269500732,
+      "learning_rate": 3.410123886450407e-05,
+      "loss": 0.6905,
+      "step": 32450
+    },
+    {
+      "epoch": 2.4998077070994538,
+      "grad_norm": 0.8606531023979187,
+      "learning_rate": 3.384151884268759e-05,
+      "loss": 0.7235,
+      "step": 32500
+    },
+    {
+      "epoch": 2.4998077070994538,
+      "eval_loss": 0.7832362651824951,
+      "eval_runtime": 17.9726,
+      "eval_samples_per_second": 55.64,
+      "eval_steps_per_second": 13.91,
+      "step": 32500
     }
   ],
   "logging_steps": 50,